home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Amiga Tools 3
/
Amiga Tools 3.iso
/
grafik
/
raytracing
/
imagine
/
tips
/
arc44.txt
< prev
next >
Wrap
Text File
|
1993-12-05
|
374KB
|
10,044 lines
This is the Imagine Mailing List (imagine@email.sp.paramax.com) Archive #44
covering messages from Nov. 01. to Nov. 30. 1993.
If you have any questions or problems with this file, E-mail Nik Vukovljak
at nvukovlj@extro.ucc.su.oz.au
Note: each message separated by '##'.
------------------------------------------------------------------------------
##
Subject: RE: Stars & Artifacts & Stuff
Date: Mon, 1 Nov 93 10:30:05 CST
From: drrogers@camelot.b24a.ingr.com (Dale R Rogers)
|Imagemaster has a wonderful filter operator called "NTSC Filter" (how
|appropriate!) which smooths out any illegal color/luminance transitions,
|like the aforementioned stars. A CompuServe user uploaded one of the
|classic Babylon V images, showing how NTSC Filter dramatically improved
|image quality.
ADPro has the same operator.
____________________________^____________________________
dale r. rogers
Intergraph Corporation
Building Design & Management MailStop: LR24A4
drrogers@b24a.b24a.ingr.com Tel: (205) 730-8294
.
##
Subject: Re: whining, and a _real_ question..
Date: Mon, 1 Nov 93 17:21:00 GMT
From: glewis@pcocd2.intel.com (Glenn M. Lewis - ICD ~)
>>>>> "Dan" == Daniel Jr Murrell <djm2@Ra.MsState.Edu> writes:
Dan> ...So I dusted off my Terrain disk (remember
Dan> that ol' program? :) I made a really nice landscape, exactly what
Dan> I needed, but then I couldn't get Imagine to load the object. I
Dan> eventually had to boot Turbo Silver up, at the Impulse guys'
Dan> suggestion, and load it in there and save it back out...
Dan> ...Has anyone had success converting these old objects?
Dan> Glenn, does T3D load them?
Hi, Dan!
Yes, if Turbo Silver can load them, then T3DLIB should be able
to load them. That concerns me, though, that loading into TS and
saving it back out solves the problem... I would have to check it out
myself. The main thing that T3DLIB will do is convert all EXTR
(external) objects into internal objects for Imagine to read. Give it
a try and let me know how it goes... I would be interested to find
out.
-- Glenn
##
Subject: Undersea F/X
Date: Mon, 01 Nov 93 11:01:00 PST
From: Stethem Ted 5721 <TedS@ms70.nuwes.sea06.navy.mil>
Well, I suppose you might have seen a few episodes of seaQuest DSV by
now. In case you didn't know, the undersea scenes are all done with
Lightwave and Video Toaster. Anyway, they have been able to achieve a
pretty decent effect of being underwater that I am having a hard time trying
to achieve with Imagine.
I have been able to get some marginal appearance of being underwater with
the Global Fog. I have also been experimenting with planes with ripple
texture placed in front of the camera and planes with waves texture also,
to try to get a "ripple"y effect. So far, the end result has been less than
satisfactory.
Also, I have a Ground object with a wave texture applied to it and it
looks fine from above the surface. When the camera is underneath the
"waves", though, there is a distinct line in the horizon. I have tried to
blur this horizon line with the Global fog to no avail. Basically, I want
everything beyond the immediate foreground to blur with distance. And I
want a ripple look through the camera.
Anybody have any tips or suggestions?
##
Subject: Enough already
Date: Mon, 1 Nov 93 11:10:19 -0800
From: stevez@rhythm.com (Steve Ziolkowski)
You know, I paid 5000 for a complete system 2 years ago. It seems I
can get the same system for less than 3000 now. That's life.
So is the fact that I have been using imagine since it's creation, and
started out in Turbo. The trick in CG is getting the results you
want with what you have, and work around the bugs. Hell,
even the high end software doesn't work right a lot of the
time.
If Imagine wants to do this free giveaway thing, it's their choice. I don't
care much one way or the other since I already paid for mine
and am very happy with it. I get the photorealistic results
I want with Imagine, and just because they decide to give it
away has no impact on the quality of the program.
So my complaint is twofold. One, we all know that Imagine does
have some bugs. Yeah. So enough bitching about them, and
find ways to work around them. I have sent out a couple of
questions that nobody has answered, and I'm still waiting for a
reply. Two, deal with the fact that we really have no control
over what a software company chooses to do with their product,
and if that means some of you will use another product, so be it.
Please leave the IML to those of us who want to use Imagine.
I realize that some of you will flame me for saying this, but hey.
I signed up for this list to learn things from all of you
that know more than I do.
steveZ Rhythm & Hues, Inc.
celia!stevez@usc.edu
"That's not Art Linkletter, that's Mickey Mouse!"
-Art Linkletter, Disneyland opening day
##
Subject: Latest from Impulse (Imagine 3.0)
Date: Mon, 01 Nov 1993 15:10:23 -0500 (EST)
From: "Andrew P. Vogel" <vogelap@ucunix.san.uc.EDU>
Well, contacted Impulse today for my weekly update on Imagine 3.0. Asked them
today when it would be shipping, and the voice said 'Sometime this week.'
I inquired, 'Early, middle, or late this week?'.
'Probably middle,' was the reply.
Ah well... At least they've reduced the 'put-offs' to mere DAYS instead of
months.
Actually, they can take just about as long as they want - I'm just getting
anxious to play with it! My wife's going to be out of town this weekend, so
I have the opportunity to play with it for about 60 hours and write a review
of it for all of us.
Here's to 'sometime this week,' then!
##
Subject: Todays story
Date: Mon, 1 Nov 1993 14:31:30 -0600 (CST)
From: kalb0003@gold.tc.umn.edu
All,
Just got off the phone with Impulse.
In the continuing saga of the ever-elusive 3.0 version of Imagine
I have todays version of how it's all going to go down.
1.Imagine 3.0 is NOT ready (Don't panic, read on...)
Because we have waited so long, Impulse is going to ship version
2.9 of Imagine starting either late tomorrow or early Wednesday.
They still want to finish a few features before dubbing it 3.0.
In my opinion, this is a wise choice: 1) We get the latest version TODAY
(Well, almost). 2) We know that the final version wasn't rushed out
the door with a slew of bugs. 3) If bugs appear in the 2.9 version, there is
still time to correct them before the 3.0 release.
2. The manual is not yet complete (I know you already know that).
As this is being written, the elves at Impulse are writing about
30-40 pages of updated supplementary documentation.
This is only todays version of how things are going. Tomorrow...
Well, we'll just have to wait and see.
Cyrus J. Kalbrener
Digital_Cel
##
Subject: Re: Undersea F/X
Date: Mon, 1 Nov 93 14:11:05 -0800
From: stevez@rhythm.com (Steve Ziolkowski)
>>Also, I have a Ground object with a wave texture applied to it and it
There is a ground plane problem that I have run into... It works as long
as you don't use any global fx like "use genlock", fog and the
backdrop picture. So, try doing your effect with an actual plane
object that switches from the ground plane once you go underwater.
Also, remember that under water, the ceiling is vague and unless
you actually look up, you can probably get away with just using
the global fog as the top of your ocean, once you enter the water.
steveZ Rhythm & Hues, Inc.
celia!stevez@usc.edu
"That's not Art Linkletter, that's Mickey Mouse!"
-Art Linkletter, Disneyland opening day
##
Subject: Re: more complaining, and a question for Glenn
Date: Mon, 1 Nov 1993 18:53:27 -0600 (CST)
From: Daniel Jr Murrell <djm2@Ra.MsState.Edu>
I can understand Michel is feeling like he was ripped off by buying
Imagine a few months before it was given away free. However, from
what I've read on net news, AF made a rush decision to get a hold of
Imagine because a rial mag was getting a toned down copy of Real3D.
So, maybe when you bought Imagine, Michel, the idea of AF giving away
2.0 hadn't surfaced just yet. Besides, it's not like the fact that a
3.0 upgrade was imminent is new news. We've been waiting for what, a
year or so? It's pretty much public knowledge on the net that the new
Imagine would ship RSN, and has been for quite a while.
Ok, my question for Glenn:
I was gonna try T3D for converting a landscape, but can't find it
now. Is it no longer on aminet? Where can I get a hold of it?
Thanks,
Dan
djm2@ra.msstate.edu
##
Subject: Terrain
Date: Tue, 2 Nov 93 02:06:00 BST
From: w.graham6@genie.geis.com
I don't know about Glenn's program, but I know that object converters such
as Interchange and Pixel 3D Pro will not read Terrain objects. Also, they
double in size once they've been run thru Turbo Silver, similar to Forms
objects in Imagine after their symmetry has been broken. Terrain objects
are also saved with Phong shading turned off. But individual features such
as peaks in the object can be set to have their own fractal roughness, a
feature that I always thought was a pretty neat trick. This also is true
for the Wave objects that Terrain generates. It was always a mystery to
me how they could render so smooth but have Phong turned off. Hmmmm. I hope
an updated Imagine version is in the works. I get tired of running them thru
Silver.
##
Subject: Instructional animation
Date: Tue, 2 Nov 93 9:09:17 CST
From: drrogers@camelot.b24a.ingr.com (Dale R Rogers)
I have a queston for the Imagine/Social Philosophies/Personal
Flame Wars Mailing List ;-)
I am an instructor with the Intergraph Corporation. I have an
Amiga at home and am getting up to speed with producing animations
on that platform. I am currently working on a training video at
work and would love to incorporate some animations, produced on my
Amiga, in our corporate video. The executive producer said that
it might be possible. I have the hardware setup to get it to BetaSP
(the DPS Personal Animation Recorder). So getting the output is
no problem. However, I have a production question about getting
an effect and the easier path to follow. Our video services
department here seems to be biased against the Amiga. Therefore,
I want the quality to be the best. It might represent future,
freelance, work if I can pull it off.
The question:
I want to illustrate a particular workflow that a user needs to
follow to use our software. I'd like to use an animated arrow
that bends and turns, like a ribbon, around various objects.
I have both a 3-D product and a 2-D product (Imagine 2.0, and
DPaint IV). Would this effect be easier to accomplish using 3-D
or my 2-D tools? I'll experiment with both, but I have a short
time line to work with and would rather not squander the time
going down a dead end road. I'm not sure if I can get the nice
3-D effect using DPaint. Does anyone have the experience?
Also, if I use Imagine, what would be the best way to model the
arrow so that I can bend it without looking terribly segmented?
B-Spline techology would come in real handy here. However,
working around our polygonal limitation, do I simply need to make
the arrow's sides with a large number of edges to get a smooth look
when bending?
Any ideas?
Dale
____________________________^____________________________
dale r. rogers
Intergraph Corporation
Building Design & Management MailStop: LR24A4
drrogers@b24a.b24a.ingr.com Tel: (205) 730-8294
.
##
Subject: Re: less complaining, and a question for Glenn
Date: Tue, 2 Nov 93 16:37:42 GMT
From: glewis@pcocd2.intel.com (Glenn M. Lewis - ICD ~)
>>>>> "Dan" == Daniel Jr Murrell <djm2@Ra.MsState.Edu> writes:
Dan> Ok, my question for Glenn: I was gonna try T3D for converting a
Dan> landscape, but can't find it now. Is it no longer on aminet?
Dan> Where can I get a hold of it?
Charles Congdon informed me of the following:
T3DLIB R39 may be found in the boing directory tree on
ftp.wustl.edu, parallel the aminet tree, under video/utils. The full
path is something like systems/amiga/boing/video/utils. Aminet is
systems/amiga/aminet.
./video/utils:
-rw-rw-r-- 1 tucker arc_amiga 13728 Feb 15 1993 T3DLIB_R39.readme
-rw-rw-r-- 1 tucker arc_amiga 360692 Feb 15 1993 T3DLIB_Exe_R39.lha
-rw-rw-r-- 1 tucker arc_amiga 98335 Feb 15 1993 T3DLIB_Src_R39.lha
Of course, if people can't find it, just e-mail me and I can
e-mail you the files.
-- Glenn
##
Subject: Re: Instructional animation
Date: Tue, 2 Nov 1993 08:10:37 -0800 (PST)
From: Doug Kelly <dakelly@class.org>
On Tue, 2 Nov 1993, Dale R Rogers wrote:
> I want to illustrate a particular workflow that a user needs to
> follow to use our software. I'd like to use an animated arrow
> that bends and turns, like a ribbon, around various objects.
>
> I have both a 3-D product and a 2-D product (Imagine 2.0, and
> DPaint IV). Would this effect be easier to accomplish using 3-D
> or my 2-D tools? I'll experiment with both, but I have a short
> time line to work with and would rather not squander the time
> going down a dead end road. I'm not sure if I can get the nice
> 3-D effect using DPaint. Does anyone have the experience?
>
> Also, if I use Imagine, what would be the best way to model the
> arrow so that I can bend it without looking terribly segmented?
> B-Spline techology would come in real handy here. However,
> working around our polygonal limitation, do I simply need to make
> the arrow's sides with a large number of edges to get a smooth look
> when bending?
IMHO, a 3D arrow will no-question look better than a DPaint 2D arrow.
I'd try making an arrow, then making it conform to path for the various
morph targets you'll need.
Regarding point/edge/face count: an Amiga artist (Brad?) once said, "If
you're not running out of RAM, you're not trying hard enough." Make your
arrow, then select the whole thing and fracture it, repeating until you
have lots of points. Load it with the rest of the animation, and try
rendering. If you don't run out of RAM, fracture your arrow again and try
it again. When you run out of RAM, go back to the previous version of the
arrow and use it. You'll have the smoothest arrow that your configuration
can render.
It's only when you're already running out of RAM that you need to worry
about point counts.
Good luck, and blow the doors off those video services dweebs!
------------------------------------------------------------------------------
Doug Kelly Information Specialist First Consulting Group
dakelly@class.org (310)595-5291x125 P.O.Box 5161, Los Alamitos,CA 90721-5161
"The difference between genius and stupidity: genius has its limits."
------------------------------------------------------------------------------
##
Subject: Terrain/vista
Date: Tue, 2 Nov 1993 7:42:40 -0700 (MST)
From: LESK@CC.SNOW.EDU
Hi all;
I have been reading about terrain and was wondering what it was?
is it like vista pro? are there new releases of these as I remember them
from some years back I was not aware they were still viable products.
Do you think some of the features they offered will be available in imagine?
Thanks for your responses
Lesk
##
Subject: Re: Imagine 2.0 & AmigaFormat...
Date: Tue, 2 Nov 93 09:48:31 PST
From: 02-Nov-1993 0652 <leimberger@marbls.enet.dec.com>
I took the time to contact Impulse. I asked about the AF disk.
Impulse told me it was a PAL only version that wouls not run on
machines in the US.(might be able to run on a 3000 or 4000 in
PAL mode????) They stated that the giveaway was basically an
amnesty program. Apparently the overseas market has many
shall we say unregistered users. In my opinion there is not much
Impulse can do short of a nasty copy protection scheme, so they
opted for the next best thing. They hope that this giveaway will
encourage everyone to purchase the upgrade at least. Now I know
that their are those that will cry foul, "Why reward the guilty
ect". But think about it. Anybody that has Imagine that never
purchased it got it from SOMEWHERE. They would most likely get
it from Somwhere again. Impulse believes their is a large % of
thse users out their, and I'm afraid they are probably right.
bill
##
Subject: Re: Instructional animation
Date: Tue, 2 Nov 93 13:12:20 CST
From: setzer@comm.mot.com (Thomas Setzer)
> The question:
>
> I want to illustrate a particular workflow that a user needs to
> follow to use our software. I'd like to use an animated arrow
> that bends and turns, like a ribbon, around various objects.
>
> Also, if I use Imagine, what would be the best way to model the
> arrow so that I can bend it without looking terribly segmented?
> B-Spline techology would come in real handy here. However,
> working around our polygonal limitation, do I simply need to make
> the arrow's sides with a large number of edges to get a smooth look
> when bending?
Well in Imagine, isn't there a way to animate an extrude? I can't remember
how, is it an FX? Anyway, I would extrude the "ribbon" along your path you
wish to follow. You could have the head of the arrow, the triangle, follow
the same path, just ahead(touching) of the growing "ribbon". This all depends
on the ability to animate an extrude. Also, this will probably require
lots of tweeking, as always, but the effect would probably be worth it.
Maybe even add an Essence texture to change the color as you move along.
Let us know if you get something to work.
Tom Setzer
setzer@ssd.comm.mot.com
"And of course, I'm a genius, so people are naturally drawn to my fiery
intellect. Their admiration overwhelms their envy!" - Calvin
##
Subject: Re: Terrain
Date: 02 Nov 93 14:22:19 EST
From: John Foust - Syndesis Corporation <76004.1763@compuserve.com>
To: >internet: imagine@email.sp.paramax.com
w.graham6@genie.geis.com writes:
> I don't know about Glenn's program, but I know that object converters such
> as Interchange and Pixel 3D Pro will not read Terrain objects. Also, they
InterChange has supported Terrain's Turbo Silver 2.0 objects since we
added Turbo Silver support, oh, four years ago?
Maybe you're thinking of the "Save" option in Terrain. Terrain has
its own proprietary format for saving landscapes. Over on the right,
there's a menu titled "Silver" that saves Turbo Silver 2.0 format
objects and cells. We've supported TS 2.0 since 1989.
And because of customer demand because of this Terrain compatibility,
we kept the Turbo Silver 2.0 Converter in InterChange Plus v2.0 and
v3.0. With it, you can read and write those old-format objects.
Pixel does not support this file format.
> double in size once they've been run thru Turbo Silver, similar to Forms
Turbo Silver 2.0 has a very different file structure than 3.0. Most
notably, I believe 2.0 didn't have a separate edge list, which would
easily double the file size when converted to 3.0. It has nothing to
do with Forms objects. (BTW, we'd love to convert Forms objects, but
Impulse has never documented them.)
##
Subject: IRC MEETING
Date: Tue, 2 Nov 1993 20:19:49 +0100
From: Hannes Heckner <hecknerh@informatik.tu-muenchen.de>
Perhaps it's the thing that I am ircing from germany and therefore
from another time zone.
I was only once able to establish an irc connection with someone
else (MR. Waif (nickname, of course:-))
Therefore I want to suggest a fixed time for our irc meetings:
It would be 20:00 PM European Time, that would be 14:00 PM EST (if
I am right )
Then to hold the standards for the irc discussions high I suggest
that we regularly announce a special irc discussion referring to
a specific matter.
Examples:
Thursday 14:00 PM EST Discussion about particle systems with Imagine
using T3D lib.
Then everybody could writoe down some notes for this discussion which
would improve discussion qualtiy a lot.
What are your opinions about this ?
Hannes
##
Subject: Imagine 3.0 RELEASED TODAY!
Date: Tue, 02 Nov 1993 17:00:59 -0500 (EST)
From: "Andrew P. Vogel" <vogelap@ucunix.san.uc.EDU>
Well, just got off the phone with the guys at Impulse, and Imagine 2.9 was
released today! \
That comes with some qualifiers...
- It was the PC version. The Amiga version will be shipping in the next
2 or so days...
- It's version 2.9. This version isn't 'really' lacking anything - the Bones
function isn't 100%. Usable, but not EXACTLY the way Impulse wants it...
- AGA compatible, promotable.
- 40+ pages of README for the new version
- 40+ pages of README for the 52 new textures included...
- Shipping US Air Mail.
This boils down to the fact that we'll have this sucker by the end of this
week or early next!
Warm up your mice, men! We've got _IMAGINING_ to do!
As I said before, if it arrives at my door this week or early weekend, I will
write a review for the net.
##
Subject: Re: Instructional Animation
Date: Tue, 02 Nov 93 14:26:00 PST
From: Stethem Ted 5721 <TedS@ms70.nuwes.sea06.navy.mil>
>I want to illustrate a particular workflow that a user needs to
>follow to use our software. I'd like to use an animated arrow
>that bends and turns, like a ribbon, around various objects.
I think Imagine could do this very effectively and fairly easy, once
you get everything set up. The set-up will take awhile but it should be
achievable in several hours. I would make an "arrow" object first, just
straight. I would do this by taking a rectangle into detail editor and
extrude along the path with several sections, maybe 10 or 20. This would be
the tail. I would probably do the same thing with the head, although with
"Make lines" for the triangular outline using 10 points per slant line. I
would add edges and faces manually so the "head" would also have several
sections. Then "Join" these two objects and save as Arrow. Then I would
take that basic arrow and edit it to make the next version, like the arrow
curving around a right corner. Go to "Pick Points", multipick one section
of points and rotate it slightly and move it into place, go to the next
section, rotate it at an increased angle, move those into place, go to the
next section and rotate it slightly, move into place, then pick the head
section and rotate it 90 degrees and move into place. This should now give
you an arrow object with a 90 degree bend, as viewed from the top, although
you could also do this so it has a 90 degree bend pointing down. Save this
object as Arrow1. Do this for all your basic positions, like 90 degree bend
right, 90 degree bend left, 90 degree bend down, up, twisted, stretched, and
so on. Be sure to set Phong in the attributes. Despite the fact that the
object is made of polygons, it will appear remarkably smooth when rendered.
There is also an option called Make Smooth when you are in Pick Edges mode.
The next step would be to create your Path object. This will take a
while to think out and get set up so it will move your arrow object around
your other objects in the way you want. You should make a path with several
axes so you can place rotation movements and different bends where you want
them. Also, it is easier to edit in the Stage Editor.
What I usually do next is go to the Stage Editor and put all my starting
objects into the first frame, get my lights in, and get some idea as to the
scale of things.
After this, go to the Action Editor and increase the frame count to the
size of your animation. Go to the Arrow object line, delete the Arrow
actor, then add the Arrow object. Make the Arrow actor follow the Path.
Create a morph to the next basic Arrow object (maybe Arrow1) at the frame
where you want the arrow to take a turn or elongate or whatever, while still
in the Add mode by going to that next frame, picking the duration of that
sequence, and loading Arrow1 as the object.. Set the transition count for
something like 2 or 3 frames, depending on the duration of that particular
scene. Doing this will allow you to load the basic Arrow objects, at
different points, and Imagine will "morph" from the previous object to the
next object. This "morph" will be very smooth and it will look very fluid.
From there, you will have to play with the Path, set acceleration and
deceleration where you want, and play with the camera so everything looks
right. Maybe you will just have the camera far enough away to see the whole
path without re-aligning and moving.
Anyway, hope this helps. It sounds like a fun project and it is
worthwhile to show the capabilities of the Amiga to the unknowing. Good
luck and happy rendering!
##
Subject: Re: IRC MEETING
Date: Tue, 2 Nov 93 16:38:10 MST
From: pringleg@cuugnet.cuug.ab.ca (Greg Pringle)
>Thursday 14:00 PM EST Discussion about particle systems with Imagine
>using T3D lib.
I wouldn't expect a lot of north american participation if you're
holding chats during the day on business days. Chances are we'll
either have classes or work. If they have to be during the day,
weekends would seem more reasonable.
I've been opening the IML channel after school here also, but haven't
run into anyone either.
Greg
##
Subject: Re: Instructional animation
Date: Tue, 2 Nov 93 16:50:29 MST
From: pringleg@cuugnet.cuug.ab.ca (Greg Pringle)
Hi Dale,
> I have a queston for the Imagine/Social Philosophies/Personal
> Flame Wars Mailing List ;-)
So you noticed it too? B^) I'd suggest people stop and think if
what they have to say is really going to be of interest to the
majority of IML readers before posting, but it's just something
ya got to put up with I guess.
> or my 2-D tools? I'll experiment with both, but I have a short
> time line to work with and would rather not squander the time
> going down a dead end road. I'm not sure if I can get the nice
> 3-D effect using DPaint. Does anyone have the experience?
The dpaint method would probably be quicker, but flashy 3d arrows
would look nicer..
> Also, if I use Imagine, what would be the best way to model the
> arrow so that I can bend it without looking terribly segmented?
> B-Spline techology would come in real handy here. However,
Just make a straight arrow, fracture it until it has a reasonable number
of faces. Reasonable would depend on how much memory, and what resolution
you're rendering to. No use wasting time rendering 10,000 polygons for
a 320X200 picture. Rember to set the phong, and set the corner edges
of the arrow to make sharp. Save the arrow, then make some 'tween'
objects for each stage of the path from this object. Don't add or
remove points from the object. Then, morph between the 'tween' objects.
A good technique for turning the arrow is just to lasso a bunch of points
in pick point mode, then just rotate the points that make up the end of
the arrow. You might figure out how big the arrow is going to get, and
then make sure there'll be enough faces in the original arrow, since it'll
be a waste of time if it starts looking segmented by the time you get
the arrow stretched out.
> Dale
Greg
--
+------------------------------+---------------------------------------+
| Greg Pringle | Amiga VBBS - Multitasking, Windowed |
| pringle@cpsc.ucalgary.ca | BBS'ing! |
| pringleg@cuugnet.cuug.ab.ca | VBBS 14.4K: (403) 284-2048 & 284-5625 |
+------------------------------+---------------------------------------+
##
Subject: Attributes and Memory
Date: 2 Nov 93 21:45:00 EST
From: "J_GEORGE" <J_GEORGE@vger.nsu.edu>
This is a question posed to all, particularly those who have written programs
that hook into Imagine. I've created a futuristic cityscape with a large number
of simple polygons, a few somewhat complex objects, and made use of Essence II
textures as well as a couple of animated brushmaps. Working on a system with
only 7 megs, it was no surprise when parts of the scene started dropping out
of the rendered image. In an attempt to scale things down, I've tailored the
concentration of building objects toward the areas where they are visible to
the camera during the fly-through, reduced the number of lights and reflective
surfaces, and so forth, but still parts of the scene are dropping out. Not
being able to afford any additional RAM at this point (unless someone knows
where I can get some GVP SIMMs cheap), I have to continue scaling the scene
down. Before taking this any further, I wanted to get an idea of what object
elements take up the most memory during rendering, everything from the number
of points in an object to the individual attributes (reflection, filters, etc)
to applied textures and brushmaps.
If anyone has an "itemized list" of how these elements are broken down
memory-wise or at least ballpark figures on these ratios, posting them would
help keep a poor college student from pulling the rest of his hair out. ;-9
Thanks in advance.
I\/Iax I\Iomad
##
Subject: Re: Attributes and Memory
Date: Tue, 2 Nov 1993 22:37:34 -0500 (EST)
From: Udo K Schuermann <walrus@wam.umd.edu>
Forwarded message:
> From imagine-relay@email.sp.paramax.com Tue Nov 2 22:19:20 1993
> Message-Id: <9311030252.AA16387@email.sp.paramax.com>
> Date: 2 Nov 93 21:45:00 EST
> From: "J_GEORGE" <J_GEORGE@vger.nsu.edu>
> Subject: Attributes and Memory
> To: "imagine" <imagine@email.sp.paramax.com>
>
> Before taking this any further, I wanted to get an idea of what object
> elements take up the most memory during rendering, everything from the number
> of points in an object to the individual attributes (reflection, filters, etc)
> to applied textures and brushmaps.
Brushmaps of any type are converted internally to 24-bit. It makes no
difference in terms of memory if they're 1-bit, 5-bit, 12-bit, or 24-bit
images. I don't know enough about the mechanics of rendering, but I would
imagine (no pun intended) that reflection and refraction add to the memory
requirement, depending on the depth you take them to.
If you have a machine with an MMU, you probably would benefit greatly from
a virtual memory system such as GigaMem. It works very well with Imagine
and the render times are, surprisingly, not as badly affected as I thought
they would be when I first started to use that combination. It would
certainly save you from going bald before your time. ;-)
._. Udo Schuermann
( ) walrus@wam.umd.edu
##
Subject: Cluttered Detail Editor
Date: Tue, 2 Nov 1993 23:13:00 -0500
From: changc9@rpi.edu (Cedric Georges Chang)
I am trying to do the not-so-original anim of a plane flying through a canyon.
I want the plane to twist and turn in accordance with the contours of the
ground.
Using VistaPro I imported a section of the Halfdome landscape. What I see in
the detail editor (turning off fastdraw) is a mess; the canyon walls block
everything in the front/right views. With basic knowlege of what the terrain
looked like and some quickrenering, I was able to get a path so that the plane
wouldn't crash into anything. I wasn't able to create the subtle twists and
turns needed for nap-of-the-earth flying.
What I need is a hide points feature, but I understand that only works when
creating an object. It would also have helpful if I could have seen the path
in the perspective view. Since I didn't create the landscape, it isn't grouped
so the canyon walls can be deleted for a less cluttered view and then reloaded
for the actual rendering.
Any suggestions?
Thanks.
Cedric
--
-------------------------------------------------------------------
|Cedric Chang | Grad Student | Aerospace Engineer |
|changc9@rpi.edu | Rennselaer Polytechnic | (Will Work For Food) |
|Amiga 3000 | Institute | |
-------------------------------------------------------------------
##
Subject: Re: Cluttered Detail Editor
Date: Tue, 2 Nov 93 22:43:41 PST
From: ue481@freenet.victoria.bc.ca (Gerard Menendez)
>
>I am trying to do the not-so-original anim of a plane flying through a canyon.
>I want the plane to twist and turn in accordance with the contours of the
>ground.
>
>What I need is a hide points feature, but I understand that only works when
>creating an object. It would also have helpful if I could have seen the path
>in the perspective view. Since I didn't create the landscape, it isn't grouped
>so the canyon walls can be deleted for a less cluttered view and then reloaded
>for the actual rendering.
>
Doesn't Vista-Pro save an object that can be edited in the detail
editor? I've only used it one week-end. If so, take it into the
detail editor, delete alot of points (like maybe select by drag box and
drag the box down so that only the canyon floor is left) use this object
to arrange your whole anim, renaming the original file only when it's
time to render.
Gerard
--
Gerard Menendez Seattle, WA
ue481@victoria.freenet.bc.ca
##
Subject: Re: Terrain files & T3DLIB
Date: Tue, 2 Nov 93 16:50:06 GMT
From: glewis@pcocd2.intel.com (Glenn M. Lewis - ICD ~)
Yikes... I had no idea. Jeff Walkup just pointed out to me
that Terrain files are completely different beasts and that T3DLIB
doesn't have the foggiest idea what they are.
Sorry about that. Thanks for the report, Jeff!
-- Glenn
>>>>> "Jeff" == Jeff Walkup <jwalkup@sfsuvax1.sfsu.edu> writes:
Jeff> Glenn M. Lewis - ICD ~ writes:
>> Yes, if Turbo Silver can load them, then T3DLIB should be able to
>> load them.
Jeff> I just tried it, it doesn't work. Terrain files are "special".
Jeff> Here's the first few lines of one:
Jeff> 00000000: 53494C56 45522049 490A4F42 4A540A01 SILVER II.OBJT..
Jeff> 00000010: 54455252 41494E00 00000000 00000000 TERRAIN.........
Jeff> readwrite (v36, I think) says "invalid chunk on line xxx: 'x'
Jeff> skipped". (xxx = every line of the file, and 'x' is some
Jeff> garbage)
Jeff> -- Jeff Walkup - jwalkup@sfsuvax1.sfsu.edu - Digital Animator /
Jeff> Videographer
##
Subject: RE: Cluttered Detail Editor
Date: Wed, 3 Nov 93 01:41:12 PST
From: 03-Nov-1993 0530 <leimberger@marbls.enet.dec.com>
Subj: Cluttered Detail Editor
>wouldn't crash into anything. I wasn't able to create the subtle twists and
>turns needed for nap-of-the-earth flying.
It would be nice if you could use makepath, and import the path
into Imagine. I think VRL should consider this.
>What I need is a hide points feature, but I understand that only works when
>creating an object. It would also have helpful if I could have seen the path
>in the perspective view. Since I didn't create the landscape, it isn't grouped
>so the canyon walls can be deleted for a less cluttered view and then reloaded
>for the actual rendering.
Would a run through rendered in wireframe buy anything ? If you
have MakePath you could preview the flight in perspective and
then render it in vistapro. This would give you a guide at least.
Then you could use wireframe in Imagine to get quick renders
with your object. Another way would be to render only the object in
imagine and then composit it over the Vista pro anim with a Imaging
program. I use ADPro on the Amiga
bill
/*
bill leimberger Nashua NH.
*/
##
Subject: joe's diner project
Date: Wed, 3 Nov 1993 19:29:41 +0100
From: Hannes Heckner <hecknerh@informatik.tu-muenchen.de>
Hi
I contributed one object to this project and now I want to know
if the whole project has been cancled ?
Is the video tape ready, when will it be ready.
Thanks
Hannes
##
Subject: More Im3.0 stuff
Date: Wed, 03 Nov 1993 15:30:28 -0500 (EST)
From: "Andrew P. Vogel" <vogelap@ucunix.san.uc.EDU>
Okay, spoke today with Arv out at Impulse, and we talked about 2.9/3.0, etc.
The Amiga version will be coming probably Thursday, or maybe Friday. He said
Monday AT THE LATEST to start shipping the Amiga version. The code is all in,
Arv said, but they're scrambling to add OpalVision & other display enhancer
stuff.
It will most likely be on 2 disks.
Dinosaur skin, giant planet (gas planets/rings, etc), planet texture, monster
(looks like open gaping sores on skin. Lovely), statue (really neat sounding),
and many other textures will be included.
The 3.0 release is expected about 30-45 days AFTER the 2.9 release, depending
on the manual and bugs that may be shaken out of the 2.9. I asked Arv about
a ring binding (or spiral binding) on the manual, and he said that the extra
expense is prohibitive, and since people complain (ahem) about a measley
$100 upgrade fee, they probably wouldn't bear _ANY_ additional expense.
Arv told me that the new set of 3.0 textures are REALLY fast (he went so far
as to compare the speed to a 3rd-party set of Imagine textures, and said
Impulse's ROCK), and that there are about 50 of them included.
I asked about an '040 optimized version, and he said the 3.0 release 'may' have
specially optimized code for an 040, but that the 2.9 will not.
Arv also told me that he anticipates the 3.0 version will allow users to
define default directories in Imagine.CONFIG for things like TEXTURES, OBJECTS,
ATTRIBUTES, etc. This has been LONG on my list...
I hope these somewhat disjointed reports are helpful for people. Are they?
Well, doesn't look like I'll be able to work all weekend on the new version and
write my report. Ah well... It'll be here soon enough!
##
Subject: Re: Instructional animation
Date: Wed, 3 Nov 93 14:59:08 PST
From: ua197@freenet.victoria.bc.ca (Christopher Stewart)
> I have both a 3-D product and a 2-D product (Imagine 2.0, and
> DPaint IV). Would this effect be easier to accomplish using 3-D
> or my 2-D tools? I'll experiment with both, but I have a short
> time line to work with and would rather not squander the time
> going down a dead end road. I'm not sure if I can get the nice
> 3-D effect using DPaint. Does anyone have the experience?
>
> Also, if I use Imagine, what would be the best way to model the
> arrow so that I can bend it without looking terribly segmented?
> B-Spline techology would come in real handy here. However,
> working around our polygonal limitation, do I simply need to make
> the arrow's sides with a large number of edges to get a smooth look
> when bending?
>
I'm working on something similar right now and might be able to help.
If you generated multiple versions of the arrow (ie, the bends, twists
etc) by extruding along a path you'd be able to morph between the
versions. As long as you kept the # of segments the same when you
generate the shafts, you should have the same number of triangles for each
version. I've been generating text using this method and have been
suprised at the quality of the results when morphing. Hope this helps.
Christopher
--
....and if there be some harder, better way ua197@freenet.victoria.bc.ca
to salvation than to follow that which we cs833@cleveland.freenet.edu
believe to be good, then are we all damned.
Lord Dunsany, "Dom Rodriguez" (1922). Join the Animation Sig!
##
Subject: RE: More Im3.0 stuff
Date: Wed, 03 Nov 93 16:16:00 PST
From: Stethem Ted 5721 <TedS@ms70.nuwes.sea06.navy.mil>
I have a couple of questions. First, how do you get the upgrade? Without
it being available from a distributor, do you just send a check to Impulse?
Do they need some verification like the original Imagine 2.0 disks (or
maybe the torn-off cover of the Imagine 2.0 manual, ala Commodore
power-up...just kidding!)? Second, now that you have paid out the upgrade
fee for 2.9, does that mean Impulse will automatically send you 3.0 when it
becomes available, probably sans manual, and then automatically send you the
manual when that becomes available? Kind of curious about these details.
----------
From: imagine-relay
To: imagine
Subject: More Im3.0 stuff
Date: Wednesday, November 03, 1993 3:30PM
Okay, spoke today with Arv out at Impulse, and we talked about 2.9/3.0, etc.
The Amiga version will be coming probably Thursday, or maybe Friday. He said
Monday AT THE LATEST to start shipping the Amiga version. The code is all
in,
Arv said, but they're scrambling to add OpalVision & other display enhancer
stuff.
It will most likely be on 2 disks.
Dinosaur skin, giant planet (gas planets/rings, etc), planet texture,
monster
(looks like open gaping sores on skin. Lovely), statue (really neat
sounding),
and many other textures will be included.
The 3.0 release is expected about 30-45 days AFTER the 2.9 release,
depending
on the manual and bugs that may be shaken out of the 2.9. I asked Arv about
a ring binding (or spiral binding) on the manual, and he said that the extra
expense is prohibitive, and since people complain (ahem) about a measley
$100 upgrade fee, they probably wouldn't bear _ANY_ additional expense.
Arv told me that the new set of 3.0 textures are REALLY fast (he went so far
as to compare the speed to a 3rd-party set of Imagine textures, and said
Impulse's ROCK), and that there are about 50 of them included.
I asked about an '040 optimized version, and he said the 3.0 release 'may'
have
specially optimized code for an 040, but that the 2.9 will not.
Arv also told me that he anticipates the 3.0 version will allow users to
define default directories in Imagine.CONFIG for things like TEXTURES,
OBJECTS,
ATTRIBUTES, etc. This has been LONG on my list...
I hope these somewhat disjointed reports are helpful for people. Are they?
Well, doesn't look like I'll be able to work all weekend on the new version
and
write my report. Ah well... It'll be here soon enough!
##
Subject: RE: More Im3.0 stuff
Date: Thu, 4 Nov 93 08:55:13 EST
From: David Watters <watters@cranel.com>
> I asked about an '040 optimized version, and he said the 3.0 release 'may'
> have specially optimized code for an 040, but that the 2.9 will not.
Give me a break. Do this guys really think anyone believes they care about the
Amiga market?
My Lightwave 3.0 gave me a 8x increase in performance over 2.0 running on the
same '040 processor. How can impulse justify not taking advantage of this in
the last year since '040s have been in Amigas.
> I hope these somewhat disjointed reports are helpful for people. Are they?
Yes.
_ ___
David ~ |_|,--' |@,__
Watters ~ ( )-_______-()`-
--
David R. Watters (watters@cranel.com) Cranel Inc. Development & Engineering
"Porsche. The very name is, to many, the last word in sports cars. Any car
blessed with these magic seven letters is sure to be the very best. Period!"
- Car and Driver, January 1993
##
Subject: Imagine 3.0 compatibility
Date: Thu, 04 Nov 93 09:16:00 PST
From: Stethem Ted 5721 <TedS@ms70.nuwes.sea06.navy.mil>
In their many phone calls to Impulse, has anybody asked about
compatibility between Imagine 2.0 and 3.0? Primarily, object compatibility
including detail, cycle, and forms, texture compatibility, special effects,
attributes, stage and action scenes, projects, etc. I seem to remember a
message awhile back that said Essence would no longer be compatible with
Imagine 3.0. I guess these questions will be answered anyway with the
release of 3.0 and it is too late to change code just to maintain backward
compatibility. Is 3.0 going to mean that you throw out 2.0 and everything
that goes with it?
##
Subject: arrow anim(was Re: Instructional animation)
Date: Thu, 4 Nov 93 12:25:27 CST
From: setzer@comm.mot.com (Thomas Setzer)
Well, last night I gave my suggestion a whirl. It worked very nice and was
very easy to do.
Just create your path in the detail editor using spline paths. Also create
the outline of the tail of the arrow. I used a square, but I suggest a circle
because I was unable to make the edges sharp(even though I used make sharp)
at the later phase. Group 'em, with the path as the parent. After you've
set up your stage, go to the action editor and add the GROW FX to the paths
fx. Ta da! Thats it! You can practice using extrude in the detail editor
to see what you object will look like.
Now just add the pointy part of the arrow, have it follow the same path. You
may need to fool around with the axis location and alignment on the pointy part
so that it starts off in the right place.
Problems: The tail seems to grow from the head. If you add a texture, the
textures will not move and the head seems to be "drawing" the tail. It looks
really nice, though.
Problem2: The edges of the box that was extruded were rounded and make sharp
didn't take effect.
I came up with an idea for an FX while I was playing with this. I will try
and implement it when(if) Impulse releases their 3.0 FX spec.
Since this is not a morph, you can't have a starting and ending texture, so
you can't animate textures. But if you had an FX to add on top of the
grow FX, say "POSITION AND ALIGN TEXTURE AXIS FX" then you could get some
motion out of the texture. If you follow the above steps using grow, you
will know what I mean.
Idea:(besides edible paper(movie reference)) try adding a grow fx on top of
a grow fx. One growing, one shrinking. Might be a wierd effect?!?!?
Tom Setzer
setzer@ssd.comm.mot.com
"And of course, I'm a genius, so people are naturally drawn to my fiery
intellect. Their admiration overwhelms their envy!" - Calvin
##
Subject: Imagine in the Movies
Date: Thu, 4 Nov 93 13:14:29 CST
From: setzer@comm.mot.com (Thomas Setzer)
Well, here is my second contribution to the "Stop the Flames" campaign.
A while back I mentioned a friend of mine doing some interesting work
with Imagine. Well, I convinced him to let me brag on him and tell his
story(he was hesitant at first because he was having problems with Impulse
and was considering switching products).
He is working for Marimac Valley Films on a film called Radio Land Murders.
Marimac(not sure if that is spelled right) is a subsidiary(or something) of
George Lucas Films. Anyway, my friend Joey worked with the production
designer to design the sets. Normally an artist is hired to draw the set
from several different angles. The production designer will say, make
those curtains blue, or make that an easy-boy chair instead, etc. This
time, however, Joey was hired and did the whole set in 3d in Imagine.
As you can imagine, this has great advantages over the traditional way of
doing things. Camera positions can be tried. Lighting can be experimented
with. Different angles can be tried without worring about the artist forgeting
something in the background, etc.
After the production designer was happy with everything, Joey created several
animations of fly throughs and different angles and recorded them to tape
with a PAR. The tape was then used to give the builders a view of how the set
should look.
George Lucas also came into town(Wilmington, NC) and met with the production
designer and Joey(he got to shake Georges hand! OOOOHHH) George viewed the
tape and was apparently very enthusiastic about it. He said that he hoped one
day he would be able to script out a whole movie using this technique, not
just design the sets.
Joey also is involved in any screen shots(ie, computer screens) that you
might see in this movie(not necessarily Imagine). He is also currently doing
this for several other movies and TV shows currently being filmed in
Wilmington. Radio Land Murders was the most exciting, however, because he got
to meet George Lucas(Ooooooh! Gee, I wish I had been there).
Joey has also done some TV ad work for which he won an Ady(*the* ad award, don't
know how its spelled though). He won local and regional and it went on to
the nationals.(don't know how it faired there)
So, if anyone tells you Imagine isn't for professional use...well, you get
the idea.
BTW, the name of Joeys company is J&K Graphics. You may be hearing more
about them in the future(and maybe not;)
Tom Setzer
setzer@ssd.comm.mot.com
"And of course, I'm a genius, so people are naturally drawn to my fiery
intellect. Their admiration overwhelms their envy!" - Calvin
##
Subject: Imagine on film
Date: Thu, 04 Nov 93 23:08:00
From: greg.tsadilas@hofbbs.com
I'm new to the Internet <spare the hello's :) > so if this has been mentioned I
appologize.
Imagine played an integral part in the layout, design, animation & creation
of the ballroom dance scene in Disney's "Beauty and the Beast". Just in case
you were wondering what was used to create it, now ya know.
GreG tsadilas
greg.tsadilas@hofbbs.com
##
Subject: Cycle Memory question
Date: Fri, 5 Nov 1993 09:01:02 -0500
From: "Rob (R.D.) Hounsell" <hounsell@bnr.ca>
Folks,
If I have an animation sequence that has a static object for a period of
time, and then suddenly starts moving (via a cycle object) is it better (uses
less memory / CPU / disk access / time ...) to, for the first static sequence:
i) use the cycle object for the entire static sequence but don't cycle it:
i.e. load it using zero cycles (and whatever phase is appropriate).
or
ii) use a snapshot (e.g. a grouped object) of one of the cycle cels for the
static part.
I guess what I'm asking is "is there any overhead using a single cel of a
cycle for a static object rather than a non-cycle grouped object".
Thx
Rob
--
+-----------------------------------------------------------------------------+
| Rob Hounsell BNR WAN: HOUNSELL@NMERH53 |
| Team Leader: UNIX INTERNET: HOUNSELL@BNR.CA |
| System Performance: PHONE: (613) 765-2904 |
| Paradigm Club Design Team. Dept. PS27 ESN: 395-2904 |
| Northern Telecom Public Switching |
+-----------------------------------------------------------------------------+
##
Subject: Re: Imagine on film
Date: Fri, 5 Nov 93 15:42:00 GMT
From: glewis@pcocd2.intel.com (Glenn M. Lewis - ICD ~)
>>>>> "Greg" == greg tsadilas <greg.tsadilas@hofbbs.com> writes:
Greg> I'm new to the Internet <spare the hello's :) > so if this has
Greg> been mentioned I appologize.
Well, "Hello" anyway! :-)
Greg> Imagine played an integral part in the layout, design, animation
Greg> & creation of the ballroom dance scene in Disney's "Beauty and
Greg> the Beast". Just in case you were wondering what was used to
Greg> create it, now ya know.
Were you involved with any part of it? Or is this from an
article in Computer Graphics World, or what? If you have more
details, please fill us all in. Maybe from the new "Disney's Art of
Animation: From Mickey to Beauty and the Beast" book?
Feel free to share on the IML, as graphics topics would be a
refreshing change.
-- Glenn
##
Subject: Re: Imagine in the Movies
Date: Thu, 4 Nov 1993 17:05:32 -0500 (EST)
From: "David A. Rollins" <drollin@seq.cms.uncwil.edu>
Hello Thomas Setzer,
>
> Well, here is my second contribution to the "Stop the Flames" campaign.
>
> A while back I mentioned a friend of mine doing some interesting work
> with Imagine. Well, I convinced him to let me brag on him and tell his
> story(he was hesitant at first because he was having problems with Impulse
> and was considering switching products).
>
> He is working for Marimac Valley Films on a film called Radio Land Murders.
> Marimac(not sure if that is spelled right) is a subsidiary(or something) of
> George Lucas Films. Anyway, my friend Joey worked with the production
> designer to design the sets. Normally an artist is hired to draw the set
> from several different angles. The production designer will say, make
> those curtains blue, or make that an easy-boy chair instead, etc. This
> time, however, Joey was hired and did the whole set in 3d in Imagine.
I live in Wilmington and know Joey Jarman. As a matter of fact, the first
time we met was on the day that CBM Computers unveiled their very first
Amiga 1000 demo unit way back in 1885. Joey was a MessyDosHead and
Macphile back then. One look at the Amiga and two weeks later, Joey
was selling Amigas for CBM. My boast is that I was the very first person
in Wilmington, North Carolina to ever operate an Amiga. Joey was the
second.
Not only has Joey worked on the project you mentioned above, but he has
also produced a set of local environmental recycling commercials that
air several times daily on all the local network affiliate channels.
I and my partner, Richard Harris, have also created station IDs and
program intros for locally produced TV programs.
My only negative comment concerning Joey is that he is not very
original in his objects, resorting to using the tutorial man from
Imagine 1.1 in one of his commercial animations as well as using the
rendered images from the supplied .dem files in SceneryAnimator instead
of designing his own.
Still, Joey has done much to enlighten the film industry in Wilmywood
(our nickname for Wilmington with regards to Carolco Studios here).
You may not know that the Amiga was widely used in the production of
every episode of The Young Indiana Jones Chronicles, which was also
produced here in Wilmington. So was Total Recall, The Teenage Mutant
Ninja Turtles, Mario Bros., and a host of nationally distributed
TV commercials. I also believe that Joey created a few of the
animations that were in Stephen King's Golden Years mini series.
>
> As you can imagine, this has great advantages over the traditional way of
> doing things. Camera positions can be tried. Lighting can be experimented
> with. Different angles can be tried without worring about the artist forgeting
> something in the background, etc.
>
> After the production designer was happy with everything, Joey created several
> animations of fly throughs and different angles and recorded them to tape
> with a PAR. The tape was then used to give the builders a view of how the set
> should look.
>
> George Lucas also came into town(Wilmington, NC) and met with the production
> designer and Joey(he got to shake Georges hand! OOOOHHH) George viewed the
> tape and was apparently very enthusiastic about it. He said that he hoped one
> day he would be able to script out a whole movie using this technique, not
> just design the sets.
George Lucas is no stranger to Wilmington. Wilmington has become the home
of quite a few producers and stars over the past 10 years since the
film studios were built.
For those of you who are not familiar, Wilmington got on the movie
map when Dino Delaurentiis built his studios here in 1983. He produced
such big screen "classics" here as KING KONG (the one where King Kong
climbs the World Trade center), Firestarter, Silver Bullet, Cape Fear,
Places In the Heart, Chinatown, and many many more. Delaurentiis went
bankrupt and his studios became North Carolina Film Studios. Very few
films were made or released during this period due to Chapter 11
proceedings. Finally, Carolco and to a lesser degree Tri-Star bought
the studios a few years ago. The new and most prosperous productions
at Carolco are the current Matlock TV Series, the Weekend at Bernies,
Mario Bros., Golden Years, Cat's Eye, Harlem Nights, MalcolmX, Pet
Sematary, Dune (which was a DeLaurentiis production, and one or two
other Swarzenegger films. Some of the films mentioned above were
produced during the DeLaurentiis years but were not released until
Carolco came along. Also, the movie Traxx was made here as well as some
segments of the Nightmare on Elm Street series.
Amigas have played a part in many of these movies. Oh, I almost forgot...
Some of the RobocopII footage was shot here and Amigas were part of that as well.
>
> Joey also is involved in any screen shots(ie, computer screens) that you
> might see in this movie(not necessarily Imagine). He is also currently doing
> this for several other movies and TV shows currently being filmed in
> Wilmington. Radio Land Murders was the most exciting, however, because he got
> to meet George Lucas(Ooooooh! Gee, I wish I had been there).
>
> Joey has also done some TV ad work for which he won an Ady(*the* ad award, don't
> know how its spelled though). He won local and regional and it went on to
> the nationals.(don't know how it faired there)
>
> So, if anyone tells you Imagine isn't for professional use...well, you get
> the idea.
The majority of Amiga users in Wilmington are using their Amigas for
professional purposes.
Richard and I have finished our object set and it will be released
as soon as I finish the documentation and create the installer
routines. Look for it real soon.
>
> BTW, the name of Joeys company is J&K Graphics. You may be hearing more
> about them in the future(and maybe not;)
You will hear more about Joey and other Amiga production companies in
Wilmington such as, Cybergraf Synthiotics, Sly Curmudgeon Productions,
Pine Valley Productions, Propel Animations, and Desktop Video Productions.
My companies, Cybergraf Synthiotics and Desktop Video Productions are
currently involved as consultants and artistic contributors to a
film in research called Neon Warriors, written by Dr. Richard Hill.
It is a story of an inner city high school held "hostage" by a neigh-
borhood youth gang alliance. The school board hires ex-military
anti-terrorist experts as teachers. These "teachers" team up with the
students who want to learn and they systematically eliminate the
gangs with computer wizardry, martial arts expertise, "practical joke"
scenarios and just plain dazzling intellect.
Dr. Hill, by the way is a very accomplished author who has written for
a variety of national and international magazines. He has written an
episode of Seinfeld that is currently being reviewed as a candidate
for airplay next season. He is an ex-Navy Seal and a former Secret
Service operative during the Kennedy and Johnson years.>
>
##
Subject: Im3.0 MORE
Date: Fri, 05 Nov 1993 14:02:44 -0500 (EST)
From: "Andrew P. Vogel" <vogelap@ucunix.san.uc.EDU>
Just a short note today...
Imagine 2.9 'should' be shipping Monday, November 8 1993.
##
Subject: Comp Graphics - The Art, The Science, The Market
Date: 5 Nov 93 15:16:00 EST
From: "J_GEORGE" <J_GEORGE@vger.nsu.edu>
Some questions posed to all:
How many subscribers to this list are (A) working in Computer Graphics and
Animation or (B) students studying to break into these fields? I'm just
curious, for several reasons:
[1] With this list, Impulse bashing aside, we have an excellent resource here
that is still unavailable to a large percentage of the world - connectivity.
A constant flow of information from sources across the globe, no matter what
the field, lends itself to accelerating learning and problem solving. What
might take days (or weeks) to figure out can sometimes be solved within hours
or minutes, strictly because often the stumbling blocks are not unique.
Someone else has come along before you, hit the problem, and found the
solution; or in other cases, you may be that person and lend invaluable aid to
another that has reached that point. At its best, this is an inside track,
maybe even an 'edge' on the other competition out there. At its worst, it's
merely a steady stream of mail in the mailbox to sift through.
[2] Being a computer graphics artist striving to break into the market, I'm
constantly gathering information and sources for my personal library and
archives. I see this as a necessary practice, if nothing else, as a means to
insure the greatest odds are in my favor as far as landing that job in the
field to not only support myself but get my hands on the high-end equipment.
To cut out more of the socio-philosophical rambling, I'm beginning to
see a definite seperation between those that use Imagine (or any other
rendering engine) as a hobby and those that are using it to make ends meet (or
as a stepping stone in that process). The time and energy I've seen spent in
posts and responses and bickering over the actions of Impulse almost warrant
another mailing list unto itself. People in serious production, whether for
art or for profit, don't have time for the whining. Time is of the essence.
Correct me if I'm wrong, but it'd seem like only hobbyists and ethusiasts have
time for this behavior. Nothing wrong with being a hobbyist, but such a person
doesn't have as much to lose as someone that is into this stuff for a living.
So... what can be done.. ?
Granted, a little civil disobedience is always a good thing, but when
it starts to consume a good portion of the mail-traffic, it's taken on a
tangent of it's own. Solution? What about the possibility of making another
mailing list, one for, let's say, IMPULSE, and letting it go on its merry way
for the complaints/criticisms/congratulations. Just like this somewhat
lengthy letter, it's not pertaining to production, techniques, or advice.
I can't speak for anyone else, but personally production techniques/advice
and resources is what this is all about, both giving and receiving aid.
Anything else [constant complaints] is irrelevant.
I noticed that some others have posted similar pleas for the complaints
and Impulse bashing to end. I avoided contributing to those threads, hoping
they'd go away in a few days, but as we can all see, they haven't.
If the complaints continue, how about we consider making a totally
seperate list for those that have alot of fuel for flames... or creating a
special mailbox that all complaints can be sent to. Otherwise, this kind of
discussion will eventually be the downfall of a mailing list like this. The
people that are serious about their craft will get fed up and unsubscribe,
which will definitely affect the quality of the information resources
currently found in the list.
Apologies for the long speech on the soap box. I got to a point where
I couldn't hold back from speaking out any longer. Thanks for your patience.
Some food for your Brain Cels,
I\/Iax I\Iomad
##
Subject: Re: Imagine in the Movies
Date: Fri, 5 Nov 1993 18:37:27 -0500 (EST)
From: "David A. Rollins" <drollin@seq.cms.uncwil.edu>
Hello Dale R Rogers,
>
> |
> |The majority of Amiga users in Wilmington are using their Amigas for
> |professional purposes.
> |
>
> Is Imagine used on a regular basis for your projects?
>
> Dale
Yes, Imagine is what we use most. We also use Lightwave via the
Newtek VideoToaster when Imagine just cannot get the job done.
We also use Caligari via the IV24 system.
AdPro and Glenn Lewis' T3DLIB also are essential tools.
##
Subject: Imagine on film
Date: Fri, 05 Nov 93 23:08:58
From: greg.tsadilas@hofbbs.com
Imagine and "Beauty and the Beast"
In a conversation with Mike Halvorson of Impulse, It was brought to my
attention that Imagine played an integral part in coreographing, defining the
character's positions, dance sequences, camera angles, etc.
To what degree the finished sequence includes Imagine output was not
discussed.
Hopefully this reinforces or clears up my last posting.
GreG tsadilas
greg.tsadilas@hofbbs.com
##
Subject: Nothing about 3.0 or upgrades
Date: Sat, 6 Nov 1993 14:55:36 -0600 (CST)
From: Peter Garza <pmgarza@ccwf.cc.utexas.edu>
Well I'm nice and toasty against the cold from all the warm fires
about the AF issue. Anyway...
I was wondering if anyone has tried to make a non-texture, somekind of
texture or brushmap that will fade out any other textures and let the
original values of the objects show through. Maybe something that
takes the Fade setting in Essence textures to zero in some pattern.
Hmm, maybe an example would be better.
Say I want a spot of diamonddeck with a little rust (fractal color)
which flattens out to a certain radius. Sortof like using ringsm to
flatten the diamonddeck and fade out the rust, so the rest of the
object is left with the original attributes given.
If that's clear, is there anything like this on Essence II (kick
myself for not getting it yet) or Essence Pro? Thanks for any
responses.
Peter Garza
pmgarza@ccwf.cc.utexas.edu
##
Subject: Comments on Imagine 2.9 ?
Date: Mon, 8 Nov 1993 01:21:11 +1100 (EST)
From: Nikola Vukovljak <nvukovlj@ucc.su.OZ.AU>
So, has anyone received it yet ? Either PC or Amiga ? Comments ?
Nik.
##
Subject: Re: Comments on Imagine 2.9 ?
Date: Sun, 7 Nov 1993 09:47:59 -0600 (CST)
From: Cliff Lee <cel@tenet.edu>
On Mon, 8 Nov 1993, Nikola Vukovljak wrote:
> So, has anyone received it yet ? Either PC or Amiga ? Comments ?
I had expected to recieve it already. From the list, I thought the PC
version had gone out Tuesday. Maybe not?!? I have not recieved it yet.
-=-=-=-=-=-=-=-=-=-=-=-=-
Cliff Lee
cel@tenet.edu
"Everything will work out if you let it!" Cheap Trick
##
Subject: New User-PC Version
Date: Sun, 7 Nov 1993 12:54:30 -0800 (PST)
From: Mark Boulding <boulding@well.sf.ca.us>
Just wanted to say hello to everyone, make an observation, and ask a question.
I am a new user of Imagine 2.0 for the PC, and have already spent many
hours burning the midnight oil doing a simple logo animation for a corporate
affiliate of the firm I work for.
The observation is that this seems to be a pretty flame intensive mailing
list.
Shouldn't we all be helping each other with this powerful (but difficult)
software, rather than expending so much bandwidth on mutual recriminations?
After all, this isn't some newsgroup...
The question is, do the Essence packages work with Imagine-PC? I have an
Amiga 2000HD and Cross-DOS, so I could easily copy the binaries from one
machine to the other.
I'm amazed at how much Imagine-PC makes my PC look like an Amiga, and I'm
anxiously awaiting 2.99 (or whatever they're calling it now).
Mark
--
+===================================================================+
| Mark Boulding c/o FB&T | Internet: boulding@well.sf.ca.us |
| 750 17th St., NW, Ste. 1100 | CIS: 72401,2400 Prodigy: TDRC42A |
| Washington, DC 20006 | AOL: boulding GEnie: M.BOULDING |
+===================================================================+
##
Subject: Re: Imagine in the Movies
Date: Sun, 7 Nov 93 13:10:30 PST
From: Joey_F_Jarman@cup.portal.com
My name is Joey Jarman and there have been a few posts recently about myself
using Imagine for film and television productions in Wilmington, NC.
Since there is a few inaccurate statements here and there, I will describe
what I am doing with Imagine (that is the purpose of the IML, isn't it?)
and disregard the personal comments.
Some background for those that care:
I have been running a 9 to 5 business for almost 4 years earning every dollar
from 3D animation and special effects for the film and television industry.
I have been involved in computer graphics for over 15 years and have a B.S.
in computer science (big deal). I know several languages fairly well and
have written some of my own 3D routines on the Amiga and SG platform.
I primarily use Imagine for modeling and animation (the point of this post).
Some recent credits include:
"Radioland Murders" which is currently still under production, "The Young
Indiana Jones Chronicles," "Lovejoy Mysteries," and (8) Recycling commercials
that were complete 30 second spots done entirely with Imagine. I also did
some legal demonstration animations of nuclear fuel rod extrusion and nuclear
rod reaction for General Electric, a 30 second, animated fair spot that won a
couple of awards, a few program intros, the typical lawyer commercials,
flying logos, blah, blah, blah.
As Setzer posted earlier, I am currently working on a feature film called
"Radioland Murders" created by George Lucas and set in the 1930's. The
producer sent Industrial Light and Magic, Western Images, and myself a test
project to see if there would be some practical applications for using 3D
animation.
The project consisted of inputting sets into the computer and rendering
perspective animations for the production designer (Gavin Bocquet) to use.
They were able to save money by moving walls, changing windows, testing color
schemes, etc. before the sets were actually built. I created over 52Mb
of original 1930's set interiors and exteriors including an auditorium with
almost 50 seats, suspended control booths, working doors, a revolving stage,
human figures for scale, etc. There were over 100,000 edges and 22 light
sources in this scene alone. This was not just an auditorium interior either.
There was an Executive Corridor with glass brick lighting effects, stainless
molding and corner treatments, working doors, skylights, etc. There was also
a Public Corridor, a Viewing Corridor, a Workers' Corridor and more. All on
multiple levels with stairs connecting the various levels. I also created an
entire city with over 20 city blocks of buildings. The centerpiece was a 25
story Radio Tower with over 50,000 edges worth of detailed 1930's styling.
Needless to say this was a challenge for one person to take on but I felt
confident with Imagine and was not about to turn down a chance to 'one
up' the big boys.
This stuff must be EXACTLY to scale and rendered in 2.35:1 aspect for film.
Not to mention 18-20 hour/6-7 days a week to meet the production schedule.
After inputting the sets, Gavin would come to my office for about 2 hours
at a time to pick camera angles and positions for the shots used in the
film. We also worked out some key angles for some special effects, and
stunts. I created over 10,000 frames of animation for which part was used
here at the studios and part was sent back to ILM for constructing the matte
and CG shots. I am proud to say that I was the only one of the three
companies to complete the project.
I did get the opportunity to meet George Lucas. I had a meeting together
with the producer, the director, the production designer, myself, and George.
He was impressed with my work and was very complimentary. He even suggested
some new ideas for using these techniques. He did ask what machine I was
using and seemed surprised when I told him I used several networked Amigas.
The producer is very excited about this new technique and has asked me to
work on 4 other projects as well. I'm not permitted to say what they are
yet (use the force) but I believe it is safe to say that anyone invloved
in graphics would give their eye teeth to work on them. It is definitely
beyond my wildest expectations.
Working with this production has been a great experience. They have an
incredible video room set up with a full blown THX sound system, full screen
videoconferenceing in 24 bits (I got to see George Lucas sitting in his
office in LA while the producer and he discussed the production, very
impressive), and the latest portable computers and phone setups. They really
put the new technology to use and it is fascinating to be a part of it.
Well, that's my "What are you doing with Imagine?" update for now.
Joey Jarman
J&K Graphics
##
Subject: Re: Comments on Imagine 2.9 ?
Date: Sun, 7 Nov 1993 16:39:29 -0500
From: Jason B Koszarsky <kozarsky@cse.psu.edu>
Assuming Imagine 2.9 is shipping, I would not expect to receive it right away.
Impulse probably has tons of copies to ship and it would be unreasonable to
expect them to get ALL the uprgrades orders out in a couple days.
Jason K.
CYBERNETIC EXPRESSIONS
##
Subject: Re: New User-PC Version
Date: Mon, 8 Nov 1993 08:03:20 -0800 (PST)
From: Doug Kelly <dakelly@class.org>
No, unfortunately Essence has to be rewritten to work on the PC. Last
word we had from Steve Worley (correct me if this has changed, Steve!) was
that they were working on the PC version.
Since Impulse has reportedly rewritten Imagine so it won't use Essence
textures in 3.0, I'm not sure what Apex is going to do now. Wait and see
how they can rewrite Essence to work with 3.0, I'd guess.
------------------------------------------------------------------------------
Doug Kelly Information Specialist First Consulting Group
dakelly@class.org (310)595-5291x125 P.O.Box 5161, Los Alamitos,CA 90721-5161
"The difference between genius and stupidity: genius has its limits."
------------------------------------------------------------------------------
##
Subject: Re: New User-PC Version
Date: Mon, 8 Nov 93 17:28:16 GMT
From: glewis@pcocd2.intel.com (Glenn M. Lewis - ICD ~)
>>>>> "Doug" == Doug Kelly <dakelly@class.org> writes:
Doug> No, unfortunately Essence has to be rewritten to work on the PC.
Doug> Last word we had from Steve Worley (correct me if this has
Doug> changed, Steve!) was that they were working on the PC version.
Doug> Since Impulse has reportedly rewritten Imagine so it won't use
Doug> Essence textures in 3.0, I'm not sure what Apex is going to do
Doug> now. Wait and see how they can rewrite Essence to work with
Doug> 3.0, I'd guess.
Right on the money, on both paragraphs. The Essence-PC port
has been extremely challenging... nothing to report yet. :-(
-- Glenn
##
Subject: I got mine...
Date: Mon, 8 Nov 1993 15:39:14 -0600 (CST)
From: Cliff Lee <cel@tenet.edu>
I got the PC version of 2.9 in my hands. I have not loaded it yet, but
have noticed that there is a typo on the 1st page of the pre-release
manual. It says "IMAGINE 2.0 Pre Release Manual". Should read "IMAGINE
3.0 Pre Release Manual". Lets hope this is not an indication of "quality"
of the rest of the package.
-=-=-=-=-=-=-=-=-=-=-=-=-
Cliff Lee
cel@tenet.edu
"Everything will work out if you let it!" Cheap Trick
##
Subject: From the Document enclosed with 2.9
Date: Mon, 8 Nov 1993 15:52:57 -0600 (CST)
From: Cliff Lee <cel@tenet.edu>
"Here are some of the features that are not included in this version and
WILL be in the next version.
BONES
KINEMATICS
IMPROVED ANTIALIASING
NEW LIGHT SOURCES WITH IMPROVED SHADOWS
SHADOW MAPPING
NEW STAGE and ANIMATION TOOLS
Remeber that this is only a short lived situation that will be taken care
of very soon"
There is also a section called "SPECIAL NOTE" which details an anual
service fee of $100/year to get continual updates (defined as 4/year)
instead of the single BIG upgrade that hold everything up.
-=-=-=-=-=-=-=-=-=-=-=-=-
Cliff Lee
cel@tenet.edu
"Everything will work out if you let it!" Cheap Trick
##
Subject: Re: From the Document enclosed with 2.9
Date: Mon, 8 Nov 93 14:29:17 PST
From: Byrt Martinez <martinez@srcsvr.cup.hp.com>
content-type:text/plain;charset=us-ascii
mime-version:1.0
>There is also a section called "SPECIAL NOTE" which details an anual
>service fee of $100/year to get continual updates (defined as 4/year)
>instead of the single BIG upgrade that hold everything up.
Gee, do you think it's a years subscription to Amiga Format? ::Ducking::
--
@******************************************************************************@
* "When you point your finger 'cause your *
* Byrt Martinez plan fell through, you've got 3 more *
* - Non-Christian 8^P fingers pointing back at you." *
* (Fred 12:45) - Dire Straits *
* *
* martinez@srcsvr.cup.hp.com, byrt@shell.portal.com *
@******************************************************************************@
##
Subject: Re: From the Document enclosed with 2.9
Date: Mon, 8 Nov 93 20:56:43 EST
From: Steve J. Lombardi <stlombo@eos.acm.rpi.edu>
>
> "Here are some of the features that are not included in this version and
> WILL be in the next version.
>
> BONES
> KINEMATICS
> IMPROVED ANTIALIASING
> NEW LIGHT SOURCES WITH IMPROVED SHADOWS
> SHADOW MAPPING
> NEW STAGE and ANIMATION TOOLS
>
Hmmm... These are all of the features I've been waiting for in 3.0.
So what exactly is in 2.9 that isn't in 2.0??? I'm real curious. If
2.9 mearly gives me a working slice I'll be extremely disappointed
as I fear that it will be quite some time before the official 3.0 is out.
So please Cliff, let us know what is new in 2.9 while we wait.
> Remeber that this is only a short lived situation that will be taken care
> of very soon"
| Hey Beavis. Essence-II's Crumpled texture
steve lombardi | really KICKS ASS. Mhhh huh. Yea. And those space
stlombo@acm.rpi.edu | textures don't suck either. Huh.
| -From a really new Beavis and Butthead
>
##
Subject: Re: I got mine...
Date: Tue, 9 Nov 1993 17:26:22 +1100 (EST)
From: Nikola Vukovljak <nvukovlj@ucc.su.oz.au>
On Mon, 8 Nov 1993, Cliff Lee wrote:
> I got the PC version of 2.9 in my hands. I have not loaded it yet, but
> have noticed that there is a typo on the 1st page of the pre-release
> manual. It says "IMAGINE 2.0 Pre Release Manual". Should read "IMAGINE
> 3.0 Pre Release Manual". Lets hope this is not an indication of "quality"
> of the rest of the package.
>
Ah, but we've grown to expect 'inventive' English use from Impulse, not
to mention their lovely manuals. :-)
I really hope that the new manual will be written by someone literate.
Why they didn't just get Steve Worley to do it is beyond me.
Never mind. Looking forward to 'Understanding Imagine 3.0'. :-)
Nik.
##
Subject: Re: From the Document enclosed with 2.9
Date: Tue, 9 Nov 1993 17:32:07 +1100 (EST)
From: Nikola Vukovljak <nvukovlj@ucc.su.OZ.AU>
On Mon, 8 Nov 1993, Cliff Lee wrote:
> "Here are some of the features that are not included in this version and
> WILL be in the next version.
>
> BONES
> KINEMATICS
> IMPROVED ANTIALIASING
> NEW LIGHT SOURCES WITH IMPROVED SHADOWS
> SHADOW MAPPING
> NEW STAGE and ANIMATION TOOLS
>
> Remeber that this is only a short lived situation that will be taken care
> of very soon"
I'd hope so. What is new ? These sund like all the major features that
3.0 was supposed to have. So, are particle system and brushtacking
present ?
>
> There is also a section called "SPECIAL NOTE" which details an anual
> service fee of $100/year to get continual updates (defined as 4/year)
> instead of the single BIG upgrade that hold everything up.
>
Owners of Image Master stand up. Does this sound like Black Belt or what ?
:-)
Nik.
##
Subject: RE:New User-PC Version
Date: Tue, 09 Nov 93 09:54:37 TUR
From: Erdem ERTAN <E73412@vm.cc.metu.edu.tr>
--Is it possible to use Essence in PC?
At last a question I can answer.
I think it is impossible.As I am a coder on Amiga,I usually look the
codes or structure of the datas.Essence textures are working like small
codes.They have their own hunk table and data structure(Also the code in
it is for 680x0 assembler not 80x86 based machines).So it is impossible
to use AMIGA vesion Essence in PC.Try to use Amiga.
------------------------------------------------------------------------
- ERDEM ERTAN E73412@vm.cc.metu.edu.tr -
------------------------------------------------------------------------
##
Subject: Slice
Date: Tue, 09 Nov 93 10:03:52 TUR
From: Erdem ERTAN <E73412@vm.cc.metu.edu.tr>
Hello to all guys||
As I am new on this mailing list,I didnt know if this is written
before,but if it didnt I think this can be useful for everyone.
When I first learn about Imagine,I read the book 'Understanding
Imagine 2.0 '.In this book,It says that when error 2 is occured while
using slice command,try to move the objects a little bit and try again
and if it doesnt work try to find another way.Here I found another way:
When slice command doesnt work,first select the smallest objects that
will slice the large object.Then enter pick faces mode and select the
faces that intersects the other object,then use the fracture command.
Then try slice command again and now it works.
Maybe this info will be useless in Imagine 3.0 ,but it is still
useful for Imagine 2.0 users.
-----------------------------------------------------------------------
- ERDEM ERTAN e73412@vm.cc.metu.edu.tr -
-----------------------------------------------------------------------
##
Subject: Imagine pc-help
Date: Tue, 09 Nov 93 02:17:11
From: greg.tsadilas@hofbbs.com
Ok fellow PC (and Amiga) Imagineers, lets see if you can help me...
I created a group composed of 118 objects, 14,168 points, 35,697 edges and
21,852 faces. No textures, no brush maps, nothing, 'cept for many and many
sharp defined edges.
Problem: I cannot render the entire Image. Imagine apparently doesn't have
enough memory to do it (I have 8 Meg in my machine). Both Im2.0 and Im2.9 will
only render what it can (ie: about 1/3 of the total objects).
Does Imagine require more memory? From the detail editor in Im2.9, after I
load in the grouped objects, I have 4,013,080 Bytes Avail, and Largest Block is
4,012,944.
P.S. I boot a clean system
GreG tsadilas
greg.tsadilas@hofbbs.com
##
Subject: Re: Imagine pc-help
Date: Tue, 9 Nov 1993 16:22:21 +0100
From: <robin@robin.lausanne.sgi.com>
On Nov 9, 2:17am, greg.tsadilas@hofbbs.com wrote:
> Problem: I cannot render the entire Image. Imagine apparently doesn't have
> enough memory to do it (I have 8 Meg in my machine). Both Im2.0 and Im2.9
will
> only render what it can (ie: about 1/3 of the total objects).
>
Are you doing a ray-traced image ?
Be aware that all objects have to be in the limit of the world. So you could
change (or add) Global axis (in the action editor) to something bigger than
1024x 1024y 1024z (as it is by default), shrink down all you scene or set the
global axis to 0x 0y 0z, and so it will be figured to be big enough for your
scene.
Maybe, you have another problem, of course it is still possible...
Robin
--
\|/
@ @
---------------------------------------------------oOO-(_)-OOo-----------
Robin Chytil, Staff Engineer Email: robin@lausanne.sgi.com
Silicon Graphics Inc. Vmail: 5-9389
Mediterranean Distribution Territory Tel: +41 21 6249737
Lausanne, Switzerland Fax : +41 21 6259184
##
Subject: Re: Imagine pc-help
Date: Tue, 9 Nov 1993 13:06:18 +0100
From: Peter Bugla <bugla@informatik.tu-muenchen.de>
> I created a group composed of 118 objects, 14,168 points, 35,697 edges and
> 21,852 faces. No textures, no brush maps, nothing, 'cept for many and many
> sharp defined edges.
>
> Problem: I cannot render the entire Image.
[stuff deleted]
>
> GreG tsadilas
> greg.tsadilas@hofbbs.com
Hi Greg,
I'm working with I 2.0 - PC, and I think it's simply insufficient
memory in terms of these cute little black ICs you plug in
your rendering-engine :-).
Up to now I have only 4 Meg in my PC so this problem happens
more often. I can't even load some object into Detail e.g.
the amazing 1701-D-Enterprise from Carmen Rizzollo (don't know if
spelled correctly). When rendering things "at the edge of memory"
lots of single faces are left out.
If SIMMs were cheaper I would go for 20 Meg.
(Anybody out there giving away free SIMMS? ;-) )
Peter
--------------------------------------------------------------------------------
#include <disclaimer.h> " If architects built buildings the way
programmers write programs the first
woodpecker that came along would
Peter Bugla destroy civilization "
E-mail: bugla@informatik.tu-muenchen.de -- Murphy's Law of Computers
--------------------------------------------------------------------------------
##
Subject: Re: Comp Graphics
Date: Tue, 09 Nov 93 09:55:59 EDT
From: "Breno A. Silva" <INF02@BRUFSE.BITNET>
I've been working with computer graphics in my town for 3 years and I'm
doing quite well (at least that's what everybody tell me), and I used
Imagine for a TV commercial. In fact I'd love to have used it in all the
other 9999 animations I had to use Lightwave 'cause of its more practical
way of achieving the effects (more interactivity in the stage editor,
PLEASE! 3.0? :). Otherwise I HATE Lightwave's wireframe-only representa-
tion. The solid view in the modeler helps, but is restricted to a moving
(bouncing) representation, GKW (God-Knows-Why). I'll really love to get
back to Imagine, if 3.0 was developed enough for my demands. By the way,
I also prefer Caligari's interface. T's a shame so bad lightning and ab-
solute lack of morphing, reflections and procedural textures.
By the way, I'm considering to evaluate 3DS usefulness, for its
extremely fast rendering, mainly of reflections. I know they are "reflec-
tion maps" and not recursive, but for most of the use its quite enough,
and if they are not in Imagine 2.9, they should be in Imagine 3.0.
Breno A. Silva (INF02@BRUFSE.BITNET)
##
Subject: Upgrade and Giveaway
Date: Tue, 9 Nov 93 9:50:20 EST
From: woovis@jcnpc.cmhnet.org (William V. Swartz)
I too am a little disapointed that all the features I've been waiting for in
3.0 are not present in 2.9. But I am glad I haven't ordered it yet! The past
has taught me to wait until some folks get new versions in their hands and
comment on the functionality before taking my plunge. Ofcourse I'm not happy
about the third party support products for Imagine being broken in 3.0 (ie.
Essence) but if it is truly for progress than we should cope. Has anyone noted
if ISL will still work with 3.0 (or 2.9) staging files? And has Impulse
released the needed info to Worley and others? These, and billions of other
questions can be yours!
And speaking of questions, the Amiga Formatters are showing up in
amiga.graphics and starting to ask all sorts of elementary things that the
manual would have even answered. They even have the nerve to complain about
the program! They'll discover us soon enough so I'm going back into hiding
now.
Keep the faith...
//
\X/ -BiL-
woovis@jcnpc.cmhnet.org (See my 'Imagine'-ary signature below)
##
Subject: Re: Imagine pc-help
Date: Tue, 9 Nov 1993 07:14:53 -0800 (PST)
From: Doug Kelly <dakelly@class.org>
RAM is your problem. I've also got 8 megs, and can load LOTS more objects
in the Detail Editor than I can render. Carmen Rizzolo's NCC-1701D is
BEAUTIFUL, but I have to borrow a 16 Meg machine to render it.
Even when you work within your RAM limits, be aware that successive frames
in an animation will have progressively less RAM to work with. I don't
know if this is sloppy cleanup on Imagine's part, but if I'm working near
the limit, I have to reboot my system (or at least quit Imagine and
restart it) to make sure ALL my RAM is available. For animations, this
sometimes means that the last object loaded will disappear in the course
of the animation. The only work-around I've found is to render each frame
alone, rebooting for each rendering, then create the animation from the
finished frames. Kinda defeats the purpose of the Project Editor, doesn't it?
Any suggestions on fixing this would be greatly appreciated. I've had
this problem since Turbo Silver.
------------------------------------------------------------------------------
Doug Kelly Information Specialist First Consulting Group
dakelly@class.org (310)595-5291x125 P.O.Box 5161, Los Alamitos,CA 90721-5161
"The difference between genius and stupidity: genius has its limits."
------------------------------------------------------------------------------
##
Subject: Pc Version - 2.9
Date: Tue, 9 Nov 1993 09:18:23 -0800 (PST)
From: Mark Boulding <boulding@well.sf.ca.us>
Thanks for the responses on the Essence textures for Imagine-PC.
Does anyone know if Imagine 2.9 PC is shipping? Or is it just the Amiga
version?
Mark
--
+===================================================================+
| Mark Boulding c/o FB&T | Internet: boulding@well.sf.ca.us |
| 750 17th St., NW, Ste. 1100 | CIS: 72401,2400 Prodigy: TDRC42A |
| Washington, DC 20006 | AOL: boulding GEnie: M.BOULDING |
+===================================================================+
##
Subject: Re: Imagine pc-help
Date: Tue, 9 Nov 93 09:30:11 PST
From: 09-Nov-1993 1312 <leimberger@marbls.enet.dec.com>
Subj: Re: Imagine pc-help
Doug,
On the amiga (I have an 18meg 4000) you can have an arexx script
watch the directory, and as it sees an image completed it
converts the frame to DCTV, or Jpeg thus freeing up disk space.
I really don't know how much ram is needed to render Carmen's
object though. The dos boxs must have virtual memory support
from someone. This may be the way to go.
In all honesty I never paid much attention to Imagine's memory use.
I often have Imagine. Pixel3d pro, ADPro and depaint loaded and seldom
have memory problems. I'll have to load the NCC-1701D again to see
what it uses for ram. Would one expect a different usage between the
PC, and Amiga versions ? Is windows or anything like that loaded while
your runnung imagine ?
bill
##
Subject: Rend24
Date: Tue, 9 Nov 93 13:16:10 -0600
From: tes@killerbee.jsc.nasa.gov (Tom Smith)
I keep hearing about how great Rend24 is. But it's location seems to be the
Amiga community's best kept secret. If anyone has it could they upload it
to an wuarchive or something? Or let me know where I can find a copy. Thanks.
Tom Smith
##
Subject: Re: Imagine pc-help
Date: Tue, 9 Nov 1993 13:06:11 -0600 (CST)
From: Schumacher Gordon C <gschumac@ux4.cso.uiuc.edu>
> I often have Imagine. Pixel3d pro, ADPro and depaint loaded and seldom
> have memory problems. I'll have to load the NCC-1701D again to see
> what it uses for ram. Would one expect a different usage between the
> PC, and Amiga versions ? Is windows or anything like that loaded while
> your runnung imagine ?
Yes! The IBM and Amiga have very different memory usage comarisons.
The minimum configuration on an IBM to do *anything*, much less graphics
is 4MB. I have 3MB on my A3000 and am working just fine, doing image
processing, Imagine, and the whole bit. I would guess this is about
equivalent to an 8MB IBM.
There's just something about the way Windoze and its apps handle memory...
--
Gordon Schumacher
/-------------------------------------------------------------------\
| Champaign- "We apologize for the inconvenience." _@_ |
| Urbana -HHGTTG / \ |
| kilroy was here | o o | |
\-------------------------------------------------------U|--U--|U---/
##
Subject: Imagine 2.9
Date: Tue, 9 Nov 93 15:25:04 PST
From: dgb@cup.portal.com
OK, so I hear that Imagine 2.9 doesn't have all of the features of 3.0...
Does 2.9 have a working SLICE command? Not having a working boolean
cutting operator is my worst hangup with 2.0......
##
Subject: Imagine pc-help
Date: Tue, 9 Nov 1993 17:03:00 -0500
From: roy.park@canrem.com (Roy Park)
>Problem: I cannot render the entire Image. Imagine apparently doesn't have
>enough memory to do it (I have 8 Meg in my machine). Both Im2.0 and Im2.9 will
>only render what it can (ie: about 1/3 of the total objects).
Solution: Get more memory.
It's something that you just NEED... can't help you much there... UNLESS you
want to use VERY SLOW virtual memory.
Currently, I have 14 MB RAM on my system and sometime I run out of memory doing
some raytracing with Imagine. I'll probably get some more when the RAM price
drops in the future.
>GreG tsadilas
>greg.tsadilas@hofbbs.com
---------------------------------------------------------------
|Roy Park | C= // A3000@25Mhz-14.4HST-2MBPicassoII|
|roy.park@canrem.com | // DM3024-ST296N-LP240S-LP240S-LP105S|
|rkpark@io.org | \X/ ViewSonic17-2MBChip/12MBFast-Emplant|
---------------------------------------------------------------
##
Subject: Re: Imagine pc-help
Date: Tue, 9 Nov 1993 18:38:12 -0600 (CST)
From: Schumacher Gordon C <gschumac@ux4.cso.uiuc.edu>
> Windows, what does THAT have anything to do with Imagine! This is not a
> "Windoze" problem - it's the memory manager Impulse uses on the PC, MSDOS
> and whatever other TSR's are running. I don't think you can even run
> Imagine when Windows is running, or in a DOS box under Windows. I've heard
> problems when trying to do so.
You mean Imagine PC *isn't* a Windows app?! My apologies. I thought
that nobody *wrote* things that ran under DOS anymore...
> The issue with memory on the PC you are referring to is not Windows' fault,
> but a limitation of MSDOS and backward compatibility, i.e. 8086 support,
> 640K conventional memory, etc. With NT, this problem is gone since the
> full power of the protected mode in '386, '486 chips can be exploited and
> take advantage of the linear memory address space.
Yes, this is true, but it's also that IBM software programmers never
seemed to realize that writing code to conserve memory is a good idea...
at least not after you could access anything above 640K at all...
And yes, I definitely think NT is a big step in the right direction!
*I AM NOT AN AMIGA BIGOT* ;)
There are things that I like/dislike about all three (Amy, Mac, IBM)
machines.
--
Gordon Schumacher
/-------------------------------------------------------------------\
| Champaign- "We apologize for the inconvenience." _@_ |
| Urbana -HHGTTG / \ |
| kilroy was here | o o | |
\-------------------------------------------------------U|--U--|U---/
##
Subject: Re: Imagine pc-help
Date: Tue, 9 Nov 93 17:55:09 MST
From: pringleg@cuugnet.cuug.ab.ca (Greg Pringle)
> > I often have Imagine. Pixel3d pro, ADPro and depaint loaded and seldom
> > have memory problems. I'll have to load the NCC-1701D again to see
>
> Yes! The IBM and Amiga have very different memory usage comarisons.
> The minimum configuration on an IBM to do *anything*, much less graphics
> is 4MB. I have 3MB on my A3000 and am working just fine, doing image
> processing, Imagine, and the whole bit. I would guess this is about
> equivalent to an 8MB IBM.
I have a 6 meg amiga, and have tried imaginepc on a 8 meg IBM, and
I'd have to disagree.. 3mb on an amiga is nowhere close to loading
in the big enterprise object BTW. I don't think I can render it with 6
megs unless I use virtual memory. From what I have seen, it looks like
imagine PC is a pretty straight port, and I can't see them using
different data structures for the objects.
> There's just something about the way Windoze and its apps handle memory...
Imagine won't run under windows.. if it did, you could use window's built
in virtual memory and not have the problem in the first place. Of course,
this also means you have to quit imagine to load up a paint program, then
quit the paint program to load imagine.. does anyone know if imagine pc
will run with OS2, WindowsNT? Would this allow virtual memory?
Greg
(of course, then I'd have to get more memory to run OS2.. B^)
##
Subject: re:Imagine pc-help
Date: Tue, 9 Nov 1993 20:14:56 -0500 (EST)
From: kingb@echonyc.com (Andrew McDonald)
> The minimum configuration on an IBM to do *anything*, much less graphics
> is 4MB. I have 3MB on my A3000 and am working just fine, doing image
> processing, Imagine, and the whole bit. I would guess this is about
> equivalent to an 8MB IBM.
>
> There's just something about the way Windoze and its apps handle memory...
> --
> Gordon Schumacher
Imagine *does* not run if Windows is loaded, so you can forget that
theory. I always ran out of memory when my 3000 had only 4 megs, 16 seems
to be the minimum for complex 3d scenes.
Still, even with the occasional crash, Imagine-PC renders so fast, I don't
care. Haven't seen any of the mentioned problems with decreasing ram
during multi-frame renderings. Just finished my 10th fifteen second anim
without a hitch.
If only Imagine would co-exist with a Novell network.....
Andrew - obligatory boring signature, feel free to add useless blanks.
##
Subject: Re: Imagine pc-help
Date: Tue, 9 Nov 1993 19:48:53 -0600 (CST)
From: Schumacher Gordon C <gschumac@ux4.cso.uiuc.edu>
> Nope, I'm running bare-bones DOS, nothing in the background, just Imagine.
> I have the same problem on both the PC AND the Amiga.
>
> Imagine is just a RAM-hog, IMHO.
*Any* 3-D package hogs memory... just like any image processor...
they're just memory-intensivce sorts of things.
--
Gordon Schumacher
/-------------------------------------------------------------------\
| Champaign- "We apologize for the inconvenience." _@_ |
| Urbana -HHGTTG / \ |
| kilroy was here | o o | |
\-------------------------------------------------------U|--U--|U---/
##
Subject: Re: Imagine pc-help
Date: Tue, 9 Nov 1993 20:33:17 -0600 (CST)
From: Daniel Jr Murrell <djm2@Ra.MsState.Edu>
> Still, even with the occasional crash, Imagine-PC renders so fast, I don't
> care. Haven't seen any of the mentioned problems with decreasing ram
> during multi-frame renderings. Just finished my 10th fifteen second anim
> without a hitch.
>
Dunno about the PC version, but Amiga Imagine has a setting for
controlling the amount of RAM used during multi-frame rendering. It
stores however many frames you specify so that it can calculate the
deltas for anims properly. It speeds up ANIM rendering, but if you
don't have enough RAM, then set the number of screens low, like 2 or
1. I believe the setting is NUMS.
I'm not exactly sure if this is the problem with whoever first asked,
but it's what I would check first.
Dan
djm2@ra.msstate.edu
##
Subject: Imagine pc-help
Date: Tue, 09 Nov 93 22:40:00
From: greg.tsadilas@hofbbs.com
Thanks to all that confirmed that I lacked sufficient memory to render my
181 object (corrected number) file.
Something I also learned is that Scanline mode takes up much more memory
than ray TRACE mode. This temporarily solves my problems. I also learned
from Mike Halvorson that each point in an object occupies about 250 bytes
worth of descriptive data structures.
Thought I'd share the following message with you, I'll withhold the
senders address.
>>Ok fellow PC (and Amiga) Imagineers,
>Just making this distinction powers up my MEGA FLAME mode.We are all Imagine
>users. I'm not in the greatest of moods this morning but will hold back
>the mega flame this time. Next time you'd better be wrapped in asbestos.
Ahem, I know we are all "Imagine users" but we are not all using the same
machine. Imagine is new to the PC, so if another PC user has/had the same
problem and has rectified it _they_ would be better qualified to answer my
question.
I've used Silver back when it was in alpha version with only primitive
sphere, plane & triangle, through to Turbo Silver (yes I own 2 Amiga's)
and never had this happen with far less memory than I have now, so it
seemed odd to me.
So if the opening to my question set you off, flame away next time. It
seems you look for anything to set you ablaze.
>Anyway, YES, you need more memory. Imagine won't tell you that, it
>will just render what it can.
Thank you. This should have been the only content of your reply.
>But before you run out to buy more memory perhaps you can tell me why you
>have so many objects with so many points? You could probably do what you
>want with the memory you have just by going about it a different way.
Because when I create an object I create it highly detailed. There is
no other way to create it, brushmaps won't do it, textures won't do it,
what else is there? Any non-flamatory suggestions would be welcomed.
To anyone who is wondering what the object in question is/are, it's
called the "AFS-mk I", which stands for "ARMOURED FIGHTING SUIT-mark 1".
Basically it's a motorized exoskeleton that one climbs into for protection
during combat. This is a whim of mine, nothing I'm working on for anyone
but myself.
Again,thanks to all the replies.
GreG tsadilas
greg.tsadilas@hofbbs.com
##
Subject: Twist and Shout...
Date: Tue, 9 Nov 1993 23:34:18 -0600 (CST)
From: Cliff Lee <cel@tenet.edu>
I've finally had some time to play with Imagine 2.9. Some nice stuff.
The real time graphics in the perspective window is nice.
I love the deformations. There is twist, bend, pinch, taper, stretch, and
shear. Thats the good news. The bad news is that they don't all work.
At least for me. Maybe I'm screwing something up?!? OK here the deal...
Create a Cylinder. Transform the axis to scale the z axis to the full
length of the cylinder. Select it, then try the bend command.
Interactively the bounding box is transformed nicely. Things "look"
encouraging. Click on OK. Object snaps back to its original shape. :(
I have got the twist, shear, stretch, pinch, and taper to work, but no go
on the bend.
One of the biggest dissapointments with Imagine 2.0 PC was the total lack
of textures. I believe that 8 were shipped with it orignially. And no
Essence for the PC. Imagine 2.9 has MANY textures. A nice refreshing
change for PC users. I'm still investigating. Trying multiple in
combination. DINOSKIN, WORMVIENS, STAINGLAS are neato!
Looks like Particles are included, but I've not tried it yet. Says in the
document that comes with it: "Particles F/X only works on objects for the
moment. In the next version of IMagine you will be able to do this to any
object in the group".
If I'm babbling on about things that everyone knows, let me know...
-=-=-=-=-=-=-=-=-=-=-=-=-
Cliff Lee
cel@tenet.edu
"Everything will work out if you let it!" Cheap Trick
##
Subject: PAR question
Date: Tue, 9 Nov 93 23:15:18 CST
From: drrogers@camelot.b24a.ingr.com (Dale R Rogers)
I know that one of the outputs of the DPS PAR is NTSC composite.
Does this mean I can output a signal to my video monitor on my
television? I would also like to record test animations to my
VHS recorder. Is this possible?
Dale
____________________________^____________________________
dale r. rogers
Intergraph Corporation
Building Design & Management MailStop: LR24A4
drrogers@b24a.b24a.ingr.com Tel: (205) 730-8294
.
##
Subject: Re: Twist and Shout...
Date: Wed, 10 Nov 1993 00:26:48 -0600 (CST)
From: Daniel Jr Murrell <djm2@ra.msstate.edu>
> I love the deformations. There is twist, bend, pinch, taper, stretch, and
> shear. Thats the good news. The bad news is that they don't all work.
> At least for me. Maybe I'm screwing something up?!? OK here the deal...
> Create a Cylinder. Transform the axis to scale the z axis to the full
> length of the cylinder. Select it, then try the bend command.
> Interactively the bounding box is transformed nicely. Things "look"
> encouraging. Click on OK. Object snaps back to its original shape. :(
>
> I have got the twist, shear, stretch, pinch, and taper to work, but no go
> on the bend.
Oooooh! This sounds really good. I can't wait til I get my Amiga
version. Hmm. Think about what you're trying to do. Can you really
bend a cylinder, when there's no points in the middle of the faces?
Maybe it should work, but it sounds like you're using the wrong kind
of object. Try this: make a vertical line, with many points in the
line. Sweep it around the axis to make a tube. THEN try the bend
transform. (Assuming it wasn't really disabled) It should then bend
properly, since it actually has points to bend around.
>
> One of the biggest dissapointments with Imagine 2.0 PC was the total lack
> of textures. I believe that 8 were shipped with it orignially. And no
> Essence for the PC. Imagine 2.9 has MANY textures. A nice refreshing
> change for PC users. I'm still investigating. Trying multiple in
> combination. DINOSKIN, WORMVIENS, STAINGLAS are neato!
>
> Looks like Particles are included, but I've not tried it yet. Says in the
> document that comes with it: "Particles F/X only works on objects for the
> moment. In the next version of IMagine you will be able to do this to any
> object in the group".
Wow. Imagine is starting to sound more like Wavefront every day. :)
Ok, more KINDA like Wavefront. (I do use Wavefront at work, so I know
what I'm talking about :)
> If I'm babbling on about things that everyone knows, let me know...
How would we know without folks like you giving us reports? :)
Danimal
djm2@ra.msstate.edu
##
Subject: Re: Twist and Shout...
Date: Wed, 10 Nov 1993 00:57:05 -0600 (CST)
From: Cliff Lee <cel@tenet.edu>
On Wed, 10 Nov 1993, Daniel Jr Murrell wrote:
> > I have got the twist, shear, stretch, pinch, and taper to work, but no go
> > on the bend.
> version. Hmm. Think about what you're trying to do. Can you really
> bend a cylinder, when there's no points in the middle of the faces?
OK, so I tried it with a plane instead. This time I got it work. I am by
no means, an expert at this. Major learning curve parity bit ON.
Thanks for the advice.
-=-=-=-=-=-=-=-=-=-=-=-=-
Cliff Lee
cel@tenet.edu
"Everything will work out if you let it!" Cheap Trick
##
Subject: Re: Imagine pc-help
Date: Wed, 10 Nov 1993 18:25:41 +1100 (EST)
From: Nikola Vukovljak <nvukovlj@ucc.su.OZ.AU>
On Tue, 9 Nov 1993 greg.tsadilas@hofbbs.com wrote:
>
> Ok fellow PC (and Amiga) Imagineers, lets see if you can help me...
>
> I created a group composed of 118 objects, 14,168 points, 35,697 edges and
> 21,852 faces. No textures, no brush maps, nothing, 'cept for many and many
> sharp defined edges.
>
> Problem: I cannot render the entire Image. Imagine apparently doesn't have
> enough memory to do it (I have 8 Meg in my machine). Both Im2.0 and Im2.9 will
> only render what it can (ie: about 1/3 of the total objects).
>
> Does Imagine require more memory? From the detail editor in Im2.9, after I
> load in the grouped objects, I have 4,013,080 Bytes Avail, and Largest Block is
> 4,012,944.
>
> P.S. I boot a clean system
I have found that you would need about 12 Meg to render a scene with
about 20000 faces. This is on the Amiga. The PC version seems to be a bit
more memory hungry (needs minimum 4M to run) so you may find that you
need even more than 12M.
Nik.
##
Subject: Twist and Bend w/ IM2.9
Date: Wed, 10 Nov 1993 10:39:42 -0600 (CST)
From: Cliff Lee <cel@tenet.edu>
OK, so I've got the Twist to work, and the bend to work. But not
together. I took a disk made from 4 points (a square really) and removed
the middle point. Extruded it to a lenght of 100 with 20 section. Nice
square tube. Now I transform my Axis to force the Z axis to extend along
the axis of the tube (and extend to at least the end of the object). Use
the twist command. It works. Nice. Undo that. Now use the Bend
Command. Nice. Now try to Twist it after you Bent it. Nada. What I was
trying to do was to creat a twisting arch.
Interesting thing about bend is that the endpoint of the object don't
move. From a strucutral engineering standpoint you would say that the
ends were fixed with respect to rotation and translation. It would be
nice to allow the end point to rotate so that the base of the object is
perpindicular to the central axis. But hey, they just get his out and I'm
whining for more... such is life.
-=-=-=-=-=-=-=-=-=-=-=-=-
Cliff Lee
cel@tenet.edu
"Everything will work out if you let it!" Cheap Trick
##
Subject: Imagine Animation Speeds
Date: Wed, 10 Nov 1993 11:10:53 -0500
From: "Rob (R.D.) Hounsell" <hounsell@bnr.ca>
Folks,
What is the actual playback rate of the animation preview in the perspective
view of the stage editor? If the speed control slider is set all the way to the
right (fastest), is it meant to represent 30fps? (assuming ideal conditions
i.e. no additional CPU load, etc.). I'd like to have some accurate method of
determining the timing of a segment in the stage editor preview without actually
rendering the thing in the project editor and then playing it (trial and error
method).
I'll ask the same question about playing Imagine-format animations from the
project editor, if the animation was generated without a movie file. What is
the default rate? Does the playback software in Imagine read the movie file? If
so, why don't changes in the movie file come into effect until the next time
the animation is generated? Are they somehow incorporated into the specs file,
or in the anim.xxxx files themselves?
Thx
Rob
--
+-----------------------------------------------------------------------------+
| Rob Hounsell BNR WAN: HOUNSELL@NMERH53 |
| Team Leader: UNIX INTERNET: HOUNSELL@BNR.CA |
| System Performance: PHONE: (613) 765-2904 |
| Paradigm Club Design Team. Dept. PS27 ESN: 395-2904 |
| Northern Telecom Public Switching |
+-----------------------------------------------------------------------------+
##
Subject: PAR connectors
Date: Wed, 10 Nov 93 11:14:19 CST
From: drrogers@camelot.b24a.ingr.com (Dale R Rogers)
Thanks for the responses I've gotten so far.
Since now I know that it is possible to send the NTSC signal to my
TV and/or VHS VCR (I thought that was the case), I guess my
problem is concerning the cable from the PAR board to my VCR/TV.
The output of the PAR is a different connector than the input to
my VCR/TV. (This is where my ignorance shows.) I beleive the PAR
has a BNC type connector (is that the name?) and the VCR/TV
connectors have the RCA type connectors. Is there a common patch
cable that will go between these connections? That's what
prompted my previous question. I wasn't sure if a the connector
type on the PAR was indicating a signal type that my VHS machine
could not handle. If there is no signal problem, then I just need
the proper cable. Any one have input on this? Is this a cable
that I can pick up at a Radio Shack? Or would I be better off
going to a video supply house and getting a sheilded cable?
Thanks in advance. I have an Imagine animation I dieing to get to
tape.
Dale
____________________________^____________________________
dale r. rogers
Intergraph Corporation
Building Design & Management MailStop: LR24A4
drrogers@b24a.b24a.ingr.com Tel: (205) 730-8294
.
##
Subject: Re: PAR question
Date: Wed, 10 Nov 93 09:44:49 PST
From: CarmenR@cup.portal.com
Dale R. Rogers writes...
> I know that one of the outputs of the DPS PAR is NTSC composite.
> Does this mean I can output a signal to my video monitor on my
> television? I would also like to record test animations to my
> VHS recorder. Is this possible?
Yup.. That's pretty much how it works. :) The output in composite, but
it's using a superior BNC type connector. You can go to Radio Shack and buy
a 6' BNC to BNC cable, and a BNC -> to RCA adaptor to use at the end of the
cable so it'll fit into a consumer VHS deck. Then you can have the Output
of your VCR feed your television. Neato how tht works, eh? :)
Carmen Rizzolo - Crazed Artist
CarmenR@cup.portal.com
##
Subject: Re: PAR question
Date: Wed, 10 Nov 93 12:49:51 -0500
From: mbc@po.CWRU.Edu (Michael B. Comet)
>
>
>
> I know that one of the outputs of the DPS PAR is NTSC composite.
> Does this mean I can output a signal to my video monitor on my
> television? I would also like to record test animations to my
> VHS recorder. Is this possible?
>
> Dale
>
Yes. It has an NTSC composite BNC connector, a SVHS S-Video
output, and 3 BNC's for MII or Betacam.
This is what it is made for. I use it to play my 24 bit anims in
real time.
--
+------------------------------------------------------------------------+
| Michael Comet, mbc@po.CWRU.Edu, CWRU Software Engineer/Graphics Artist |
+------------------------------------------------------------------------+
##
Subject: Re: imagine pc-help
Date: Wed, 10 Nov 93 09:41:00 PST
From: Stethem Ted 5721 <TedS@ms70.nuwes.sea06.navy.mil>
To GreG tsadilas greg.tsadilas@hofbbs.com
> To anyone who is wondering what the object in question is/are, it's
>called the "AFS-mk I", which stands for "ARMOURED FIGHTING SUIT-mark 1".
>Basically it's a motorized exoskeleton that one climbs into for protection
>during combat. This is a whim of mine, nothing I'm working on for anyone
>but myself.
Have you seen that new cartoon show on Saturday called "Exo-Squad" or
something like that? It is about a squadron of future soldiers with
Exoskeleton suits. Pretty detailed for a cartoon show. They are in a
struggle with the Neohumans from Mars who have their own Exoskeleton suits,
different designs.
BTW, have you had any luck with any Extended Memory Managers and Imagine
PC? I have found that Imagine PC doesn't seem to "like" any EMM's at all
but I think this is the reason for lack of memory problem. It appears to me
that Imagine PC will use mostly DOS memory, and needs almost the whole 640K
to run effectively. On the other hand, it seems to be able to use expanded
memory during rendering. I haven't tried QEMM yet, maybe that will work.
Imagine PC sure doesn't like EMM386 in DOS and it doesn't like XMS for
Windows. I haven't been able to get Imagine PC to run at all when the
Windows config is used for booting up, even if I don't call Windows. I
think it is possible that Imagine PC would run more effectively if there
were some EMM that it is compatible with it but I haven't found it yet. If
you have had any better luck, please let me know.
##
Subject: Re: Imagine Animation Speeds
Date: Wed, 10 Nov 93 14:09:38 -0500
From: mbc@po.CWRU.Edu (Michael B. Comet)
>
>Folks,
>
> What is the actual playback rate of the animation preview in the perspective
>view of the stage editor? If the speed control slider is set all the way to the
>right (fastest), is it meant to represent 30fps? (assuming ideal conditions
>i.e. no additional CPU load, etc.). I'd like to have some accurate method of
>determining the timing of a segment in the stage editor preview without actually
>rendering the thing in the project editor and then playing it (trial and error
>method).
>
I don't think they seem to be correct for a fps. I use the second
to fastest as an approximation on my A3000....it also seems that that plays
closer to true 30fps when I do a Play Big (only available on the Amiga
side).
> I'll ask the same question about playing Imagine-format animations from the
>project editor, if the animation was generated without a movie file. What is
>the default rate? Does the playback software in Imagine read the movie file? If
>so, why don't changes in the movie file come into effect until the next time
>the animation is generated? Are they somehow incorporated into the specs file,
>or in the anim.xxxx files themselves?
>
I belive it gets incorporated. BTW, you can play with the speeds
whileyou are playing back. Space will move 1 frame at a time, and F1-F8 I
think will adjust the speeds...F1 being fastest...
--
+------------------------------------------------------------------------+
| Michael Comet, mbc@po.CWRU.Edu, CWRU Software Engineer/Graphics Artist |
+------------------------------------------------------------------------+
##
Subject: Thanks for the PAR info
Date: Wed, 10 Nov 93 13:41:57 CST
From: drrogers@camelot.b24a.ingr.com (Dale R Rogers)
It will take less time to respond to the list, as opposed to all
the persons who responded to my questions.
Thanks for the PAR connection info. It was very timely and
helpful information.
BTW... the list consistently beats the manufacturers on giving me
support. DPS hasn't got back to me yet. Thanks for the help.
Dale
____________________________^____________________________
dale r. rogers
Intergraph Corporation
Building Design & Management MailStop: LR24A4
drrogers@b24a.b24a.ingr.com Tel: (205) 730-8294
.
##
Subject: Re: PAR connectors
Date: Wed, 10 Nov 1993 00:51:00 -0500
From: j#d#.moore@canrem.com (J. Moore)
DRR> The output of the PAR is a different connector than the input to
DRR> my VCR/TV. (This is where my ignorance shows.) I beleive the PAR
DRR> has a BNC type connector (is that the name?) and the VCR/TV
DRR> connectors have the RCA type connectors. Is there a common patch
DRR> cable that will go between these connections? That's what
DRR> prompted my previous question. I wasn't sure if a the connector
DRR> type on the PAR was indicating a signal type that my VHS machine
DRR> could not handle. If there is no signal problem, then I just need
DRR> the proper cable. Any one have input on this? Is this a cable
DRR> that I can pick up at a Radio Shack? Or would I be better off
DRR> going to a video supply house and getting a sheilded cable?
Just an adaptor will do. Check with a video supply house if you have one
nearby. I get my cables done at one here in Toronto and not only are
their connectors better quality than Radio Shack, they're 2/3s the price.
I use BNC ends on both end and an RCA adaptor (so they can do double
duty) but you could have a cable made up with different connectors on
each end.
* Q-Blue 0.93 [NR] *
##
Subject: Re: Imagine pc-help
Date: Wed, 10 Nov 1993 16:58:21 -0500 (EST)
From: Udo K Schuermann <walrus@wam.umd.edu>
Nikola Vukovljak <nvukovlj@ucc.su.oz.au> writes:
> I have found that you would need about 12 Meg to render a scene with
> about 20000 faces. This is on the Amiga.
This is by no means universally applicable, as CastleRoom consisted of more
than 20000 faces and rendered in less than 6MB of RAM. That was with a
fair bit of refraction and reflection, including three light sources and
two bitmap images. There are many more factors involved than just #faces,
although I'd be hard pressed to enumerate them all.
._. Udo Schuermann
( ) walrus@wam.umd.edu
##
Subject: Re: Upgrade and Giveaway
Date: Wed, 10 Nov 93 11:59:04 PDT
From: Jeff.Saffold@lookout.com (Jeff Saffold)
> And speaking of questions, the Amiga Formatters are showing up in
> amiga.graphics and starting to ask all sorts of elementary things that the
> manual would have even answered. They even have the nerve to complain about
> the program! They'll discover us soon enough so I'm going back into hiding
> now.
Well, I just talked to Impulse, and they said that the AF Upgrade
offer was ONLY for the European people, not for the American/NTSC
users... So anyone wanting to go that route is SOL...:)
// Jeff Saffold
\X/ Only the Amiga makes it possible.
... We give advice, but we cannot give the wisdom to profit by it.
___
X MsgView V1.13 [R029] X
--
*******************************************************************************
* Cuerna Verde BBS FidoNet Gateway Data/Fax: 1-719-545-8572 *
* Pueblo, Colorado USA FidoNet: 1:307/18 *
*******************************************************************************
##
Subject: Bye bye,
Date: Thu, 11 Nov 93 10:57:12 -0500
From: ac394@leo.nmc.edu (Adam Benjamin)
Well, I'm droping my name off the list as I get ready for the big adventure
(life that is..)
I have accepted a job with IBM's customer support, and will be
moving about 850 miles.
Keep on Keeping on, my friends
Adam Benjamin
--
---
Adam Benjamin ac394@leo.nmc.edu A.K.A. A.Benjamin@mi04.zds.com
Christian Animator Computer junkie 4000/40 CD-ROM 486DX33
Get a life! (John 3-16)
##
Subject: question
Date: Thu, 11 Nov 93 11:14:34 CST
From: setzer@comm.mot.com (Thomas Setzer)
Any ideas on how to model a candle flame? Animated ofcourse.
Tom Setzer
setzer@ssd.comm.mot.com
"And of course, I'm a genius, so people are naturally drawn to my fiery
intellect. Their admiration overwhelms their envy!" - Calvin
##
Subject: More on the arrow
Date: Thu, 11 Nov 93 11:27:46 CST
From: setzer@comm.mot.com (Thomas Setzer)
I played around with the arrow a little. Came up with something a little
better(still needs work, however)
Action editor:
12345678901234567890
ARROWBODY&PATH position __________----------
fx1 ==========
fx2 ==========
PATH actor --------------------
position 1-10 is just tween and you don't move it
position 11-20 is follow path, where path is the same path that the arrow
grows along. You have to have saved it as a different object, say PATH,
and have it lined up in the stage editor so that the begining of the path is
touching the axis of the arrowbody&path group.
fx1 is grow, have it grow along path(see earlier post, read Understanding
Imagine 2.0, etc)
fx2 is grow along path but with time reversed. Maybe you dont need this
effect, but it was what I used when I was playing with it.
This gives you the effect of the arrow squirming along your path, across the
screen, and the squirming off the screen. Probably looks best if the head
of the arrow is offscreen when you start frame 11.
Again this is just an experiment and you may want to play with this yourself.
It still needs work, but it looked nice.
Tom Setzer
setzer@ssd.comm.mot.com
"And of course, I'm a genius, so people are naturally drawn to my fiery
intellect. Their admiration overwhelms their envy!" - Calvin
##
Subject: Re: candle flame
Date: Thu, 11 Nov 1993 10:22:22 -0800 (PST)
From: Doug Kelly <dakelly@class.org>
I used a morphing fog object, but that's only useful if you'll never have
one flame pass in front of another. Also, I'm crippled by not having
Essence; I've heard of some neat techniques for flames using that approach.
When you're making the fog object, remember that it's not a simple
teardrop; the interior, where actual combustion initiates, is a small oval
that is completely transparent. I sliced a distorted sphere out of the
inside of a teardrop, and that has worked fairly well.
Make your wick several colors, white shading up to black to red to bright
yellow-orange at the tip, and make it BRIGHT so it's like a glowing coal.
I made mine a sort of question-mark shape.
Needless to say, I went half-blind staring at candle flames to model this...
------------------------------------------------------------------------------
Doug Kelly Information Specialist First Consulting Group
dakelly@class.org (310)595-5291x125 P.O.Box 5161, Los Alamitos,CA 90721-5161
"The difference between genius and stupidity: genius has its limits."
------------------------------------------------------------------------------
##
Subject: question
Date: Thu, 11 Nov 1993 15:05:00 -0500
From: roy.park@canrem.com (Roy Park)
>Any ideas on how to model a candle flame? Animated ofcourse.
Try using 'gas' object... use multiple/different coloured gas objects will
do it very nicely (blue for the inside, yellow for the middle, organge for
the outer side of the flame).
Similar trick was used to create my "JetThrust.lha" (which I haven't uploaded
to FTP yet due to a technical difficulty) :)
>Tom Setzer
>setzer@ssd.comm.mot.com
+--------------------------------------------------------------+
|Roy Park | C= A3000@25Mhz-14.4HST-2MBPicassoII|
|roy.park@canrem.com | // DM3024-ST296N-LP240S-LP240S-LP105S|
|rkpark@io.org | \X/ ViewSonic17-2MBChip/12MBFast-Emplant|
+--------------------------------------------------------------+
##
Subject: CD32 and Imagine
Date: Thu, 11 Nov 1993 11:52:15 -0700
From: "Jeff Wahaus, CAPS, ATL, 404-640-3529" <JEFF_W1@sfov1.verifone.com>
Wouldn't it be neat if you could hook up a CD32 to your amiga and
use it as some additional horsepower during rendering. I really
hope someone comes out with some kind of SCSI interface for the
CD32 and writes the software to network it to any Amiga. Not
only would you have a CD-ROM drive, but also a display buffer,
an additional CPU, and optional hardware MPEG support.
Just think, you could send your MPEG animation to the CD32 via
SCSI and use its hardware MPEG decoding to play animations at
a full 30 fps in 24-bit at full screen resolution!
Or you could use the CD32 to process images as they complete rendering
and them send them back. The image processing part would not slow down
(very much) the additional frames rendering.
I wish I had to time to write something like this myself. It would
be a very interesting project. I would have the CD32 look sort of like
a disk drive to the host computer. Packets sent to the CD32 would be
similar to writing a small file to a SCSI drive. The software would
periodically read a small file from the CD32 (maby 5 times a second)
so that the CD32 could send packets back to the host computer.
All types of input could be sent to the CD32 via packets over the SCSI
bus: keyboard, mouse, disk drives, requests to play animations, requests
to get files from the CD-ROM drive, requests to run programs.
I've been wanting to get a CD-ROM drive to add to my accumulation of
Amiga computing machinery. How much better it would be if I could
add a CD32 for the same price and get all of these additional features.
I think Commodore could sell a lot more of these CD32 things if they
provide a good interface via SCSI to the rest of the Amiga family.
-Jeff Wahaus-
##
Subject: Re: candle flame
Date: Thu, 11 Nov 93 14:03:24 PST
From: Byrt Martinez <martinez@srcsvr.cup.hp.com>
content-type:text/plain;charset=us-ascii
mime-version:1.0
>I used a morphing fog object, but that's only useful if you'll never have
>one flame pass in front of another. Also, I'm crippled by not having
I'm starting to wonder about this cancelling out effect of Imagine. I had a
similar type of thing happen when modelling a phaser beam, but I got the
effect I wanted, though I still don't understand why. Apparantly, when the
cancelling phenomenon occurs, it looks as if it is doing some XORing. I THINK
that if you put say a third flame in front of the second one passing in front
of the first one, the flame comes back when all three are in line with each
other.
Unfortunately, I'm at work and can't test this theory right now. But I do
have this cool pic of my phaser beam which shows this weird "feature."
Although, the phaser beam doesn't use fog, it looks as though it should. :-/
If anyone wants, I can ftp the pic somewhere, maybe Aminet. Let me know.
--
@******************************************************************************@
* "When you point your finger 'cause your *
* Byrt Martinez plan fell through, you've got 3 more *
* - Non-Christian 8^P fingers pointing back at you." *
* (Fred 12:45) - Dire Straits *
* *
* martinez@srcsvr.cup.hp.com, byrt@shell.portal.com *
@******************************************************************************@
##
Subject: scallop shell
Date: Fri, 12 Nov 93 00:58:26 +0100
From: humez@idnges.decnet.citilille.fr
I'd like to model a scallop shell. Any idea on how to create
easily the stripes ?
Stanis humez@idnges.decnet.citilille.fr
##
Subject: CD32 and Imagine
Date: Thu, 11 Nov 1993 18:26:00 -0500
From: roy.park@canrem.com (Roy Park)
> Wouldn't it be neat if you could hook up a CD32 to your amiga and
>use it as some additional horsepower during rendering. I really
>hope someone comes out with some kind of SCSI interface for the
>CD32 and writes the software to network it to any Amiga. Not
>only would you have a CD-ROM drive, but also a display buffer,
>an additional CPU, and optional hardware MPEG support.
A SCSI interface won't let you control CD32 remotely... you'd be lucky
if you just get the CD-ROM drive in CD32 working on other Amiga. Also,
as for using CD32 as a 'rendering' engine, 68020 isn't too fast and
CD32 doesn't have math coprocessor.
If you want to get CD-ROM capability that bad, just buy a CD-ROM drive
or just get CD32. Why all the trouble? Nobody's going to bother
buying CD32 FOR THEIR AMIGA. Weird thought...
> Just think, you could send your MPEG animation to the CD32 via
>SCSI and use its hardware MPEG decoding to play animations at
>a full 30 fps in 24-bit at full screen resolution!
The MPEG is another issue... Commodore will have FMV module for every
(ok, some) Amiga in the future soon or later, so this isn't necessary.
> I've been wanting to get a CD-ROM drive to add to my accumulation of
>Amiga computing machinery. How much better it would be if I could
>add a CD32 for the same price and get all of these additional features.
And you can't have it for the same price... much higher.
> I think Commodore could sell a lot more of these CD32 things if they
>provide a good interface via SCSI to the rest of the Amiga family.
Dream on! :)
##
Subject: Re: question
Date: Thu, 11 Nov 1993 18:26:00 -0500
From: roy.park@canrem.com (Roy Park)
> Gas object? Is this in 2.0? I'll have to back and look at my
> manual. ~:o
Yes, it's in Attributes requester. Read it carefully, and you will see it!
:)
##
Subject: CD32 and Imagine
Date: Thu, 11 Nov 93 16:20:13 PST
From: CarmenR@cup.portal.com
With a sparkle in his eyes, Jeff Wahaus writes...
> Wouldn't it be neat if you could hook up a CD32 to your amiga and
> use it as some additional horsepower during rendering. I really
> hope someone comes out with some kind of SCSI interface for the
> CD32 and writes the software to network it to any Amiga. Not
> only would you have a CD-ROM drive, but also a display buffer,
> an additional CPU, and optional hardware MPEG support.
Jeff...
This is POSSIBLE, but it's not probable. First of all, CD32 does not
have SCSI. Second of all, CD32 is powered by a low-end 60EC020, without a
math co-processor. You're not gonna get a signifigant boost of rendering
speed from CD32 unless you're maybe used to a 68000 for rendering.
Also, CD32 has 2 megs of RAM [I think].. That's just enough to play
games, hardly enough to use a 3D program.
The MPEG would be a plus tho, as would it's display abilities as a near-
24-bit frame buffer. But I think the closest you'll get to this idea is to
get an A1200 [with some Xtra memory and an 030] and wait for the CD32 add-on
module promised from Commodore.
Myself, I've got a similar setup, but with all different equipment than
you mentioned.. Personal Animation Recorder to play anims instead of MPEG
and OpalVision [with OpalPaint] as a true 24-bit display device. It may be
more expensive than the setup you suggested, but it's a cleaner display,
broadcast quality and it's possible. :)
Sorry to shoot your idea full of holes like that..
Carmen Rizzolo - Crazed Artist
CarmenR@cup.portal.com
##
Subject: CD32 and Imagine
Date: Thu, 11 Nov 93 20:26:34 EST
From: mart4678@mach1.wlu.ca (Phil Martin u)
>
> Wouldn't it be neat if you could hook up a CD32 to your amiga and
> use it as some additional horsepower during rendering. I really
I know what you mean... I've been thinking about this sort of thing
too. (Not rendering specifically, just trying to think of something
cool I could have a 'slave' CD32 doing...
>
> Just think, you could send your MPEG animation to the CD32 via
> SCSI and use its hardware MPEG decoding to play animations at
> a full 30 fps in 24-bit at full screen resolution!
Yes, plus it has colour composite out (and supposedly it's a quite
high-quality signal - I've heard MPEG movies look better than SVHS,
although not broadcast quality) so it would be ideal for that sort
of thing. You could even use the CD32 to build the MPEG as the anim
is being rendered, send it back to the A4000 (or whatever) which
would have some big HD (and act like a 'server'), then when it's done,
the CD32 could play it off the A4000's HD....
>
> Or you could use the CD32 to process images as they complete rendering
> and them send them back. The image processing part would not slow down
> (very much) the additional frames rendering.
Another neat idea, since (in general) most image processing would be
a lot faster than rendering, but still quite time (and memory) consuming.
>
> I wish I had to time to write something like this myself. It would
> be a very interesting project.
I have to agree... Maybe someday...
>
> -Jeff Wahaus-
>
Phil Martin.
##
Subject: Candle flame
Date: Thu, 11 Nov 93 23:20:41
From: greg.tsadilas@hofbbs.com
Read the info on the FIREBALL texture. It describes how to make a candle
flame. If you don't have time to hunt for it, here it is:
"To do fire: create a candle flame shaped object. Apply the texture and
selext "Edit Axis" from the texture requestor. Move the center of the texture
axis to the base of your flame. Now scale the texture axis until ... [the]
axis extends beyond the top of your object. In the texture requestor, set the
noise to 0.3 and the "Filter 2" param to 1.0.
Make the object bright."
I haven't tried this yet, let me know how it works.
GreG.tsadilas
##
Subject: Imagine v2.9
Date: Fri, 12 Nov 93 02:22:28 PST
From: CarmenR@cup.portal.com
Could someone please post a more in-depth review of Imagine 2.9? Tell us how
the new features, has imagine retained it's old feel [especially the Amiga
versions]? Does it have a different look? Is the light source for the
interface buttons still coming from the upper-right hand corner? How is the
realtime Perspective? How is the particle stuffs? Thanks.
Carmen Rizzolo - Crazed Artist
CarmenR@cup.portal.com
##
Subject: Re: CD32 and Imagine
Date: Fri, 12 Nov 93 09:37:09 CST
From: setzer@comm.mot.com (Thomas Setzer)
>
> Wouldn't it be neat if you could hook up a CD32 to your amiga and
> use it as some additional horsepower during rendering. I really
> hope someone comes out with some kind of SCSI interface for the
> CD32 and writes the software to network it to any Amiga. Not
> only would you have a CD-ROM drive, but also a display buffer,
> an additional CPU, and optional hardware MPEG support.
>
You could use parnet, I think. I doubt the usefulness for rendering, but
it would give older Amiga owners AGA playback(MMMMMPEG) and a cd rom drive
accessable through parnet. Does CD32 have a paralell port?(I think thats
what parnet uses)
Tom Setzer
setzer@ssd.comm.mot.com
"And of course, I'm a genius, so people are naturally drawn to my fiery
intellect. Their admiration overwhelms their envy!" - Calvin
##
Subject: Re: Several posts...
Date: Fri, 12 Nov 93 10:00:11 CST
From: setzer@comm.mot.com (Thomas Setzer)
Speaking of fire(FIRE FIRE FIRE:), is fog fixed in 2.9? For those of you who
have 2.9, please try rendering two fog objects, one in front of the other.
Also, I beleive its FOG, not Gas. Aladin uses 'gas', and implements it quite
differently than Imagines fog. From what I've heard, Aladins gas has true
volumetric properties.
Is brush/texture tacking implemented in 2.9? Make an cylinder, apply a texture,
save it. Bend the cylinder. save it as a different name. Morph 'em. How do
the textures look? Do they 'swim'?
Hmm, what else? Anyone play with particles?(I would have done this first!:)
What does it allow? Whats it do for you?
Tom Setzer
setzer@ssd.comm.mot.com
"And of course, I'm a genius, so people are naturally drawn to my fiery
intellect. Their admiration overwhelms their envy!" - Calvin
##
Subject: Re: Imagine v2.9
Date: Fri, 12 Nov 1993 11:30:15 -0500
From: Jason B Koszarsky <kozarsky@cse.psu.edu>
>Could someone please post a more in-depth review of Imagine 2.9?
Has it been optimized for different processors(030/040)? How much faster
does it render? Are we still limited to 640x400?
I'd also like to here more about the audio track support.
Jason K.
##
Date: Sat, 13 Nov 1993 05:37:11 +1100
From: imagine-relay@email.sp.paramax.com
> Nope, I'm running bare-bones DOS, nothing in the background, just Imagine.
> I have the same problem on both the PC AND the Amiga.
>
> Imagine is just a RAM-hog, IMHO.
*Any* 3-D package hogs memory... just like any image processor...
they're just memory-intensivce sorts of things.
--
Gordon Schumacher
/-------------------------------------------------------------------\
| Champaign- "We apologize for the inconvenience." _@_ |
| Urbana -HHGTTG / \ |
| kilroy was here | o o | |
\-------------------------------------------------------U|--U--|U---/
##
Subject: CD32 as a Slave continued...
Date: Fri, 12 Nov 1993 08:00:06 -0700
From: "Jeff Wahaus, CAPS, ATL, 404-640-3529" <JEFF_W1@sfov1.verifone.com>
> With a sparkle in his eyes, Jeff Wahaus writes...
>> Wouldn't it be neat if you could hook up a CD32 to your amiga and
>> use it as some additional horsepower during rendering. I really
>> hope someone comes out with some kind of SCSI interface for the
>> CD32 and writes the software to network it to any Amiga. Not
>> only would you have a CD-ROM drive, but also a display buffer,
>> an additional CPU, and optional hardware MPEG support.
Carmen Rizzolo recently interjected...
> Jeff...
>
> This is POSSIBLE, but it's not probable. First of all, CD32 does
> not have SCSI. Second of all, CD32 is powered by a low-end 60EC020,
> without a math co-processor. You're not gonna get a signifigant boost
> of rendering speed from CD32 unless you're maybe used to a 68000 for
> rendering.
Well I don't know why you think it's not probable. A product like
I suggested would take off like sliced bread. The CD32 does not have SCSI
but it does have an expansion bus (you know where the MPEG add on goes).
So all you need is 1 expansion card which has on it MPEG and SCSI. Ideally
you would have MPEG, SCSI, and RAM and maybe even a math coprocessor
socket.
As far as rendering goes, no it wouldn't be practical to use a CD32 for
rendering although it would be possible. What you could use the CD32 for
is 24-bit to HAM6 or HAM8 conversion. And then even HAM to ANIM
conversion. If your rendering animation is taking 1 hour per frame then
the 14 MHz 68020 in the CD32 could easily do the above conversions in that
time. The CD32 could even preview the completed part of the animation
while the rest is still rendering. A program like HamLab (around 150K)
which supports using disk space as memory could be run on the CD32 without
modification. You could control it using ARexx. That's assuming you can
pass ARexx messages to the CD32 which would be quite possible with the
right software.
> The MPEG would be a plus tho, as would it's display abilities as a
> near- 24-bit frame buffer. But I think the closest you'll get to
> this idea is to get an A1200 [with some Xtra memory and an 030] and
> wait for the CD32 add-on module promised from Commodore.
Um, well I don't want a A1200 or even an A4000. I quite satisfied with
my Amiga 2000. It's 5 years old now. It has got 5 video output ports:
Monochrome Composite, Standard ECS RGB, 31KHz ECS RGB (via an A2320),
GVP's Spectrum RGB, and SVGA. It runs at 25MHz with its 68030/68882. Has
a bridge board with a Quantum HardCard+, dual serial ports, SVGA graphics.
It has 10MB of RAM (9MB 32-bit), fast SCSI, Quantum 120MB SCSI, Commodore
SCSI tape drive, 3 1/2" high density floppy, 5 1/4" high density floppy.
Oh yea did I mention 68000 fall back mode for old games. And all of this
is inside the case (except the tape drive). And there is even room for
more expansion cards and maybe even another SCSI hard disk.
The point here is that there is no way I'm going to dump this for an
Amiga 1200. An Amiga 4000 wouldn't come close to holding all of my
expansion cards. A CD32 daisy chained externally to my SCSI bus would be
the perfect way to add AGA compatibility, a CD-ROM drive, and a CD32! :)
> Myself, I've got a similar setup, but with all different equipment
> than you mentioned.. Personal Animation Recorder to play anims instead
> of MPEG and OpalVision [with OpalPaint] as a true 24-bit display device.
> It may be more expensive than the setup you suggested, but it's a
> cleaner display, broadcast quality and it's possible. :)
The PAR will cost you about $2300 when you include the price of the
required IDE (yuk!) hard disk. The CD32 will cost you about $300 plus
another $200 if you want MPEG. Of course for the CD32 setup to be of any
use someone will have to write the networking software and design the CD32
to Amiga SCSI interface card for the CD32.
The CD32 can display AGA graphics over the top of high quality MPEG video
which is something that I don't think the PAR can do.
> Sorry to shoot your idea full of holes like that..
> Carmen Rizzolo - Crazed Artist
Well don't be sorry, I think you just missed the target. Aim again. :)
-Jeff Wahaus-
jeff_w1@verifone.com
##
Subject: Still more on CD32 as a Slave
Date: Fri, 12 Nov 1993 08:40:25 -0700
From: "Jeff Wahaus, CAPS, ATL, 404-640-3529" <JEFF_W1@sfov1.verifone.com>
Another unbeliever recently wrote:
> A SCSI interface won't let you control CD32 remotely... you'd be lucky
> if you just get the CD-ROM drive in CD32 working on other Amiga.
I must disagree here, a SCSI interface and properly written SCSI based
networking software would allow you to do whatever you wanted. This is
assuming that you were a skilled enough programmer to write such a thing.
It just so happens that I am. ;->
> Also, as for using CD32 as a 'rendering' engine, 68020 isn't too
> fast and CD32 doesn't have math coprocessor.
I don't think I mentioned anything about using the CD32 as a rendering
engine (which is possible). It could be used to do lots of other useful
things though, like playing a MPEG animation in 24-bit color for example.
> If you want to get CD-ROM capability that bad, just buy a CD-ROM drive
> or just get CD32. Why all the trouble? Nobody's going to bother
> buying CD32 FOR THEIR AMIGA. Weird thought...
Actually I was speaking of doing just that. You purchase a CD32 and a
SCSI adapter for it (assuming someone makes one) and you have:
(1) A double speed CD-ROM disk drive
(2) A CD32 (of course)
(3) A slave computer capable of running off the shelf Amiga Software.
(4) Optional hardware MPEG support
You will not be able to run CD32 software if you just get any CD-ROM
drive. Also, why would you want a CD32 if you couldn't access standard
format CD-ROM's with your Amiga?
> The MPEG is another issue... Commodore will have FMV module for every
> (ok, some) Amiga in the future soon or later, so this isn't necessary.
Did you use the words "Commodore" and "soon" in the same sentence?
Actually using any of the words "soon", "shortly", "quickly", or "next
month" in the same sentence with "Commodore" is an oxymoron and should be
avoided.
> Dream on! :)
Hieratic!
-Jeff Wahaus-
jeff_w1@verifone.com
##
Subject: Another Imagine 2.9 Q
Date: Fri, 12 Nov 93 09:43:27 PST
From: CarmenR@cup.portal.com
Oh yeah.. Could someone tell me how the Field rendering is in Imagine 2.9
now? Someone with single-frame abilities or a PAR? Thanks!
Carmen Rizzolo - Crazed Artist
CarmenR@cup.portal.com
##
Subject: CD32 as a Slave continued
Date: Fri, 12 Nov 1993 13:06:00 -0500
From: roy.park@canrem.com (Roy Park)
> Well I don't know why you think it's not probable.
Ok, not improbable, IMPOSSIBLE.
>The CD32 does not have SCSI
^^^^^^^^^^^^^^^^^^^^^^^^^^^
And that's EXACTLY WHY it makes so tricky to make such hardware... and the
CD-ROM drive used in CD32 is NOT SCSI! Get it? NOT SCSI!
> As far as rendering goes, no it wouldn't be practical to use a CD32 for
>rendering although it would be possible. What you could use the CD32 for
>is 24-bit to HAM6 or HAM8 conversion. And then even HAM to ANIM
>conversion.
Why bother? If you have 68030 already, it's a LOT faster at doing image
conversion than 68020 that CD32 has!
[stuff about how much things you have on your A2000 deleted]
> The point here is that there is no way I'm going to dump this for an
>Amiga 1200. An Amiga 4000 wouldn't come close to holding all of my
>expansion cards.
Of course it WOULD! According to what you said, you are using 3 Zorro slots
(Bridgeboard, Spectrum, dual serial) 2 internal 3.5" bays and 1 internal
5.25" bay... which WILL fit into A4000 without any problem.
>A CD32 daisy chained externally to my SCSI bus would be
>the perfect way to add AGA compatibility, a CD-ROM drive, and a CD32! :)
Ya, ya.. dream on...
Remember, this is different from A500 expansion chassis with SCSI interface
because CD32's CD-ROM drive is NOT on the SCSI chain.
Also, just by having a SCSI interface on CD32 does not guarantee that you
will be able to access the CD-ROM drive from another Amiga with SCSI
controller (because of two SCSI host conflict situation).
> The PAR will cost you about $2300 when you include the price of the
>required IDE (yuk!) hard disk. The CD32 will cost you about $300 plus
>another $200 if you want MPEG.
However, it will never work... so people take PAR. (not that I'm thinking that
PAR is the best solution)
>Well don't be sorry, I think you just missed the target. Aim again. :)
Mr. Rizzolo hit the target right on the bull's eye.... it's just that you
can't face the reality.
+--------------------------------------------------------------+
|Roy Park | C= A3000@25Mhz-14.4HST-2MBPicassoII|
|roy.park@canrem.com | // DM3024-ST296N-LP240S-LP240S-LP105S|
|rkpark@io.org | \X/ ViewSonic17-2MBChip/12MBFast-Emplant|
+--------------------------------------------------------------+
##
Subject: Re: Imagine v2.9
Date: Fri, 12 Nov 93 12:48:49 -0700
From: Steve Koren <koren@hpfcogv.fc.hp.com>
Jason B Koszarsky <kozarsky@cse.psu.edu> wrote:
> Has it been optimized for different processors(030/040)? How much faster
> does it render? Are we still limited to 640x400?
Howzat? Limited to 640x400? Since when?
I've rendered plenty of images at 1280x1024, and I think Imagine has no
internal limit. I recall setting the image size at up to 4000x4000
before, and Imagine accepted it just fine. I always render final images
in 1280x1024 or 896x628. 640x400, heck, that's *preview* resolution :-)
Either your Imagine is a lot different than mine, or you meant something
else by "limited to 640x400" and I'm completely missing your point.
- steve
##
Subject: New Speed
Date: Thu, 11 Nov 93 17:58:54 EST
From: imagine@bknight.jpr.com (Yury German)
So with a few people gettign 2.9 on PC's and soon the rest will start
gettign 3.0 and such we still do not know about one thing@!!!!!
HOW IS THE RENDERING SPEED??????
I mean we all learned (or were taught) some tricks for speeding up the
rendering but it was still not close to Lightwaves speed in 2.0. So
basically can one of you new 2.9 - 3.0 people render and let us know if
there is a speed increase??? It is pretty important for all of us..
considering the extended rendering time with Imagine
##
Subject: Re: CD32 and Imagine
Date: Fri, 12 Nov 93 12:08:00 PST
From: Stethem Ted 5721 <TedS@ms70.nuwes.sea06.navy.mil>
The CD32 doesn't have a parallel port. There was a photograph of the
motherboard in a recent Amiga magazine, forget which one right now. Anyway,
it does have a big "expansion" connector, or more accurately, edge card
connection. The only thing that Commodore will fess up to is that it will
have a keyboard and disk drive expansion that hooks up to that edge card.
The disk drive is being provided strictly for game saves. Commodore is
taking the official position that the CD32 is a game machine, period.
Apparently, Commodore's attempt to promote the CDTV as a multimedia,
information database, game, educational, multipurpose machine that could be
expanded to become a computer, too (A500), confused and perplexed the mass
market so almost nobody bought it. Now, Commodore is saying, " The game
comes on a compact disk, put it in and play!", hoping the mass market will
understand that much.
Anyway, I heard that Fred Fish was at the World of Commodore in Berlin,
with a CD32 with the keyboard expansion and he has found out that some C++
development program WILL run on the CD32, as is. So, there is the
possibility that Imagine could run on the CD32 directly, especially if the
disk drive comes out. The CD32 is basically a stripped down A1200 anyway,
AGA chipset, 2 MB of RAM, although it does have a special chunky-to-planar
chip which may make for some graphics incompatibilities. Don't really see
the point in running Imagine on the CD32 anyway, even if it were possible.
The CD32 has a pretty slow 68020 and the RAM is fixed. Too many
limitations. The really positive outcome is that the CD-ROM is supposed to
become available for the Amiga computer late this year or early next year.
The CD-ROM drive is a double speed, faster access type. Commodore
developed a special version of Workbench to control the drive and they are
supposed to be in negotiation with Kodak to include Kodak-CD compatibility.
So, Commodore is saying that they will probably be coming out with a new
Workbench which will include the CD-ROM additions. Lew Eggebrecht is also
saying that the next Workbench will not only put a speech synthesizer
program back in (but not Say) and it will be multi-lingual.
OK, this doesn't have anything to do with Imagine so I better stop wasting
bandwidth. It would be nice to have some kind of high-speed rendering
engine, though. Warp, Screamer, Resolver, Excalibur ..... ?
----------
From: imagine-relay
To: imagine
Subject: Re: CD32 and Imagine
Date: Friday, November 12, 1993 9:37AM
>
> Wouldn't it be neat if you could hook up a CD32 to your amiga and
> use it as some additional horsepower during rendering. I really
> hope someone comes out with some kind of SCSI interface for the
> CD32 and writes the software to network it to any Amiga. Not
> only would you have a CD-ROM drive, but also a display buffer,
> an additional CPU, and optional hardware MPEG support.
>
You could use parnet, I think. I doubt the usefulness for rendering, but
it would give older Amiga owners AGA playback(MMMMMPEG) and a cd rom drive
accessable through parnet. Does CD32 have a paralell port?(I think thats
what parnet uses)
Tom Setzer
setzer@ssd.comm.mot.com
"And of course, I'm a genius, so people are naturally drawn to my fiery
intellect. Their admiration overwhelms their envy!" - Calvin
##
Subject: Re: Imagine v2.9
Date: Fri, 12 Nov 1993 14:53:43 -0600 (CST)
From: Schumacher Gordon C <gschumac@ux4.cso.uiuc.edu>
> Either your Imagine is a lot different than mine, or you meant something
> else by "limited to 640x400" and I'm completely missing your point.
Yup... he meant working res.
--
Gordon Schumacher
/-------------------------------------------------------------------\
| Champaign- "We apologize for the inconvenience." _@_ |
| Urbana -HHGTTG / \ |
| kilroy was here | o o | |
\-------------------------------------------------------U|--U--|U---/
##
Subject: Re: Imagine v2.9
Date: Fri, 12 Nov 93 14:23:39 -0700
From: Steve Koren <koren@hpfcogv.fc.hp.com>
In a fit of stupidity, I wrote:
> Either your Imagine is a lot different than mine, or you meant
> something else by "limited to 640x400" and I'm completely missing your
> point.
Turns out the latter is true. He obviously meant "promoting Imagine's
screen to more than 640x400", not "rendering to more than 640x400".
I hope they fixed that too.
- steve (mr. stupid) k
##
Subject: Re: Imagine v2.9
Date: Fri, 12 Nov 1993 15:27:43 -0600 (CST)
From: Trin Yuthasastrackosol <yuthas@cc.umanitoba.ca>
On Fri, 12 Nov 1993, Steve Koren wrote:
>
> Jason B Koszarsky <kozarsky@cse.psu.edu> wrote:
>
> > Has it been optimized for different processors(030/040)? How much faster
> > does it render? Are we still limited to 640x400?
>
> Howzat? Limited to 640x400? Since when?
>
[Stuff deleted for the sake of brevity]
> Either your Imagine is a lot different than mine, or you meant something
> else by "limited to 640x400" and I'm completely missing your point.
I think he means the resolution at which you can use Imagine in it's
various editors. Some people like to have the capability to work in screens
much larger than 640 x 400 or whatever, particularly those with the new
graphics boards like Picasso, Spectrum etc...You can do it with Imagine
2.0 by patching some bytes here and there, but the program has some trouble
working in the larger modes (I *THINK*...i remember reading the thread on
here not long ago but my memory gets worse and worse as I approach age 30
(^:). It would be pretty nice if the size could be easily configured.
Trin Dominic Yuthasastrakosol | Definition: Bigot (noun). Someone
Dept. Pharmacology and Therapeutics | who zealously holds a set of ideas,
University of Manitoba | beliefs or opinions which are
A5002MbChip1MbFastSupraturbo28GVPII354Mb| different from yours.
##
Subject: Imagine 2.9 Questions
Date: Fri, 12 Nov 93 14:11:55 PST
From: Scott Lundholm <scottl@hpsadmq.sad.hp.com>
Hi All,
Just a few short questions about Imagine 2.9 that I have not seen asked:
1: Is the Amiga Version out yet?
2: If yes to #1, does it load v2.0 objects?
3: Does it use Essence I & II textures on 2.0 objects?
4: Does it use Essence I & II at all????
5: Does it load v2.0 stage files?
Thats 'bout it. Take care All, and keep rendering. Kudos (spelling?) to
Essence II, FANTASTIC product!!!!!!!!
Scott Lundholm
*----------------------------------------------------------------------------*
| Amiga 500/030 at 38Mhz w/68882 at | Email (UNIX)-scottl@hpsadmq.sad.hp.com |
| 50Mhz, 8Meg of 32-bit RAM, Bodega | Memeber |
| Bay w/105 & 450 Meg Hard Drives, | Redwood Empire Amiga Users Group |
| HP PaintJet, 16" HP Programmable | Imagine 2.0 rendering enthusiast and |
| Multisynch w/Flicker Free Video. | graphics fanatic!!!!!!!!!!!!!!!!!!!! |
*----------------------------------------------------------------------------*
##
Subject: Re: CD32 and Imagine
Date: Fri, 12 Nov 1993 18:09:00 -0500
From: roy.park@canrem.com (Roy Park)
> I think SCSI would be the best way to implement this sort of thing. The
>CD32 should be able to transfer about 1.5M bytes per second with a properly
>designed SCSI interface.
If you want a POSSIBLE and FEASIBLE solution, it would be to have Ethernet
interface for CD32... it doesn't cost that much on hardware over the price
of SCSI (whatever, the one you are suggesting) interface.
+--------------------------------------------------------------+
|Roy Park | C= A3000@25Mhz-14.4HST-2MBPicassoII|
|roy.park@canrem.com | // DM3024-ST296N-LP240S-LP240S-LP105S|
|rkpark@io.org | \X/ ViewSonic17-2MBChip/12MBFast-Emplant|
+--------------------------------------------------------------+
##
Subject: Re: Imagine v2.9
Date: Sat, 13 Nov 1993 00:54:41 -0500
From: Jason B Koszarsky <kozarsky@cse.psu.edu>
>Howzat? Limited to 640x400? Since when?
>
>I've rendered plenty of images at 1280x1024, and I think Imagine has no
I'm asking about the work screen size not the rendering output. I'd like to
be able to work on a screen larger than 640x400.
Jason K.
##
Subject: Re: Imagine pc-help
Date: Sat, 13 Nov 1993 18:35:57 +1100 (EST)
From: Nikola Vukovljak <nvukovlj@ucc.su.OZ.AU>
On Wed, 10 Nov 1993, Udo K Schuermann wrote:
> Nikola Vukovljak <nvukovlj@ucc.su.oz.au> writes:
> > I have found that you would need about 12 Meg to render a scene with
> > about 20000 faces. This is on the Amiga.
>
> This is by no means universally applicable, as CastleRoom consisted of more
> than 20000 faces and rendered in less than 6MB of RAM. That was with a
> fair bit of refraction and reflection, including three light sources and
> two bitmap images. There are many more factors involved than just #faces,
> although I'd be hard pressed to enumerate them all.
>
> ._. Udo Schuermann
> ( ) walrus@wam.umd.edu
>
Ok, well that is strange. The object I created (a huge group actually),
had no textures, no refractions, some parts had minimal reflection settings.
Many didn't use Phong. Couldn't be simpler, render engine wise (hard to
model thanks to multiple layers though).
Maybe because I rendered everything in Scanline ? The object was 1.2M in
size.
Nik.
##
Subject: Re: Upgrade and Giveaway
Date: Sat, 13 Nov 1993 18:44:03 +1100 (EST)
From: Nikola Vukovljak <nvukovlj@ucc.su.OZ.AU>
On Wed, 10 Nov 1993, Jeff Saffold wrote:
>
> > And speaking of questions, the Amiga Formatters are showing up in
> > amiga.graphics and starting to ask all sorts of elementary things that the
> > manual would have even answered. They even have the nerve to complain about
> > the program! They'll discover us soon enough so I'm going back into hiding
> > now.
>
> Well, I just talked to Impulse, and they said that the AF Upgrade
> offer was ONLY for the European people, not for the American/NTSC
> users... So anyone wanting to go that route is SOL...:)
>
> // Jeff Saffold
> \X/ Only the Amiga makes it possible.
I wonder what they'll say about Australians. No one ever thinks of the
market here and yet Imagine is the no.1 Amiga 3D program here (Real 3D is
fast caching up). BTW we use PAL so I wonder what Impulse will do.
Not that I care either way since I'm waiting for 'almost 3.0'.
Speaking of which, has Amiga version shipped yet ?
Nik.
##
Subject: Re: New Speed
Date: Sat, 13 Nov 1993 19:13:50 +1100 (EST)
From: Nikola Vukovljak <nvukovlj@ucc.su.OZ.AU>
On Thu, 11 Nov 1993, Yury German wrote:
> So with a few people gettign 2.9 on PC's and soon the rest will start
> gettign 3.0 and such we still do not know about one thing@!!!!!
>
>
> HOW IS THE RENDERING SPEED??????
>
> I mean we all learned (or were taught) some tricks for speeding up the
> rendering but it was still not close to Lightwaves speed in 2.0. So
> basically can one of you new 2.9 - 3.0 people render and let us know if
> there is a speed increase??? It is pretty important for all of us..
> considering the extended rendering time with Imagine
>
>
Well, someone on here said that the Amiga version of Imagine 2.9 will not
be optimized for '040. They 'may' optimize v 3.0.
Great! Everyone else can do it but not Impulse! They'll probably tell us
to get a PC if we want more speed. I don't wanna do that. I know that
they can optimize Imagine after some speed tests I've done. LW has been
optimized, Real 3D will be as of v2.40. So, I think that it is time for
all the '040 owners (and '030) to contact Impulse and let them know that
we'd like to see this happen to Imagine. Imagine started on the Amiga and
its original users should be properly supported.
Anyway, that is my two cents worth, so it is up to all of you guys to act
now before it is too late.
I intend to. (Can someone repost that Fax no. of Impulse please ?)
Nik.
##
Subject: Re: Imagine pc-help
Date: Sat, 13 Nov 1993 11:04:31 -0500 (EST)
From: Udo K Schuermann <walrus@wam.umd.edu>
Nikola Vukovljak <nvukovlj@ucc.su.oz.au> writes
> On Wed, 10 Nov 1993, Udo K Schuermann wrote:
>
> > Nikola Vukovljak <nvukovlj@ucc.su.oz.au> writes:
> > > I have found that you would need about 12 Meg to render a scene with
> > > about 20000 faces. This is on the Amiga.
> >
> > This is by no means universally applicable, as CastleRoom consisted of more
> > than 20000 faces and rendered in less than 6MB of RAM. That was with a
> > fair bit of refraction and reflection, including three light sources and
> > two bitmap images. There are many more factors involved than just #faces,
> > although I'd be hard pressed to enumerate them all.
> >
> Ok, well that is strange. The object I created (a huge group actually),
> had no textures, no refractions, some parts had minimal reflection settings.
> Many didn't use Phong. Couldn't be simpler, render engine wise (hard to
> model thanks to multiple layers though).
>
> Maybe because I rendered everything in Scanline ? The object was 1.2M in
> size.
That would explain it. Scanline is hungrier than rayrace; my scene
was raytraced because of reflection, refractions, and the shadows.
._. Udo Schuermann
( ) walrus@wam.umd.edu
##
Subject: Re: Imagine v2.9
Date: Sat, 13 Nov 1993 11:23:52 -0600 (CST)
From: Cliff Lee <cel@tenet.edu>
On Sat, 13 Nov 1993, Jason B Koszarsky wrote:
> I'm asking about the work screen size not the rendering output. I'd like to
> be able to work on a screen larger than 640x400.
PC Imagine 2.9's native screen size is 640 x 480. If that helps.
-=-=-=-=-=-=-=-=-=-=-=-=-
Cliff Lee
cel@tenet.edu
"Everything will work out if you let it!" Cheap Trick
##
Subject: Question: IMAGINE vs 3DS ?
Date: Sat, 13 Nov 1993 18:26:02 GMT
From: "Janusz Kalinowski" <sjk@pwm.ict-pwr.wroc.edu.pl>
Hi IMAGINE gurus !
I consider buying some flexible 3D program which include
(serious) modelling, rendering and professional animation (on the
486-based machine). I'd like to use it BOTH to create non-trivial
3D models for further realistic rendering (by third party
renderers) and, independently, for nice animation (I suppose
simpler models will be used then).
I was almost decided on 3D Studio, but in some Ray Tracing News
I've found info about PC version of IMAGINE. I know nothing about
that program except that, what was written in RTN.
Could you pass me a bit more info about possibilities of IMAGINE?
Especially, does it include professional modeller ? What are the
pros/cons comparing to 3D Studio ? What is the current
version/price ? Is there some ftp-able IMAGINE demo ?
Aswer directly to me PLEASE. As I'm not (yet?) IMAGINE user, I'm
NOT subscribed to this list.
many thanks,
Janusz Kalinowski
Technical University of Wroclaw, Poland
Computing-Science Center
##
Subject: Re: Imagine v2.9
Date: Sat, 13 Nov 93 12:23:33 MET
From: boinger@myamy.hacktic.nl (Paul Kolenbrander)
Hello Steve Koren,On 12 NOV 1993 12:48:49 you said regarding Re: Imagine v2.9:
> > Has it been optimized for different processors(030/040)? How much faster
> > does it render? Are we still limited to 640x400?
>
> Howzat? Limited to 640x400? Since when?
>
> Either your Imagine is a lot different than mine, or you meant something
> else by "limited to 640x400" and I'm completely missing your point.
I think he means the window the editor opens. Which has always been limited to
640x400. Although I'm now running it in 800x600 on my Piccolo. No weird crashes
yet. Although dragging an object erases the grid. 'Redraw' helps restoring it.
-- __
--/_/ |/ ------------+--Main : boinger@myamy.hacktic.nl-
-/aul |\olenbrander--+--Spare : paulk@stack.urc.tue.nl---
---------------------+-----------------------------------
-Reg. CD32 developer-+-A4000/040&A3000. Best of 2 worlds-
##
Subject: More on 2.9 new stuff
Date: Mon, 15 Nov 1993 12:06:57 -0600 (CST)
From: Cliff Lee <cel@tenet.edu>
I have continued my experimentation with 2.9. Upon the recomendation of
others, I have tried to play with the Particle feature. I added a sphere,
and then chose particle. Chose to inscribe random size particles and
perform. It worked. I quick rendered it and I had a tight ball of
particles. Neat!
I have also been playing with the spline editor. This is nice! It
supports postscript fonts. Load the font, type the text, then you have a
spline outline of the letters. Choose add points and tell the kind of
beveling. Viola! You have an object that can be loaded into the detail
editor. The KNOTS can be modified. Interesting stuff.
I am not an expert at using Imagine, so I sure I'm not pushing this. I
have not tried any of the cycle editor stuff
-=-=-=-=-=-=-=-=-=-=-=-=-
Cliff Lee
cel@tenet.edu
"Everything will work out if you let it!" Cheap Trick
##
Subject: More 2.9 questions
Date: Mon, 15 Nov 93 12:07:10 -0700
From: Steve Koren <koren@hpfcogv.fc.hp.com>
Cliff Lee wrote:
> I have continued my experimentation with 2.9. Upon the recomendation of
> others, I have tried to play with the Particle feature. I added a sphere,
Some questions for anyone who has 2.9. BTW, apologies in advance if
some of these have been answered already and I missed them.
* First and foremost: Does Essence work? As I consider Imagine
without Essence near worthless, I won't upgrade if it doesn't. (I
know they have supposedly added more textures, but I doubt they'll
add equivalents of all the ones I use from Essence).
* Have they fixed any of the transparency bugs?
* As someone else asked previously (and I misinterpreted), can
Imagine's screen be promoted to better resolutions without crashing
Imagine? Does it support the display database directly?
* Do shadows in scanline work right?
* Is Imagine any more multi-threaded than it was in 2.0? I always end
up running two copies of Imagine at once; one to render and one to
model. Even if they compiled with the "pure" bit so that the
overhead of two concurrent Imagines was less, I'd be happy. But a
true render-while-modeling ability would be nicer.
* I desperately want Aladdin 4-like gas handling. Is there anything
in 2.9 which will help? (Fog objects are a real pain for this and
have many limitations that true volumetric gasses don't have). I've
actually considered buying Aladdin to augment Imagine in this area,
since it seems to have wonderful abilities with gasses. If it was
$75 or so cheaper I'd have done it by now.
- steve
##
Subject: Trace time vs scanline
Date: Mon, 15 Nov 93 09:44:00 PST
From: Stethem Ted 5721 <TedS@ms70.nuwes.sea06.navy.mil>
Hello there, Imagineers!
I have been working on an animation with an undersea scenario. I am
using a 3D landscape imported from VistaPro3, with bump texture and a 24-bit
sandstone brushmap, as the ocean floor. This object is about 500K. I have
several buildings including a Greek temple, and a couple other architectural
objects representing a "lost" city on the ocean bottom, ala Atlantis. They
range from 100K to 300K each. I have a submarine traveling over the bottom,
it is about 300K. In addition, I have a Plane object acting as the ocean
surface over the whole landscape. This Plane has a seawave texture applied
to it. I also have several of these morphing over time so I get wave
motion. At this time, I am experimenting with a Plane object in front of
the camera with a linear Ripple texture applied to it. In addition, I have
the Global fog attribute set so the background fades off.
So, I tried all this in Scanline, and found out that I was getting no
refraction, ripple, or shadowing, as obviously stated for Scanline mode.
But, it was taking only about 3 to 5 minutes per frame. Since I found out
that I couldn't get the desired water effects with Scanline mode, I went to
Trace. Now, it is taking over 9 HOURS per frame to render! While the
effects are working now, pretty close to what I wanted, the render time is
just too long. At this rate, it will take a month to render 100 frames or
about 4 seconds of animation.
I am running an A3000 with a 25 MHz '040 accelerator, 14 Megs of RAM, 213
Megabyte Maxtor harddrive, 52 Meg Quantum HD. The Maxtor is averaging
about 500KB/sec transfer rate. I have Imagine on this drive. The Quantum
averages about 1 MB/sec after using TurboQuantum but it is my system disk
and I don't have room on it for Imagine. I have tried cutting up my
landscape object in smaller sections, placing them just when I needed them
in the camera's field of view, but then I get a "chopped off" look.
I guess my question is, if this is a typical time for Trace rendering
based on the number and size of the objects I am using and the amount of
textures being applied? I am thinking about getting a RAM expansion board
which would cut down on the Hardrive access time. Other than that, is there
anything obvious or simple I could be doing to reduce my rendering time? I
do have the Global size attribute set for 0,0,0. I guess most of my render
time is coming from the Global fog and multiple ripple textures. These add
the most "atmosphere" for my scenes, though, so I need them there. I was
just wondering if anybody had any experience with Lightwave versus Imagine
for Trace renderings and if Lightwave were substantially faster or not?
Real-time 3D trace renderings? Not in this lifetime!!!
##
Subject: imagine on ftp
Date: Mon, 15 Nov 93 19:08:39 +0100
From: humez@idnges.decnet.citilille.fr
Imagine2.0 is now available on ftp aminet. It can be found in
aminet/biz/demo. I didn't try it, but I think AmigaFormat's cover disk is
now on PD.
Stanis
##
Subject: Re: RAM-hogs
Date: Mon, 15 Nov 1993 10:19:15 -0800 (PST)
From: Doug Kelly <dakelly@class.org>
> > Imagine is just a RAM-hog, IMHO.
>
> *Any* 3-D package hogs memory... just like any image processor...
> they're just memory-intensivce sorts of things.
> --
> Gordon Schumacher
>
NOT TRUE. I happen to know of one that can render Carmen Rizzolo's
NCC-1701D in less than 8 megs of RAM. Stay tuned for further details.
------------------------------------------------------------------------------
Doug Kelly Information Specialist First Consulting Group
dakelly@class.org (310)595-5291x125 P.O.Box 5161, Los Alamitos,CA 90721-5161
"The difference between genius and stupidity: genius has its limits."
------------------------------------------------------------------------------
##
Subject: 2.9 is _HERE_ !!!
Date: 15 Nov 1993 21:46:50 GMT
From: <mbc@po.cwru.edu>
Yes!
2.9 Amiga version just go here today. Even though I have several
projects (homework ! ack!) to do I will try to post a comprehensive review
and answer all those questions soon....NOT!
Seriously...I will try to get into tonight and will let everyone know
what it's like. Just couldn't resist posting here. Sorry!
Render on!
Mike C.
mbc@po.CWRU.Edu
##
Subject: Re: Trace time vs scanline
Date: Mon, 15 Nov 93 16:22:58 EST
From: Mark Thompson <mark@westford.ccur.com>
> I have been working on an animation with an undersea scenario.
> I couldn't get the desired water effects with Scanline mode, I went to
> Trace. Now, it is taking over 9 HOURS per frame to render! I was
> just wondering if anybody had any experience with Lightwave versus Imagine
> for Trace renderings and if Lightwave were substantially faster or not?
LightWave has never had much effort put into its tracer and while its not
a complete pig, it is probably not as fast as Imagine when your entire
screen is filled with traced surfaces (as in your case). However, the big
difference is that LightWave's scanline renderer is powerful enough that it
can easily accomplish many tasks that Imagine must trace through. Your
senario, for example, can easily be accomplished within LW entirely in scanline
mode and look quite convincing. SeaQuest is a testament to that. Based upon
the level of detail, texturing, and water effects you describe, I would venture
to guess that you could achieve about a 10x speedup in LW over Imagine for
the scenes you are working on.
%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~%
% ` ' Mark Thompson CONCURRENT COMPUTER %
% --==* RADIANT *==-- mark@westford.ccur.com Principal Graphics %
% ' Image ` ...!uunet!masscomp!mark Hardware Architect %
% Productions (508)392-2480 (603)424-1829 & General Nuisance %
% %
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
##
Subject: Imagine3.0 Demo on Aminet
Date: Mon, 15 Nov 1993 17:21:09 -0500 (EST)
From: "David A. Rollins" <drollin@seq.cms.uncwil.edu>
There is a demo of Imagine V3.0 on aminet in /pub/aminet/biz/demo
##
Subject: RE: imagine on ftp
Date: Mon, 15 Nov 93 15:50:00 PST
From: Stethem Ted 5721 <TedS@ms70.nuwes.sea06.navy.mil>
So, Imagine 2.0 is now Public Domain!!?! BTW, the Imagine on ftp aminet
has the title of Imagine 3.0! Haven't checked it out but will do so
tonight. Maybe whoever uploaded it got the version number wrong. We shall
see ....
----------
From: imagine-relay
To: "imagine@email.sp.paramax.com"
Subject: imagine on ftp
Date: Monday, November 15, 1993 7:08PM
Imagine2.0 is now available on ftp aminet. It can be found in
aminet/biz/demo. I didn't try it, but I think AmigaFormat's cover disk is
now on PD.
Stanis
##
Subject: render times and backgrounds
Date: 15 Nov 93 03:35:00 -0800
From: Ed_Totman@ucsdlibrary.ucsd.edu
Is there a way to merge a single background frame with
parts of multiple frames to produce an animation? What I would
like to do, to save render time, is render one frame as a
background and then render only objects that move in all other
frames, then merge pics. Is there a way to merge only the part
of a pic (the part that contains the object that moves) ?
##
Subject: Re: Trace time vs scanline
Date: Mon, 15 Nov 93 18:45:12 CST
From: setzer@comm.mot.com (Thomas Setzer)
One suggestion to try is to reduce the number of points(edges and faces too) in
your object. There are some programs on the Amiga that will do this for you
and try to maintain the look of the object.
Pixel 3d and Interchange Plus, only on the Amiga AFAIK, will do this for you.
There may be others.
Tom Setzer
setzer@ssd.comm.mot.com
"And of course, I'm a genius, so people are naturally drawn to my fiery
intellect. Their admiration overwhelms their envy!" - Calvin
##
Subject: Happy, happy. Joy, joy.
Date: Mon, 15 Nov 93 20:48:51 EST
From: Steve J. Lombardi <stlombo@eos.acm.rpi.edu>
2.9 for the Amiga arrived today. Here's my list of first impressions.
bear in mind while reading this that I've only had the software a few hours.
This is not meant to start a flame fest. Instead I would like to here
what other 2.9 users think.
GOOD STUFF
----------
1. User interface cleaned up. no more shadowed menus. more logical
placement of stuff in requestors.
2. Lots of included textures. haven't tried them out.
3. Detail editor deformations are real nice. interactive or numeric
modes. taper, bend, pinch,shear, stretch aand twist. THese are
fun to play with and work well.
4. real time window in detail and stage. FINALLY. a pleasure to use.
5. Object fracture. in 2.0 you had to fracture segments or fa ces. now
entire objects can be fractured. simple and cool. kind of like the explode
f/x, but in the detail editor.
6. PARTICLE SYSTEM. There's been a lot of talk of particles in this list
recently. never having used a particle system, and only understanding the
basic concept behind them, I was unsure what all the fuss was about. Now
I see. This is the bifggest enhancement in 2.9. The particle f/x requestor
fills 2/3rds of the screen and will probably take me until 3.0 arrives
to master, but it seems like time well spent.
7. STATES. although the cycle editor is still in 2.9 (perhaps only
for compatability), I think it may be on it's way out. States let
you save many poses in one object (like a cycle) but all of the work is
done in the detail editor. The manual mentions that 3.0's bones will be
tightly tied to the states feature. States is also the heart of brush
and texture tacking.
8. LAYERS. as you load objects into the stagethey can be assigned to layers.
then individual layers can be turned on or off to reduce clutter in the
work area. THis is a great convenience.
BAD STUFF
---------
1. No support for 3.0's display database. This sucks. come on Impulse.
Please support the features that the worlds best OS gives us. A standard
file requestor is also needed. Yours still does not sort the list. This
is silly.
2. In the stage editor, when the Camera is tracked to an object, the
alignment (tracking) is not updated when you move the camera. you must
complete your move, hit the space bar and finally A-k to retrack. This
all but defeats the beauty of a real time window. This is at the top of
my wish list for 3.0
3. an 040 optimized version would be most welcome.
4. OPALVISION BUG. when Opalvision is selected in preferences, the machine will
lock up upon attempting to display to the card. This only applies to
the Amiga 4000 with Opal card. Most other software I've used with Opal
has had this bug, and is easily fixed.
5.I was hoping for sound support. I guess this will be in 3.0.
6. Essence is no longer compatable. Hey APEX- please upgrade to
3.0 compatability. Imagine without essence is like Bert without
Lonnie (or Ernie). WE WANT ESSENCE!! Need in fact. Yes, I'm begging.
As I learn more I'll post it here. BTW- They claim 3.0 should be out
with a manual by years end. Also, the mini manual that comes with 2.9 mentions
that for 100 dollars or so, 4 updates a year will be availible to those
who want them. THis is great, I only wish they had adopted this policy
sooner.
|
steve lombardi | She won't get out of the tub. She
stlombo@acm.rpi.edu | has morning wood. -- Beavis
##
Subject: Re: Trace time vs scanline
Date: Mon, 15 Nov 93 18:49:12 PST
From: jwalkup@sfsuvax1.sfsu.edu (Jeff Walkup)
Stethem Ted writes:
> I guess my question is, if this is a typical time for Trace rendering
> based on the number and size of the objects I am using and the amount of
> textures being applied?
Yes actually, 9 hours is not unheard of. You didn't mention what
resolution you are rendering at, but I've gotten 10 hour times on my
A3000 when using Trace w/shadows and the whole screen almost filled with
a transparent & bump-mapped object. (At 736x480 rez.)
> Other than that, is there anything obvious or simple I could be doing
> to reduce my rendering time?
You could try any of these things:
Reduce the Index of Refraction on your objects - the higher, the longer
trace times will be.
Reduce the "Resolve Depth" (the number of recursive reflections). This
is in the Config editor.
Reduce the antialiasing level, also in the Config. It defaults to 30,
but you might be able to get away with a smaller value, especially if
you are only rendering in 640x400 (736x480 etc) or so.
Alternately, or in addition, you could reduce the actual number of
reflective/refractive objects in your scene. And you didn't mention
whether you're using shadowed lights, but you could reduce or eliminate
those if so.
Lastly (or maybe firstly...) as someone else mentioned - reduce the
polygon counts of your objects.
All of these things will affect the look of your images, but if the
render times are too large you will have to make decisions on where to
sacrifice. The two things that might affect your images the least, or
not at all, depending, are Antialiasing and Resolve Depth. I suggest
you start with those if you can't reduce the detail (polys) of your
objects.
--
Jeff Walkup - jwalkup@sfsuvax1.sfsu.edu - Digital Animator / Videographer
##
Subject: Re: Imagine3.0 Demo on Aminet
Date: Mon, 15 Nov 93 18:59:06 PST
From: jwalkup@sfsuvax1.sfsu.edu (Jeff Walkup)
David A. Rollins writes:
>
> There is a demo of Imagine V3.0 on aminet in /pub/aminet/biz/demo
I looked at disk 1 (Imagine3-A.lha), and did not find a demo of 3.0, but
instead what looks like the regular Imagine 2.0 INT version. It didn't
work either - complained that I had the wrong "NTSC/PAL" version, and
promptly GURU'd my machine. Thank gopod I wasn't rendering...
In other words - BEWARE. I think, as someone on Usenet pointed out
already, that this is a hacked version of 2.0, possibly illegal, and
maybe virus-infected for all I know.
--
Jeff Walkup - jwalkup@sfsuvax1.sfsu.edu - Digital Animator / Videographer
##
Subject: Re: Imagine 2.9 - It's HERE!
Date: Mon, 15 Nov 93 22:24:17 -0500
From: mbc@po.CWRU.Edu (Michael B. Comet)
Well, here are some notes for people:
0] There is a new opening screen with a new rendering etc...
1] I do not think the 640x400 has been addressed....but I haven't
totally looked into it.
2] The new 3D perspective thing is WAY cool. This alone makes the
program a pleasure to use now....here are some more details on it:
- The perspective window no longer has those 2 blocks/bars
- You can use the mouse to interactively rotate, the view,
zoom in/out of the view, and change the perspective
distortion of the view (FOV) in real time with bounding
box reps.
- There is now a great "reset" feature so you can get the perp
view back to default, (it is now very easy to get messed
up as there is no way to tell what angle you are look at)
- All of this works in wireframe, hidden, shaded and new mode.
- There is a new "New mode" This in itslef is pretty slick:
- New mode draws items in wireframe BUT...draws them
in the purple of orange blue black etc...color based
on the pick select stuff etc....
- Allows you to move and edit things in the PERSPECTIVE
view.. ie: You can Click on an object in the
perp view and move it around and see it as a bound-
ing box etc.....very slick!
3] Quick Edges: This is similar to marking edges for sharp/soft etc..
Some of the problems with wireframe is: A] They show all
those nasty triangles where all you need is certain edges, and B]
They slow down the draw time.
With this, you mark certain edges to be drawn in window,
then when
you do a "Quickdraw Pick/ALL" it will ask you bounding box
or edges....if you choose the later, only the edges that are
selected as Quick Edges are drawn.
4] States: Haven't gotten totally into this yet, but it is how
they are going to implement bones, and also brushmap tacking and
help to fix the cycle editor.
- I don't think this affects the cycle editor...you can still
use it if you want.
- It saves all the poses in 1 file which you can
edit from the lovely Detail editor.
- It lets you do brush tacking and morphin by the 1 file/object
5] Slick Stage/Animation stuff:
There really is too much but I'll try:
- Quick edges works in stage view as well
- You can make a preview animation by:
- Full objects...ie: normal
- Quick edges: ie: save time and look only at
whatever quick edges are defined for the object
- Bounding box: Use bounding boxes for all objects
These 3 optional alone will save so much
TOOOOONS of time those old damn previews
where you had to insert
t emporary objects rep days are gone!
- Camera/Light lines....
- You can now see dotted lines in
Quad view (once again
including perspective view if you
are in "new mode" where
the bounds for the light and camera
view are shown.
There is also a wireframe rep of a
camera in the "new mode"
perspecive view (ala Toaster?!)
- There is a new layers thing which i haven't gotten to yet but
sounds very slick.
- Here is a wierd one....what do you think moving the perp view
as desribed now by the mouse stuff above
does when you are in
view camera mode in the Stage editor:
Answer: In normal perp view it works like normal, but
in Camera View mode, you can move
the camera and rotate the scene around the point of
view using the mouse in real time in the perp
window. Very nice
- Also with the "new mode" perp view, when you
position objects around each other for animations, the
object is drawn in real time as a bounding box in 3D so you can
tell how it will be placed relative to other objects in the
perp view.
You can actually place objects and position the
camera and angle without using track objects!
6] I did not notice any changes in the forms or cycle editor,
however, in the forms editor it seems you CANNOT use
the "new mode" for editing. :(
7] The spline editor seems nice....if i could figure it out!
Oh well...enough for now....suffice it to say there are MANY new
things and this is not even the full release with bones yet!
I also messed with particles....made a particle sphere. It looks
likeit figure out particle based on the number of faces or something.
Dunno,I haven't read all of the "pre-manual" yet so i'll probably get to it
sooner or later.
Sorry for all thje tyopes and formatting erros but it's late!
--
+------------------------------------------------------------------------+
| Michael Comet, mbc@po.CWRU.Edu, CWRU Software Engineer/Graphics Artist |
+------------------------------------------------------------------------+
##
Subject: Re: CD32 and Imagine
Date: Mon, 15 Nov 93 21:26:14 PST
From: DonD@cup.portal.com
>With a sparkle in his eyes, Jeff Wahaus writes...
>
>> Wouldn't it be neat if you could hook up a CD32 to your amiga and
>get an A1200 [with some Xtra memory and an 030] and wait for the CD32 add-on
>module promised from Commodore.
>Carmen Rizzolo - Crazed Artist
>CarmenR@cup.portal.com
>
WARNING... CBM has already stated that the CD32 option for the A1200
will occupy the "trap door" so kiss your 030 goodbye when you add
that CD32 add-on. If you want a CD32 buy one, don't bother with the
A1200 add-on.
Don DeCosta
DonD@cup.portal.com
##
Subject: Re: Imagine 2.9 - It's HERE!
Date: Mon, 15 Nov 93 21:48:32 PST
From: jwalkup@sfsuvax1.sfsu.edu (Jeff Walkup)
The most important question I have is: does it automatically smooth the
keyframes like LightWave???
And can you move between keyframes for an object easily? Heck, can you
even move forward & back thru the frames without having to wait for it
to re-load the objects each time?
--
Jeff Walkup - jwalkup@sfsuvax1.sfsu.edu - Digital Animator / Videographer
##
Subject: Re: New Speed
Date: Tue, 16 Nov 1993 20:00:59 +1100 (EST)
From: Nikola Vukovljak <nvukovlj@ucc.su.OZ.AU>
This is something Rick Rodriguez and I have been talking about via E-mail
concerning the speed of Imagine & LW. (and the A4000/040)
He thought that it might be interesting for the rest of the IML.
So did I so here goes.
On 15 Nov 1993, Rick Rodriguez wrote:
> 2.9 should be winging its way to you any moment. Let me know what
> you think. I hate to disappoint you on this, but I have empirical
> data which supports my contentions regarding speed.
>
> Yesterday, after my post to you, I created a simple scene which could
> be duplicated on both Imagine and LW. It consisted of one light, my
> nighttable object from the VRS Media Master Designers Objects Series
> Volume 1 Bed & Bath. This object may also be found in Syndesis'
> 3D-ROM.
>
> The nighttable is mapped completely with an IFF wood brush and has
> a nice combination of flat surfaces and complex curves.
>
> I filled the frame with object against a black background.
>
> On a 4000, the image took 6 minutes in LW with rendering set to 752x480,
> antialiasing set to low, adaptive sampling, 16. In Imagine, the image
> took just under 5 minutes with edge level set to 8.
> On a 2000, the Imagine image took 2:40. I couldn't test LW on the
> 2000, but I assume the render would be around 3 minutes.
>
> I then traced the scene. On the 4000, the Imagine image took 8:08.
> Lightwave took 13:03! On the 2000 (which has a Progressive 040, 28 Mhz
> and 16 megs of RAM), the Imagine picture took 4:52!
>
> I shadow-mapped the Lightwave scene and re-rendered and the image took
> about 6-1/2 minutes. Unfortunately, I couldn't get a usable shadow
> map. I tried scaling the map all the way to 2048, but kept getting
> a serrated edge on the shadow. Blurring it helped somewhat.
>
> For grins, I tested the scene in Imagine PC on a 486/66. The raytrace
> took under 2 minutes! The scanline was 1:01.
>
> It's my conclusion that folks who talk about increased speed are
> talking apples and oranges. Since I've had the "luxury" of building
> duplicate scenes in Imagine, Lightwave and 3D Studio, I can speak very
> matter-of-factly about rendering speed.
>
> Clearly, a lot of the talk of LW's speed is based on folks doing low-res
> previews with zero antialiasing. These do seem perky. However, lowering
> the res and eliminating antialiasing gives Imagine a speed burst, too.
>
> As for professional applications, LW is an excellent package because
> Alan Hastings has been very responsive to the needs of producers and
> has implemented features as they have been requested. Unfortunately,
> Impulse has preferred to implement more esoteric features.
>
> Thank you for the opportunity to set the record straight. Please do me
> a favor and repost this message to the entire list.
>
> --Rick Rodriguez
> VRS Media
>
>
##
Subject: Archive#43
Date: Tue, 16 Nov 1993 20:33:36 +1100 (EST)
From: Nikola Vukovljak <nvukovlj@ucc.su.OZ.AU>
I'm sure that I already posted this but can't remember it appearing on
the list.
Anyway, Archive #43 of the IML has been uploaded onto Aminet since I
couldn't upload it to the usual spot on wuarchive. I'll upload it there
when waurchive lets me.
The archive is in gfx/3d directory of aminet. File arc-43.lha.
The archive covers messages from Oct 1 - 31.
Nik.
(Your friendly archivist)
##
Subject: So called Imagine 3.0 demo
Date: Tue, 16 Nov 1993 05:33 EDT
From: SPICE@DRYCAS.CLUB.CC.CMU.EDU
The so called Imagine 3.0 demo is nothing more than a hacked/cracked
version of the Imagine 2.0 that appeared on the Amiga Format coverdisk.
If you've downloaded it please erase it and if you have uploaded to
any BBS or FTP site tell the sysadmin to erase it. Impulse gives away
Imagine 2.0 for $10 and the Euro Pirates still steal it and have the
gall to upload it to a FTP site. Impulse is crazy if they think the
Euro Pirates are going to pay $100 to upgrade to Imagine 3,0. All
they've done with this coverdisk giveaway is alienate people who
are likely to actually buy the upgrade in favor of people who wont
buy it at any cost. Smart move Impulse.
Scott Corley
spice@drycas.club.cc.cmu.edu
##
Subject: Imagine 3.0 demo
Date: Tue, 16 Nov 1993 06:17:01 -0700
From: "Jeff Wahaus, CAPS, ATL, 404-640-3529" <JEFF_W1@sfov1.verifone.com>
> The so called Imagine 3.0 demo is nothing more than a hacked/cracked
> version of the Imagine 2.0 that appeared on the Amiga Format coverdisk.
Why would anyone need to crack Imagine 2.0? It was not copy protected in
any way. In what way is the file hacked?
I got the "Imagine 3.0 demo" and could not de-arc it. All of the files
in the archive seemed to not have a filename. I guess I need a newer
version of LHA. I think I'm using v1.16.
Has anyone checked this file for viruses?
-Jeff Wahaus-
jeff_w1@verifone.com
##
Subject: Re: Imagine 3.0 demo
Date: Tue, 16 Nov 93 8:17:47 PST
From: jwalkup@sfsuvax1.sfsu.edu (Jeff Walkup)
> Has anyone checked this file for viruses?
Well, I erased it, so I can't tell you if there's any virii in it. But
I checked my system afterwards with Virus Checker 6.32, and it didn't
find anything.
--
Jeff Walkup - jwalkup@sfsuvax1.sfsu.edu - Digital Animator / Videographer
##
Subject: Re: Imagine 2.9 - It's HERE!
Date: Tue, 16 Nov 93 12:05:07 -0500
From: mbc@po.CWRU.Edu (Michael B. Comet)
>
>You mention editing in Detail editor in perspective. Can you actually
>move points and faces or not ? This would be way cool if possible.
>
>Nik.
>
>
>
>
Yes, it draws the points faces etc in perspective, and you can
select and move them in the perspective window with clicking, bounding box
and lasso etc...
Pretty neat.
--
+------------------------------------------------------------------------+
| Michael Comet, mbc@po.CWRU.Edu, CWRU Software Engineer/Graphics Artist |
+------------------------------------------------------------------------+
##
Subject: More on 2.9
Date: Tue, 16 Nov 93 12:21:11 -0500
From: mbc@po.CWRU.Edu (Michael B. Comet)
I had more time to play around with 2.9 here are some more goodies:
- States: Basically this makes the cycle editor obsolete, though
they said they will keep it in the program since some
people have things like cycle man etc...
- All you do is make an object or group of objects in a heirarchy
and in any configuration settings etc you want.
Then you create a state. Now you can manipulate, mash and
mutilate your object and also move the groups arounds etcc.
then you make another state.
Finally you just morph the object by state names. Thus, one
object can hold all the morph positions etc...1 file.
Better yet, the texture and brush map requestors have a slot
where you enter the "lock state". This is the state where
it figures how to tack the brushmap and texture to the
object.
- so, to tack a brushmap, you just apply it, make a state, and tell
it to lock to that state, then you can make other states
etc....and morph them and the brushmap/texture will follow.
- Also, there is no longer any problems with having to worry about
axis size and placement (ie: cycle setup) etc...if you use
states.
Particles: Here is what I think i have figured out about it. I
may be wrong of course:
- There are 2 particle "Things" in imagine. 1 are particle objects
where when you render your object each face is rendered as
a small block, pyramid, dodecohedran etc..... and the
other is particle motion FX.
- The object part basically renders an object so that where each
face would be there is now one of these nifty objects that
you specify. Then can tell it how to align etc..etc...
- The motion will work on any object...is breaks up each face and
manipulates it etc...thus you can render particle motion
for particle objects, or normal objects whose faces just
fly all over the place.
- I managed to make a particle anim where this sphere explodes and
graviates to the ground plus bounces off the ground until
everything comes to rest.
- There is also a wind thing which blows particles etc..etc...
TONS of options etc...
Gotta go!
--
+------------------------------------------------------------------------+
| Michael Comet, mbc@po.CWRU.Edu, CWRU Software Engineer/Graphics Artist |
+------------------------------------------------------------------------+
##
Subject: Re: Imagine3.0 Demo on Aminet
Date: Tue, 16 Nov 1993 12:29:53 -0600 (CST)
From: Trin Yuthasastrackosol <yuthas@cc.umanitoba.ca>
On Mon, 15 Nov 1993, Jeff Walkup wrote:
> David A. Rollins writes:
> >
> > There is a demo of Imagine V3.0 on aminet in /pub/aminet/biz/demo
>
> I looked at disk 1 (Imagine3-A.lha), and did not find a demo of 3.0, but
> instead what looks like the regular Imagine 2.0 INT version. It didn't
> work either - complained that I had the wrong "NTSC/PAL" version, and
> promptly GURU'd my machine. Thank gopod I wasn't rendering...
Yah..I get that problem until I switch to PAL mode, then it runs fine, except
that about 20% of the screen is missing at the bottom. Easy to fix tho',
just change the bytes in the program which specify the vertical
resolution from 256 to 200 and it will open with the whole screen.
Stil have to run it in PAL unfortunately.
Anyways, I don't think Imagine2.0 should have been uploaded to aminet
. Has anyone asked Impulse about this?
Expecting 2.9 anyday now :-).
___
Trin Dominic Yuthasastrakosol | Definition: Bigot (noun). Someone
Dept. Pharmacology and Therapeutics | who zealously holds a set of ideas,
University of Manitoba | beliefs or opinions which are
A5002MbChip1MbFastSupraturbo28GVPII354Mb| different from yours.
##
Subject: morphing
Date: Tue, 16 Nov 1993 11:14:18 -0700 (MST)
From: LESK@CC.SNOW.EDU
Hello all;
It seems I have failed to grasp the morphing everyone is
talking about. I have a logo I am trying to morph into a drill bit,
I created the bit in vertex and made sure aces etc. were all the same count
as the logo. In the cycle editor the logo change postion but not shape.
So I tried a box in the detail editor saved it, then distorted the box by
moving points here and there stretching it out of shape then saving that.
now in cycle editor I loaded the box then went to frame 15 then loaded
(I think it was position or picture?) but again the box moved its location
on the screen but no distortion of the box itself. Is there an actual
morph menu or selector I am missing somewhere?
Thanks All
Lesk
##
Subject: RE: render times and backgrounds
Date: Tue, 16 Nov 93 10:58:42 PST
From: 16-Nov-1993 0514 <leimberger@marbls.enet.dec.com>
> Is there a way to merge a single background frame with
> parts of multiple frames to produce an animation? What I would
> like to do, to save render time, is render one frame as a
> background and then render only objects that move in all other
> frames, then merge pics. Is there a way to merge only the part
> of a pic (the part that contains the object that moves) ?
I did this once. I had rendered Steve Worleys fire tutorial and wanted
to use it as a background, So what I did was I made a cycleman anim,
and I set the reflective map to a frame of the fire anim, and also
the background. Then I went into the action editor, and incremented
the fire pic frames in the globels. I ended up with a chrome cycle
man that when he moved had all the right reflections. I did it manually,
but ISL or such should allow you to do it with less manual overhead.
The completed anim was quite nice. This has it's limits because you
are using both the reflected, and background in the action editor, so
their not available for oher effects. However it served me well for
this anim. Pretty crude but then it was just an experiment.
bill
##
Subject: Could you upload 2.9 particle animations?
Date: Tue, 16 Nov 1993 13:45:00 -0600 (CST)
From: DCG9367@tntech.edu
Michael Comet said:
> - The motion will work on any object...is breaks up each face and
> manipulates it etc...thus you can render particle motion
> for particle objects, or normal objects whose faces just
> fly all over the place.
>
> - I managed to make a particle anim where this sphere explodes and
> graviates to the ground plus bounces off the ground until
> everything comes to rest.
>
> - There is also a wind thing which blows particles etc..etc...
Could you post some of these animations for us who don't have 2.9 yet?
I would really like to see some of what 2.9 can do before I upgrade.
Thanks
##
Date: Wed, 17 Nov 1993 12:30:49 +1100
From: imagine-relay@email.sp.paramax.com
> Doew anybody know what is the DPS PAR system. I know that it has to do
> with real time animation. I also want the address and the phone/fax numbers
> for that company.
>
Yep. Real time 24 bit full screen animation playback. CAn also be used
with the DPS TBC 4 for real time capture of video ie: nice for rotoscoping
too though i don't have a tbc 4.
Anyhow, I am using the product with Imagine, Vista Pro 2 etc.... Unbelievable.
BTW, if you don't have a recent copy of ADPro 2.3.0 or whatever you may have
some little problems. I have 2.1.5 and sometimes the machine crashes. I have
found out that it is a problem with ADPro and NOT the PAR....upgrading adpro
will fix the problem.
The interface is great, and easy to use. I am quite happy with the quality
SVHS AND VHS is all I use so I can't vouch for the Betacam output....though
I've heard it's really good.
If you want to do 24 bit anims and have the money, this is a must for any
animator. No more drop outs etc...
The best thing is you can render different anims/scenes and can then edit them
how you want.
Also hold about 5,000-10000 still frames too.
Here is the address:
DPS - Digital Processing Systems
11 Spiral Drive, Suite 10
Florence, KY 41042
Phone: (606) 371 5533
Fax: (606) 371 3729
Mike C.
PS: I am not affiliated with DPS, I'm just a happy customer.
##
Subject: Re: IMAGINE PC
Date: Sat, 16 Oct 93 20:52:12 MDT
From: pringleg@cuugnet.cuug.ab.ca (Greg Pringle)
Thanks for all the great info re: imagine PC, everyone! I
really appreciate everyone resisting the urge to launch into
IBM vs. Amiga bashing. I've been tracking down the utility programs
mentioned, and after I get a chance to test them out, I'll try
and post a summary of what is available where.
Greg
--
+------------------------------+---------------------------------------+
| Greg Pringle | Amiga VBBS - Multitasking, Windowed |
| pringle@cpsc.ucalgary.ca | BBS'ing! |
| pringleg@cuugnet.cuug.ab.ca | VBBS 14.4K: (403) 284-2048 & 284-5625 |
+------------------------------+---------------------------------------+
##
Date: Wed, 17 Nov 1993 15:41:44 +1100
From: imagine-relay@email.sp.paramax.com
> Hello all;
> It seems I have failed to grasp the morphing everyone is
> talking about. I have a logo I am trying to morph into a drill bit,
> I created the bit in vertex and made sure aces etc. were all the same count
> as the logo. In the cycle editor the logo change postion but not shape.
> So I tried a box in the detail editor saved it, then distorted the box by
> moving points here and there stretching it out of shape then saving that.
> now in cycle editor I loaded the box then went to frame 15 then loaded
> (I think it was position or picture?) but again the box moved its location
> on the screen but no distortion of the box itself. Is there an actual
> morph menu or selector I am missing somewhere?
>
> Thanks All
> Lesk
Morphing is done in the Stage and Action editors, NOT the cycle editor,
though you can morph Cycle objects.
I am pretty sure there is a morph tutorial in the manuals.
All you have to do is to add an object, and then add to its actor timeline
in the Action editor to tell it the morph target, and then set the transition
frame count in the same requestor.
Mike C.
mbc@po.CWRU.Edu
##
Subject: Imagine Screen Sizes
Date: Sat, 23 Oct 1993 05:44:07 -0700
From: "Jeff Wahaus, CAPS, ATL, 404-640-3529" <JEFF_W1@sfov1.verifone.com>
> True, the rendered sizes can be set from the Imagine.config file, but I'd
> be happy for you to show me which numbers to change to swap the interfaces
> between PAL/NTSC screen sizes. I'm pretty sure it can't be done so easily.
>
> AFAIR, the NTSC version will use screen sizes in the region of 640 x 480
> for the interfaces, whilst PAL screens are 640 x 512 (at least the ones I
> grabbed with GrabIFF are...).
>
> Gary
Actually the screen size for Imagine can be easily changed. All you
have to do is find a good HEX editor and patch the executable. The
bytes to change were posted here not long ago.
-Jeff Wahaus-
##
Date: Wed, 17 Nov 1993 15:47:14 +1100
From: imagine-relay@email.sp.paramax.com
You don't morph in the Cycle editor, you do it in the Action editor. The
first tutorial in the 1.1 manual (if you have it) is a nice morph of a
sphere into a cube.
Basically, add your first actor in the first frame. Then add the second
actor ON THE SAME ACTOR LINE, and specify how many transition frames you
want the morph to take. Save your work. Go into Stage, and look at one of
the transition frames. Should be what you're looking for.
BTW, point count ain't all you have to match. All the points, edges and
faces also have to be drawn in the same ORDER, or no go. That's why it's
usually best to start with your most complex object, mash it into a
plane/sphere/whatever primitive, and morph from that. If you have to
morph a complex shape from a complex shape, try using a sphere as an
intermediate object, and just swapping spheres in a single frame, e.g.:
Object1------->sphere1
sphere2--------Object2
Where each sphere has the same color/brushmap/texture, but point count
matching the respective Object. Other primitives work well, too. A plane
magnetically drawn up into an amorphous blob is good for T2 effects.
------------------------------------------------------------------------------
Doug Kelly Information Specialist First Consulting Group
dakelly@class.org (310)595-5291x125 P.O.Box 5161, Los Alamitos,CA 90721-5161
"The difference between genius and stupidity: genius has its limits."
------------------------------------------------------------------------------
##
Date: Wed, 17 Nov 1993 16:05:15 +1100
From: imagine-relay@email.sp.paramax.com
> From: Rick Rodriguez
>To: TedS
>Subject: Re: Trace time vs scanline
>Date: Monday, November 15, 1993 10:20PM
>
>Nine hours seems like a long time. Maybe you can reduce index of
>refraction. That lengthens rendering considerably. Also play with
>your edge level in the Preferences editor. Make sure it's no lower
>than 8. You might also lower the number of iterations if you don't
>have a lot of reflectivity. I've gotten away comfortably with 3
>recursions.
>
>Of course, the best bet is an '040 with lots of 32-bit RAM! Or better
>still, a 486/66 or Pentium (mine arrives tomorrow).
>--Rick Rodriguez
__________________________________________________________
Yes, I have received several suggestions to do these things. So far, no
decrease in rendering time. I haven't tried less anti-aliasing yet or
investigated change in size in Stage Editor and resulting differences in
rendering time.
I do have an '040 with 32-bit RAM but maybe not enough. Of course, based
on a past speed comparison between '040/33MHz and 486/66MHz, the rendering
time could be half what it is with a 486/66 MHz! And then there is those
privileged few that will be getting a Pentium system! I am glad to hear
that you will be leaving me behind in the dust |:^(
##
Subject: trace rendering
Date: Tue, 16 Nov 1993 19:38:12 -0500 (EST)
From: galandt@cgrg.ohio-state.edu (Gigi Alandt)
I'm having a problem rendering a project. Actually, I can get the system
to render I just can't get any of my objects to show up. I've got
reflective glass surfaces so I definately want to use trace (I can get the
images to show up if I'm in scanline) I set my global size at 0 and I
scaled down all my elements to as small as I thought was necessary, and I'm
talking pretty darn small!! still no luck....
My presets are hires, ILBM 12bit. For some odd reason stereo 3D wont
turn off. I don't know if this has anything to do with why I can't get my
images to render.
Does anybody have any advice??
I'd appreciate your help thanks.
##
Subject: Re: DPS PAR
Date: Tue, 16 Nov 1993 21:13:59 -0500 (EST)
From: kingb@echonyc.com (Andrew McDonald)
For people using a PC, DPS says that a version of PAR will be available in
Q1/94. Still working on the software interface.
Andrew.
##
Subject: 2.9 particle question
Date: Tue, 16 Nov 93 18:30:07 EST
From: Steve J. Lombardi <stlombo@eos.acm.rpi.edu>
Someone here (perhaps Jeff) mentioned they made a rain particle animation. I
am trying the same thing but have a problem setting the particles in motion.
here's what I did:
1. made a 4X4 plane in detail editor. in particle requester I went for small
inscribed spheres. saved it.
2. in stage placed the plane 300 units above z.
3. in action added the particle effect from 1 to 20 frames. checked the
rain box. set ttravel distance to 300. set ground Z coordinate to 0 (the
ground)
Now the wierd part. if I return to the stage at frame 1 the particles are
already scattered from the origin point to the ground. well, ok maybe
that''s normal and in the next 19 frames the rest will follow. However
there is no movement on the next 19 frames. all particles remain
in place. I've tried adding bounce and even altering some of the
parameters that don't seem to apply to what I was doing. No difference.
What am I missing here?? anyone?
|
steve lombardi | She won't get out of the tub. She
stlombo@acm.rpi.edu | has morning wood. -- Beavis
|
##
Subject: No Demo From Impulse
Date: Tue, 16 Nov 1993 22:29:45 -0500 (EST)
From: MCA94ISA09@RCNVMS.RCN.MASS.EDU
Greetings All :)
I spoke with Arv at Impulse this morning to check if the "demo" was in any way
sponsored or endorsed by Impulse. Surprise, surprise, they don't. I
downloaded it the night it came out, and checked the files for viruii. The
files are clean, but they aren't the real thing either. Impulse says that the
only quasi-PD release they have made is the Amiga Format cover disk.
I also would LOVE to see the particle animation(s) uploaded for FTP. I don't
have 2.9 (yet), and am curious to see what the new version is capable of.
-Bob
Robert J. O'Connell Producer/Videographer/Amigan/Imagineer
The Massachusetts College of Art / Studio for Interrelated Media
##
Subject: re: 2.9 pro/con
Date: Tue, 16 Nov 1993 21:03:22 -0500 (EST)
From: kingb@echonyc.com (Andrew McDonald)
> 2.9 for the Amiga arrived today. Here's my list of first impressions.
>
> 2. In the stage editor, when the Camera is tracked to an object, the
> alignment (tracking) is not updated when you move the camera. you must
> complete your move, hit the space bar and finally A-k to retrack. This
> all but defeats the beauty of a real time window. This is at the top of
> my wish list for 3.0
>
I have been using 2.9 furiously since last Tuesday night, (ack! get a
life), depending on the animation, a camera track object may not be
necessary. The real-time window in the stage editor allows you to track
the camera right there, just use the "A"ngle and "Z"oom buttons to move
the camera in and out of the scene and rotate.
The real-time window will also handle camera rotations performed in other
windows using the usual x,y,z axis' stuff, and camera lines combined with
the real time preview allow much more precision of camera field of view
(scale on x and y).
RE: Essence Textures
Seems Impulse took a good look at the Essence textures and incorporated
most of the ideas into the new texture set included with v2.9. While some
of the Apex implementations of fractal noise are better, the new Impulse
textures go far beyond the old collection. I can get a pretty good grass
out of one of the bump textures, and coolfir is way cool. Also, crumpled,
if done right, will give that "underwater in a swimming pool" look.
I'm a little disappointed with the limitations of particles, but for the
money Imagine 2.9 gives me a good start. Not too many desktop 3D programs
have done much in this area besides the IPAS routines for 3D Studio. I
really haven't used the 3DS particles, so can't make any useful
comparisons. Note, I'm not sure if my disappointment with particles is
with Imagine, or the particle concept. As it stands, the particle f/x is
just another exploding triangle effect, while the detail editor particle
objects can be very cool. Unfortuneately, there is no way to easily morph
an object exploding into nice cubed particles since I don't have the math
background to compute the increase in points/edges/faces when an object is
made particle.
Whew. Sorry to use up so much bandwidth.
Andrew.
##
Subject: Re: Could you upload 2.9 particle animations?
Date: 17 Nov 1993 03:19:17 GMT
From: <mbc@po.cwru.edu>
>
> Could you post some of these animations for us who don't have 2.9 yet?
> I would really like to see some of what 2.9 can do before I upgrade.
>
Sure....if anyone can point to me to a site where I can upload some
JPEG stills and anims, i will do so. I wonder if wuarchive is still down.
Anyone know of ANOTHER site?
Mike C.
mbc@po.CWRU.Edu
##
Subject: Imagine 3.0, uhhh...
Date: Tue, 16 Nov 1993 22:47:00 -0500
From: charles.blaquiere@canrem.com (Charles Blaquiere)
...well, not quite. (WARNING: this message is one big spoiler! If you don't want
to spoil the surprise when you receive Imagine, don't read this!)
What I got was Imagine 2.9. Yep, Impulse has yet some more delivery problems. Not
all bugs have been squashed, so they have released an interim version whimsically
called "2.9 -- in construction". There's no manual (that'll be included in the
no-charge "real" 3.0 upgrade), but Mike Halvorson did include a 25-page
description of the new features, written in Mike's inimitable style.
Here's what DIDN'T make it:
BONES
KINEMATICS
IMPROVED ANTI-ALIASING
NEW LIGHT SOURCES WITH IMPROVED SHADOWS
SHADOW MAPPING
NEW STAGE & ANIMATION TOOLS
And here's what DID make it:
DETAIL
------
-QUICK EDGES: Instead of a bounding box in Quickdraw, you pick some
representative edges of your object.
-MOUSE PERSPECTIVE VIEW CONTROL: Change alignment, zoom level and perspective
level by moving the mouse.
-ACTIVE PERSPECTIVE VIEW: Pick and edit anything by clicking on it in the
Perspective view, just like the other 3 views.
-LOAD/SAVE DXF: exchange AutoCAD objects with your friends.
-OBJECT DEFORMATION TOOLS: Twist, shear, taper, pinch, bend, stretch. Numerical
or mouse-interactive with live redraw.
-LATTICIZE: Your object is turned into a latticework where only its edges will
show; adjustable edge width.
-APPLIQUE: You select an IFF map which will be wrapped around your object, and
move its points in and out like an altitude map -- except the points are _really_
displaced. Different from Lightwave since the displacement map is not separate
from the object, i.e. you can't make the map travel across the length of your
swimming shark. Imagine's implementation results in a single deformed object.
-WAVE: Like the Ripple F/X, only in Detail.
-CHECK OBJ: When importing an "Imagine format" object from a 3rd-party program,
you can check its validity to make sure it's fully Imagine-legal.
-STATES: A morphing tool, a superset of the Cycle editor. You create an object
(your "base object"). You then de
form it, change its attributes and/or move its
elements if it's a hierarchy. The new object is called a "state"; for example,
"sitting man", "standing man" and "man with raised hand and big nose" are states;
the third state also incorporated some point deformations. All those states are
saved in the same object file. You can then morph, in the Action editor, from one
state to any other, in any order, revisiting some previous states if you wish. It
blends object morphing and cycle objects in an elegant manner.
-RENDER PALETTE: You can force Imagine to use a locked palette, or that from an
image on disk. Great for PC users who wait forever for Imagine to "cleanup" a
quickrender -- i.e. bring it down from 24-bit to 256 colors for viewing.
-PARTICLE: Will associate a particle object (e.g. a small pyramid -- can be any
Imagine object) with each triangular face of an object. The faces will not
render, just the particles. Particles are animated by applying a "particle" F/X
to the object in the Action editor. The F/X affects the invisible faces (like
"explode", but infinitely more complex), which in turn affects the particles.
More in "Action editor", below.
SPLINE
------
A brand new editor! Select any Postscript font, enter some text and your words
appear in 2-D. You can then extrude and bevel, with a choice of bevel styles, and
save as an Imagine object. Face subgroups will automatically be created, with
names such as "BACK FACE" or "FRONT BEVEL". Heaven. Sheer heaven. With user-
defined detail level, too, for those times when a 10,000-face logo is just too
much. Apart from importing Postscript fonts, you can draw your own (closed)
spline curves, with full spline control, not just the old Imagine "path" control
point rotation. For example, you can drag a control point's handles independently
to make a sharp corner. Again, this is a 2-D editor; its only 3-D capabilities
are straight extruding and bevelling.
ACTION
------
-ASSOCIATE: Forces a "slave" object (e.g. the camera) into a fixed distance and
angle from a "d
river" object (e.g. a rollercoaster car). Much easier than
creating separate motion paths or fiddling around with track objects.
-LAYERS: Each actor lives in a layer (0-99); many actors usually share the same
layer. This lets you turn whole groups on or off in the Stage editor, to speed up
scene adjustment.
-PARTICLE F/X: As described in "Detail editor", a particle object has one small
object assigned to each (invisible) of the object's triangular faces. This F/X
controls face movement, which in turn controls the visible particles.
-REVERSE, GO BACK: Like "explode".
-DELAY motion start depending on particle distance from object axis.
To make your particles start moving, wave-like, at different times.
-RAIN: vertical movement control. Gravity, bounce plane Z position,
elasticity (how strongly particles bounce back)
-WIND: horizontal movement control. Direction, speed, start/end frames.
-PARTICLE SCALING & DELAY: Like "explode".
-MIN/MAX ROTATION, ANGLE FROM X, FROM Z: Random individual variation.
-ACCELERATION FRAMES
-EMISSION AMOUNT: How many particles/frame are ejected from the object,
a la "Deep space 9" meteor intro.
STAGE
-----
-LAYERS: Make any layer(s) invisible to reduce redraw times. As a safeguard, all
layers will automatically be used when _rendering_ the project.
-CAMERA & LIGHT LINES: shows lines beaming out of the camera and spotlight-type
lights, to visualize where their field of vision/influence ends.
-CAMERA VIEW: Adjusting the Perspective view will change the corresponding camera
position and angle. An alternative to adjusting the camera.
-CLONE: Once you've put in all the work needed to define an actor's fancy
timelines, you can create another actor with the same data, then make only those
adjustments needed to differentiate the second actor from the first.
As expected, Imagine feels a whole lot more like Lightwave now. Even if such
niceties as Bones and Kinematics aren't available yet, what _is_ there is more
than enough to play with for many, many moons. I'm sure you have m
any questions,
but for now, it's time I went to sleep.
##
Subject: re: 2.9 pro/con
Date: Wed, 17 Nov 1993 14:21:00 -0500
From: rosario.salfi@canrem.com (Rosario Salfi)
According to Steve Worley, it won't take him long to update the essence
textures to work with Imagine 3.0. I have faith in him and Apex's
ability to keep up with Imagine.
##
Subject: re: 2.9 pro/con
Date: Wed, 17 Nov 93 18:26:01 EST
From: Steve J. Lombardi <stlombo@eos.acm.rpi.edu>
> I have been using 2.9 furiously since last Tuesday night, (ack! get a
> life), depending on the animation, a camera track object may not be
> necessary. The real-time window in the stage editor allows you to track
> the camera right there, just use the "A"ngle and "Z"oom buttons to move
> the camera in and out of the scene and rotate.
>
Yea, but I still like to track to an object and move the object around.
since the real time window doesnt recalculate camera angle it defeats
the purpose here. I called Impulse on this. They have something in the
works for 3.0 to take care of this.
> with Imagine, or the particle concept. As it stands, the particle f/x is
> just another exploding triangle effect, while the detail editor particle
> objects can be very cool. Unfortuneately, there is no way to easily morph
> an object exploding into nice cubed particles since I don't have the math
> background to compute the increase in points/edges/faces when an object is
> made particle.
>
COMBINE THEM!!!! The particle f/x (although it will work with any objects
triangles) is meant to work with those cool detail editor particle objects.
try it. make an object in detail. assign particles to it. save it.
load in action editor. assign the particle f/x and render.
|
steve lombardi | She won't get out of the tub. She
stlombo@acm.rpi.edu | has morning wood. -- Beavis
##
Subject: Re: Sphere primitive doesn't work
Date: Wed, 17 Nov 1993 14:04:27 -0800 (PST)
From: Doug Kelly <dakelly@class.class.org>
The sphere primitive (as in 1.1) is a mathematical construct, not a set of
discrete points. The wireframe you see is just an approximation. Use the
Sphere listed with the other primitives, where you can define the point
count, and you'll be able to edit it.
------------------------------------------------------------------------------
Doug Kelly Information Specialist First Consulting Group
dakelly@class.org (310)595-5291x125 P.O.Box 5161, Los Alamitos,CA 90721-5161
"The difference between genius and stupidity: genius has its limits."
------------------------------------------------------------------------------
##
Subject: Rotation
Date: Wed, 17 Nov 1993 16:13:33 -0600
From: christopher arthur <amadaeus@ccwf.cc.utexas.edu>
Hello...I have Imagine 2.0, and I'm trying to make an animation
of a ball that spins for a bit, then slows down and stops.
I figured the best way to do this was to define a circular path
in the plane intersecting the axis of rotation. The path was centered
with the ball. I set a track object on the path, and then locked the
alignment of the ball to the position of the track object. This way,
I can control the rate of angular acceleration of the ball with the
speed of the track object. It should work fine, except that
when the rotation of the ball exceeds a total of 90 degrees, it flips on
it's z-axis 180 degrees. That is, it flips 180 degrees on an axis
purpendicular to rotation... WHY???? How can I avoid this problem?
Chris Arthur
##
Subject: Sphere primitive doesn't work
Date: Wed, 17 Nov 93 20:11 GMT0
From: Jacek Artymiak <jartymiak@cix.compulink.co.uk>
I've got my 2.9 PAL upgrade a couple of days
ago. I noticed that it impossible to apply
Deformations or different Pick/Add modes to
the sphere primitive (Function/Add/Sphere).
Anyone encountered similar problems?
BTW. There will be a full review/tutorial in
the next issue of bitmap. Sorry couldn't
resist the plug ;-)
-----------------------------------
bitmap: The Amiga Graphics Magazine
-----------------------------------
Jacek
##
Subject: Volcanoes
Date: 17 Nov 1993 16:32:48 U
From: "Shalini Govil" <shalini_govil@maca.sarnoff.com>
Subject: Time:5:42 PM
OFFICE MEMO Volcanoes Date:11/17/93
Hi all..
I wanted to create a volcanoe, so I presume the best way is to give it a
fractal terrain texture? Is this provided in Essence? If so, can somepne please
give me the no of Apex software? Thanks a ton
Shalini
##
Subject: Re: Sphere primitive doesn't work
Date: Wed, 17 Nov 1993 16:12:13 -0600 (CST)
From: Daniel Jr Murrell <djm2@Ra.MsState.Edu>
>
>
>
> I've got my 2.9 PAL upgrade a couple of days
> ago. I noticed that it impossible to apply
> Deformations or different Pick/Add modes to
> the sphere primitive (Function/Add/Sphere).
> Anyone encountered similar problems?
No, you can't deform the sphere primitive. It doesn't have any real
points to deform around. Deformations only work when there is a point to
pivot/twist/whatever around. Try the faceted sphere instead.
A question:
I was going to wait until 3.0 is out to upgrade. Y'all think I ought
to go ahead and do it now? I'm getting really jealous listening to the
stories about the particle systems, deformations, real-time perspective
window updating, etc.
>Jacek
Danimal
##
Subject: 2.9 - Improved keyframe system?
Date: Wed, 17 Nov 93 9:51:02 PST
From: jwalkup@sfsuvax1.sfsu.edu (Jeff Walkup)
Does 2.9 do splined keyframes (automatically smoothing the motion;
ease-in/ease-out etc...) like LW?
Also very important, are there "next/prev key" buttons? Can I move
thru the frames without waiting for the scene to reload from disk?
--
Jeff Walkup - jwalkup@sfsuvax1.sfsu.edu - Digital Animator / Videographer
##
Subject: Re: 2.9 particle question
Date: Wed, 17 Nov 93 12:32:09 -0500
From: mbc@po.CWRU.Edu (Michael B. Comet)
>
>Someone here (perhaps Jeff) mentioned they made a rain particle animation. I
>am trying the same thing but have a problem setting the particles in motion.
>here's what I did:
>
>1. made a 4X4 plane in detail editor. in particle requester I went for small
>inscribed spheres. saved it.
>
>2. in stage placed the plane 300 units above z.
>
>3. in action added the particle effect from 1 to 20 frames. checked the
>rain box. set ttravel distance to 300. set ground Z coordinate to 0 (the
>ground)
>
>Now the wierd part. if I return to the stage at frame 1 the particles are
>already scattered from the origin point to the ground. well, ok maybe
>that''s normal and in the next 19 frames the rest will follow. However
>there is no movement on the next 19 frames. all particles remain
>in place. I've tried adding bounce and even altering some of the
>parameters that don't seem to apply to what I was doing. No difference.
>What am I missing here?? anyone?
>
> |
>steve lombardi | She won't get out of the tub. She
>stlombo@acm.rpi.edu | has morning wood. -- Beavis
> |
>
>
>
>
Try raising your Gravity Constant, and decreasing the Maximum time
to H velocity and Z velocity or whatever it is labeled.
--
+------------------------------------------------------------------------+
| Michael Comet, mbc@po.CWRU.Edu, CWRU Software Engineer/Graphics Artist |
+------------------------------------------------------------------------+
##
Subject: Re: Wine glass from balls
Date: Wed, 17 Nov 93 08:37:20 PST
From: The_Doctor@nesbbx.rain.com (Michel J. Brown)
In <199311170403.UAA27198@mail.netcom.com>, bdady@netcom.com (Brian Dady)
writes:
> Is there an easy way to just join the two spheres at the edge by
> pairing the points?
>
Sure, just set a plane on top of the to hemispheres, then slice the object, an
d delete the unwanted material from the find requester. I have used this
approach, and guess I'm the only person who has no problem with the slice
function :^) Actually, I use a Ginsu! Hope this helps, good luck!
Virtually yours,
Michel
||
__||__ The opinions expressed by this author
Michel_J._Brown@nesbbx.rain.COM __ __ are mine, and mine alone, and anybody
|| claiming any resemblance to ideations
|| on my part should be ashamed to admit
|| it publicly! God Bless, and BCNU!
##
Subject: Thanks all
Date: Wed, 17 Nov 1993 8:03:40 -0700 (MST)
From: LESK@CC.SNOW.EDU
Thank you for your input everyone;
It seems I really blundered in the cycle editor, all the morphing
takes place in the action editor. It is very important though that not
only the same quantity of points, faces, edges be the same but that they
have to be drawn in the same order. It sounds like I cannot make two
distinct objects and then morph them I actually have to modify my most
complex of the two into the second form. I am afraid to get the great
beveled edge along the blade etc. is going to be beyond my time frame
for this project. It was a good idea though.
Thanks to all of you who responded it was a great help.
Lesk
##
Subject: Imagine on Aminet
Date: Tue, 16 Nov 1993 18:42:55 -0600 (CST)
From: kalb0003@gold.tc.umn.edu
Talked to Impulse about Imagine 2.0 on Aminet.
They were aware.
They were NOT pleased.
They arranged to have the file taken off-line. I successfully
ftp-ed it from wuarchive, including the readme file which went like this:
Short: Imagine 3.0 demo, disk 1 Uploaded by
xterm/jph@anaconda.bloomington.in.us
Impulse said that they want to find the individual(s) responsible
and have a little wall-to-wall counseling session with same (paraphrased,
what the guy I talked to said would be unrepeatable in a family mailing list).
Hope this clears things up.
##
Subject: re: 2.9 pro/con
Date: Wed, 17 Nov 93 07:18:46 -0700
From: Steve Koren <koren@hpfcogv.fc.hp.com>
> RE: Essence Textures
> Seems Impulse took a good look at the Essence textures and incorporated
> most of the ideas into the new texture set included with v2.9. While some
> of the Apex implementations of fractal noise are better, the new Impulse
> textures go far beyond the old collection. I can get a pretty good grass
I'm sure the new Impulse textures are a lot better than the old ones
(which were extremely primitive, IMHO). However, I think it still won't
be nearly up to what Essence can do for you. There are a couple of
problems:
* I have a lot of old projects that I'd like to be able to render
under 3.0, and in order to do this, I need Essence to work since
the objects use a lot of Essence textures.
* Impulse has added, what, 30 new textures? Essence has something
like 140, as I recall. Many of them are specialized to certain
applications yet are invaluable if you need them. Obviously most
won't be available in the Impulse list.
* The Essence textures are extremely well designed (getting the most
from the limited parameters available, working cooperatively with
other textures or not as you see fit, etc), and judging from what
I've seen from Impulse before, I really doubt theirs will be as
good or as powerful.
I still think I'm going to wait for Essence on 3.0 before I upgrade. It
sounds like even 2.9 has some nice features (I really would like the
"layers" feature), but loosing the equivalent of maybe 1/2 to 3/4 of my
currently available textures is too steep a price to pay. If it comes
to that, I'll just buy Real-3D and ditch Imagine entirely. Real-3D has
a lot of other features that we Imagine types can only drool over
anyway. Those Aladdin gas objects are pretty tempting too - I really
want true volumetric gasses, and it sounds like 3.0 won't give them to
me.
- steve
##
Subject: trace rendering
Date: Wed, 17 Nov 93 07:28:04 -0700
From: Steve Koren <koren@hpfcogv.fc.hp.com>
> images to show up if I'm in scanline) I set my global size at 0 and I
> scaled down all my elements to as small as I thought was necessary, and I'm
In *theory* setting the global size to 0 makes the world "just big
enough" to cover all of your objects. In *practice* I haven't found
that to be true at all. I have 2 or 3 projects in which, if i leave the
global size set to 0, some objects don't get rendered, but if I manually
set the global size, they do. I don't know if this is your trouble, but
its something to check at least.
Also, check your attributes. If your glass is *too* transparent, it
might not be showing up. Make sure you have some specular highlights
set, and turn hardness way up.
BTW, the unfortunate problem with global size is that, due to the way
octrees work, the larger you make the size compared to the objects in
your world, the slower your scene renders. This can be a huge
difference. I have seen 5 or 10X differences by adjusting the world
size. It usually isn't a problem, but sometimes if I want a single
small object to be far from my "main" scene, it forces my world size to
grow, and thus slows down the rendering considerably.
- steve
##
Subject: Could someone post me...
Date: Thu, 18 Nov 1993 00:13:37 +1300
From: Oren_Ben@kcbbs.gen.nz (Oren Ben)
Sorry but could someone post me the info ..that was apparently up a
short while ago on the hack needed to change the defined screen sizes in
Imagine 2.0 please...
Thanks
Oren Ben
##
Subject: Upload site
Date: Wed, 17 Nov 93 2:13:01 EST
From: mart4678@mach1.wlu.ca (Phil Martin u)
> >
> > Could you post some of these animations for us who don't have 2.9 yet?
> > I would really like to see some of what 2.9 can do before I upgrade.
> >
>
> Sure....if anyone can point to me to a site where I can upload some
> JPEG stills and anims, i will do so. I wonder if wuarchive is still down.
> Anyone know of ANOTHER site?
I like wcarchive.cdrom.com (I think it's the newest member of aminet).
They usually have all of the aminet files, and I almost always get on
first try.
>
> Mike C.
> mbc@po.CWRU.Edu
>
>
Phil Martin.
##
Subject: trace rendering
Date: Wed, 17 Nov 93 00:01:33 -0800
From: stevez@rhythm.com (Steve Ziolkowski)
I can think of two reasons why you might be having this problem. The first
one that comes to mind is RAM, or lack thereof. The second, is that
if all your objects are glass, and reflect stuff, well it could just be
that it's reflecting so much, that it's invisible. Turn down your
reflection values and transparancy values a bit and see if that helps.
Any other suggestions out there?
steveZ Rhythm & Hues, Inc.
celia!stevez@usc.edu
"That's not Art Linkletter, that's Mickey Mouse!"
-Art Linkletter, Disneyland opening day
##
Subject: RE: Rotation
Date: Wed, 17 Nov 93 17:10:00 PST
From: Stethem Ted 5721 <TedS@ms70.nuwes.sea06.navy.mil>
Why don't you just use the Rotate special effect? You could just go into
the action editor and set up the F/X line for different frames at different
rates to make it appear the ball is slowing down.
----------
From: imagine-relay
To: imagine
Subject: Rotation
Date: Wednesday, November 17, 1993 4:13PM
Hello...I have Imagine 2.0, and I'm trying to make an animation
of a ball that spins for a bit, then slows down and stops.
I figured the best way to do this was to define a circular path
in the plane intersecting the axis of rotation. The path was centered
with the ball. I set a track object on the path, and then locked the
alignment of the ball to the position of the track object. This way,
I can control the rate of angular acceleration of the ball with the
speed of the track object. It should work fine, except that
when the rotation of the ball exceeds a total of 90 degrees, it flips on
it's z-axis 180 degrees. That is, it flips 180 degrees on an axis
purpendicular to rotation... WHY???? How can I avoid this problem?
Chris Arthur
##
Subject: constant backgrounds
Date: Wed, 17 Nov 93 09:08:27 -0700
From: Steve Koren <koren@hpfcogv.fc.hp.com>
> > Is there a way to merge a single background frame with
> > parts of multiple frames to produce an animation? What I would
> > like to do, to save render time, is render one frame as a
> > background and then render only objects that move in all other
> > frames, then merge pics. Is there a way to merge only the part
> > of a pic (the part that contains the object that moves) ?
Yep, it can be done, but with certain limitations. Actually there are a
couple of ways to do it.
Given certain conditions (such as, all the moving objects are in *front*
of the static objects, and cast no shadows on the rest of the scene,
etc), you can do the following. Render the first frame with all the
objects *except* the moving ones. Save this picture. Use it as a
backdrop image for the other frames, in which you render only the
*moving* objects. As long as you don't move the camera, and don't have
your moving objects interact in any way with the still ones, it will be
indistinguishable from rendering everything for every frame.
One other alternative is to get clever with some image processing
program such as ImageFX or ADPro and merge your rendered foreground onto
the right place in the background.
- steve
##
Subject: Larger work area
Date: Wed, 17 Nov 93 19:42:33
From: greg.tsadilas@hofbbs.com
In a conversation with Mike Halvorson, he confirmed that Imagine 3.0
will allow you to increase the size of the editor's work area.
GreG tsadilas
##
Subject: re: 2.9 pro/con
Date: 18 Nov 1993 01:16:34 GMT
From: <mbc@po.cwru.edu>
> > I have been using 2.9 furiously since last Tuesday night, (ack! get a
> > life), depending on the animation, a camera track object may not be
> > necessary. The real-time window in the stage editor allows you to track
> > the camera right there, just use the "A"ngle and "Z"oom buttons to move
> > the camera in and out of the scene and rotate.
> >
>
> Yea, but I still like to track to an object and move the object around.
> since the real time window doesnt recalculate camera angle it defeats
> the purpose here. I called Impulse on this. They have something in the
> works for 3.0 to take care of this.
>
Yep...Actually...while you can now fairly easily set the camera to
look at what you want by angle etc....I still think I will use track alot.
Here is why: If you have to have the camera aiming at a couple of different
places over an animation, and you still wan't smooth interpolation, you've
got to use a track object on a path and have the camera follow that. Otherwise
the camrea will jsut rotate from a to b, then jump and start rotating from
b to c etc....
>
> COMBINE THEM!!!! The particle f/x (although it will work with any objects
> triangles) is meant to work with those cool detail editor particle objects.
> try it. make an object in detail. assign particles to it. save it.
> load in action editor. assign the particle f/x and render.
I am beginning to see that it does look just like another explode effect.
However, I have a feeling once we figure more out about it, it will be more
powerfull. Especially the emmision thing so simulate comet tails etc...
Mike C.
mbc@po.CWRU.Edu
PS: I am about to upload some demo JPG's....and hopefully some anims tomorrow!
##
Subject: Re: Larger work area
Date: Wed, 17 Nov 93 22:10:11 EST
From: Steve J. Lombardi <stlombo@eos.acm.rpi.edu>
>
>
> In a conversation with Mike Halvorson, he confirmed that Imagine 3.0
> will allow you to increase the size of the editor's work area.
>
> GreG tsadilas
>
Finally!! This is a vital feature. Was he more specific? Will they support
the display database or merely offer a few pre built size defaults?
|
steve lombardi | She won't get out of the tub. She
stlombo@acm.rpi.edu | has morning wood. -- Beavis
##
Subject: Re: 3D anim software rated
Date: Thu, 18 Nov 93 12:13:53 WST
From: Jason S Birch <jasonb@cs.uwa.edu.au>
> One of the catagories is "alpha channel compositing", which is not
> supported in Imagine. Is this related to "field rendering?".
> What do these two terms mean and how do they relate to 3d
> animation?
An Alpha Channel is typically an 8bit buffer the size of the image (at
least in my experience), where the value of the pixel in the alpha
channel is used to determine the amount of effect to apply to the
corresponding pixel in the image. Real3D allows you to specify an
Alpha Channel value for an object depicting it's transparency, so
when combined with another image (eg. a live video background) you
can "see through" transparent objects in the scene.
Field rendering relates to recording animations to video. Video is
interlaced, so each frame consists of two fields, one with all the even
scanlines, one with all the odd ones. If you record 25 complete frames
every second to video (30 for NTSC), so that each pair of fields make
up the same frame, then the motion won't be as smooth as possible. But,
if you try to record 50 frames every second, only half of each frame
(the even field for one, the odd for the next, etc) will be recorded,
which means you've wasted 50% of your rendering time. Field rendering
allows you to only render one field from each frame of animation, so
you record at a full 50 frames per second, but only render the pixels
you need. Real3D can do this. (I hope mentioning that isn't a no-no. ;)
--
Mr Jason Birch _--_|\ Internet: jasonb@cs.uwa.edu.au
Department of Computer Science / \ Tel (work): +61 9 380 1840
The University of Western Australia *_.--._/ Fax (work): +61 9 380 1126
Nedlands W. Australia 6009 v Tel (home): +61 9 386 8630
##
Subject: Re: 3D anim software rated
Date: Thu, 18 Nov 1993 00:39:44 -0600 (CST)
From: Schumacher Gordon C <gschumac@ux4.cso.uiuc.edu>
> which means you've wasted 50% of your rendering time. Field rendering
> allows you to only render one field from each frame of animation, so
> you record at a full 50 frames per second, but only render the pixels
> you need. Real3D can do this. (I hope mentioning that isn't a no-no. ;)
If you only rendered one frame out of a field, wouldn't you be missing
half of your vertical resolution? That is, after all, what they are
doing... even lines and odd lines. Now, if you rendered *both* fields
of a frame and combined them back into a field for recording, *then* I
could understand this. Am I missing something here? (I'm sure I am,
because this just doesn't make *sense*!)
--
Gordon Schumacher
/-------------------------------------------------------------------\
| Champaign- "We apologize for the inconvenience." _@_ |
| Urbana -HHGTTG / \ |
| kilroy was here | o o | |
\-------------------------------------------------------U|--U--|U---/
##
Subject: Re:Long Render Times
Date: Wed, 17 Nov 1993 22:39:21 -0800 (PST)
From: Steve Lopez <lopez@cyberspace.com>
>
> Date: Tue, 16 Nov 1993 21:13:31 -0800 (PST)
> From: Steve Lopez <lopez@cyberspace.com>
> Subject: Re: your mail
> To: imagine-relay@email.sp.paramax.com
> In-Reply-To: <9311170509.AA00901@cyberspace.com>
> Mime-Version: 1.0
> Content-Type: TEXT/PLAIN; charset=US-ASCII
>
>
>
> On Tue, 16 Nov 1993 imagine-relay@email.sp.paramax.com wrote:
>
> > I do have an '040 with 32-bit RAM but maybe not enough. Of course, based
> > on a past speed comparison between '040/33MHz and 486/66MHz, the rendering
> > time could be half what it is with a 486/66 MHz! And then there is those
> > privileged few that will be getting a Pentium system! I am glad to hear
> > that you will be leaving me behind in the dust |:^(
I run a 2000w/gvp040'33 and have had no probs with rendering time..I now
have 8 megs, but before I only had 4 and really did not notice any
increase in speed with the extra ram. I dont know what I'm doing right,
but my render times seem to hover around 1-2 minutes per frame. The most
I've been able to slow it down was 15 mins. That was full anti aliasing,
on a DCTV picture with a whole bunch of chrome and reflective glass
spheres in trace mode. I find that if you change the aliasing up to 30 or
so, it really goes much faster. Maybe you need to change your Octree
values? Like I say, I dont know what I'm doing right, but it's not really
that slow.
/------------------------------------------\
/ Steve Lopez | lopez@cyberspace.com \
/ Student at: | Art Institute of Seattle \
/ Program: //| Audio & Video Production \
\ Computer: // | Amiga 2000 GVP040'33 DCTV /
\ \\// | DSS8+ 9megs Flicker fxr /
\ IRC:Vis \Y | Midi DJ500c Supra 14.4k /
\------------------------------------------/
##
Subject: Re: 3D anim software rated
Date: Thu, 18 Nov 93 17:27:06 WST
From: Jason S Birch <jasonb@cs.uwa.edu.au>
> > which means you've wasted 50% of your rendering time. Field rendering
> > allows you to only render one field from each frame of animation, so
> > you record at a full 50 frames per second, but only render the pixels
> > you need. Real3D can do this. (I hope mentioning that isn't a no-no. ;)
>
> If you only rendered one frame out of a field, wouldn't you be missing
^^^^^ ^^^^^
Other way around.
> half of your vertical resolution? That is, after all, what they are
> doing... even lines and odd lines. Now, if you rendered *both* fields
> of a frame and combined them back into a field for recording, *then* I
> could understand this. Am I missing something here? (I'm sure I am,
> because this just doesn't make *sense*!)
:-) It took me a bit of thought to work it out at first, too.
OK, imagine we have two frames of an animation. Frame one looks like:
aaaa1111
aaaa2222
aaaa3333
aaaa4444
Frame two:
bbbb1111
bbbb2222
bbbb3333
bbbb4444
(Don't laugh at the "pictures" - it's the best way I could think of to
get my point across, vis. "a" and "b" to distinguish frames, and
"1".."4" to distinguish scanlines).
Now, to play back the animation at 50fps (PAL), we need one field from
frame "a" and one from frame "b" every 1/25 sec - ie:
aaaa1111
bbbb2222
aaaa3333
bbbb4444
(Note that the "b" fields aren't displayed until *after* the "a"
fields, and so it doesn't look "skewed" - ie. a moving object would
*not* look like:
********
********
********
********
because the first set of lines have faded by the time you see the
second. This is how a video camera would record it - remember, the
video camera records the image one field at a time, and it takes time
to record each field. So one frame, consisting of two consecutive
fields, does not represent a "snapshot" of time like with a photograph.
In fact, even one field doesn't represent an instant of time either,
because almost 1/50th of a second has elapsed from the time the field
began being recorded to the time it ended.)
So, since the even scanlines from "a" are never seen, and the odd
scanlines for "b" are never shown, we don't need to render them.
Instead, we render the scanlines that appear in the final "frame",
composit the fields into frames, and play back those frames at 25fps on
an interlaced display (which gives us 50 fields/sec, each field being
from a different frame, hence an apparent 50 frames/sec).
(For those using NTSC, convert each 25 to 30, and each 50 to 60. Oh,
and make the frames three lines tall instead of four. ;-)
> Gordon Schumacher
PS: I hope I've got that right. ;-) I haven't actually put it into
practice, but it seems to make sense.
--
Mr Jason Birch _--_|\ Internet: jasonb@cs.uwa.edu.au
Department of Computer Science / \ Tel (work): +61 9 380 1840
The University of Western Australia *_.--._/ Fax (work): +61 9 380 1126
Nedlands W. Australia 6009 v Tel (home): +61 9 386 8630
##
Subject: Particle Animation with Imagine 2.9
Date: Thu, 18 Nov 1993 11:15:02 +0100
From: Hannes Heckner <hecknerh@informatik.tu-muenchen.de>
Hi all,
I followed the conversation about the particle features in Imagine 2.9
But all I hear is that Imagine will break up an existing Object into
a lot of small objects. This is NOT what particla animation is supposed
to be.
1) There should be point objects with no faces as particles which
are rendered not like the other objects
2) There should be parameters about the lifetime velocity, splitting
etc. of particles. You should be able to define sources for
particles etc. etc.
Any comments about this ?
Hannes
PS: I still have no copy of Imagine 2.9 so I cannot figure out more
about this
##
Subject: Re: render times and bac
Date: Tue, 16 Nov 93 13:42:00 PDT
From: Jeff.Saffold@lookout.com (Jeff Saffold)
> Is there a way to merge a single background frame with
> parts of multiple frames to produce an animation? What I would
> like to do, to save render time, is render one frame as a
> background and then render only objects that move in all other
> frames, then merge pics. Is there a way to merge only the part
> of a pic (the part that contains the object that moves) ?
Sure.. Render the non moving frame, and then in the Action editor,
add that frame in the globals as a background image. If the camera
doesn't move, then you won't have any problems with this.
// Jeff Saffold
\X/ Only the Amiga makes it possible.
___
X MsgView V1.13 [R029] X Dyslexic Devil worshipper sells soul to Santa!
--
*******************************************************************************
* Cuerna Verde BBS FidoNet Gateway Data/Fax: 1-719-545-8572 *
* Pueblo, Colorado USA FidoNet: 1:307/18 *
*******************************************************************************
##
Subject: Volcanoes
Date: Thu, 18 Nov 93 06:50:49 -0700
From: Steve Koren <koren@hpfcogv.fc.hp.com>
> I wanted to create a volcanoe, so I presume the best way is to give it a
I've found that its best to create such a thing in VistaPro 3, save in
"Turbo Silver" format, and import the object to Imagine. Once in
imagine, I turn on phong shading, and add some Essence textures to add
bump maps and such. You can get pretty decent results that way.
If you want to make your own volcano instead of use whatever comes out
of your landscape program, try their (VRLI's) "Terraform" utility or
the similar thing from Natural Graphics.
- steve
##
Subject: Re: Particle Animation with Imagine 2.9
Date: Thu, 18 Nov 1993 07:46:41 -0600 (CST)
From: Cliff Lee <cel@tenet.edu>
> I followed the conversation about the particle features in Imagine 2.9
> But all I hear is that Imagine will break up an existing Object into
> a lot of small objects. This is NOT what particla animation is supposed
> to be.
> 1) There should be point objects with no faces as particles which
> are rendered not like the other objects
> 2) There should be parameters about the lifetime velocity, splitting
> etc. of particles. You should be able to define sources for
> particles etc. etc.
Actually there is 2 parts to what Impulse is calling particles. 1st is
the ability to take each face of an object and make it another (there is a
large dialog box allowing you to choose the particle type/size/number).
Once you've defined a particle object you may apply the particle effect to
it in the stage editor. This where the acceleration/wind/velocity factors
are set. I think when you see this that you be comfortable with what your
expecting.
Only difference is that the individual particle are not just points. They
are (or can be) other Imagine objects or primatives.
I'm getting excited about this. I was experimenting with it last night
and made the most basic anim (not worth U/Ling) of a sphere of particles.
I chose randomize colors which made each particle a different color. Then
applied the particle f/x to it and rendered it in 30 frames. It gave me a
kind of firework explosion effect. First the particle displaced then they
started to fall. I tweeked the paramters a bit until it looked more
realistic. Nice feature. Now all I have to do is find a use for it in my
work!
##
Subject: Re:Long Render Times
Date: Thu, 18 Nov 1993 09:19:07 -0500
From: Jason B Koszarsky <kozarsky@cse.psu.edu>
> > I do have an '040 with 32-bit RAM but maybe not enough. Of course, based
> > on a past speed comparison between '040/33MHz and 486/66MHz, the rendering
> > time could be half what it is with a 486/66 MHz! And then there is those
>>> increase in speed with the extra ram. I dont know what I'm doing right,
>>> but my render times seem to hover around 1-2 minutes per frame. The most
It would be nice if Impulse would optimize Imagine for running on an 040
machine. Slower rendering times now are partly because Imagine ISN'T optimized
for an 040.
Jason K.
CYBERNETIC EXPRESSIONS
##
Subject: Demo stuff of 2.9 Uploaded!
Date: Thu, 18 Nov 93 11:15:03 -0500
From: mbc@po.CWRU.Edu (Michael B. Comet)
By popular demand...the following has been to aminet.
I JUST uploaded it to the /systems/amiga/aminet/new directory
of wuarchive.wustl.edu.
It will probably be a while until things propoate.
---
pix/trace:
I29Demo1.JPG
I29Demo2.JPG
I29Demo3.JPG
The first 2 are pics of a rectangle and the result of using the new
deformation tools. Also shows brushtacking.
The last is 4 pics in 1 of brushtacking and particle objects.
gfx/anim:
Particle.lha
BrushTack.lha
The first is a 60 second particle anim....unfortunately i stopped the
anim halfway and then restarted...so I think it may use multiple palettes
now.
The second is a quick anim of a morph to show brushtacking.
Enjoy!
--
+------------------------------------------------------------------------+
| Michael Comet, mbc@po.CWRU.Edu, CWRU Software Engineer/Graphics Artist |
+------------------------------------------------------------------------+
##
Subject: Re: 2.9 particle question
Date: Thu, 18 Nov 93 10:24:51 CST
From: setzer@comm.mot.com (Thomas Setzer)
> >
> >Someone here (perhaps Jeff) mentioned they made a rain particle animation. I
> >am trying the same thing but have a problem setting the particles in motion.
> >here's what I did:
> >
> >1. made a 4X4 plane in detail editor. in particle requester I went for small
> >inscribed spheres. saved it.
> >
Could you replace the inscribed spheres with, say, cat and dog objects? Or
Snowflake objects? Or any object you have? Beethoven(sp?) busts falling
from the sky? Cows, even better!(Note: I am not advocating cruelty to cows:)
I get the impression that this is possible, but I haven't heard that anyone
has done it yet.
Tom Setzer
setzer@ssd.comm.mot.com
"And of course, I'm a genius, so people are naturally drawn to my fiery
intellect. Their admiration overwhelms their envy!" - Calvin
##
Subject: Mac 3D software
Date: Thu, 18 Nov 93 09:17:33 PST
From: CarmenR@cup.portal.com
I went up to SDSU last night, where a Mac user group was demonstrating
their latest 3D software "Strata Studio Pro." Believe me guys, they have
a long way to catch up to Imagine 2.9. Basically they have 2 effects, tweens
and morphs. Wow. The Modelling is terrible. I think the program is
basically a rendering engine and they expect you to go out and buy a
seperate modelling program to do serious modelling for it. It does have
a 'pretty' interface, when you apply a texture to an object, say, a flat
projection style map, there's a nice little illustration showing your image
being projected on a sphere. And you can do a 'quickrender' of your object
with the texture in a little window. He was on an 840AV Mac I believe, and
the rendering was fairly quick for the example model he was playing with.
You can save your textures and attribute presets and they all appear in
colourfull doc images on the bottom of the screen. The motion of most of the
anims I saw that was created with it were very mechanical in motion.. Kinda
like animation in Imaigne using tweens. The price for this power-package?
$1,400.
Carmen Rizzolo - Crazed Artist
CarmenR@cup.portal.com
##
Subject: Re: 2.9 particle question
Date: Thu, 18 Nov 1993 11:23:45 -0600 (CST)
From: Daniel Jr Murrell <djm2@Ra.MsState.Edu>
With all the talk about particle rendering, I'm wondering what kind of render
speed you guys are getting. My understanding of particle anim rendering is that
you take very little of a hit, whether you render 10 or 10,000 particles. Is
that correct, and more importantly, does Imagine comply? Thanks.
Dan
##
Subject: Backgrounds...
Date: Thu, 18 Nov 93 09:55:16 -0800
From: stevez@rhythm.com (Steve Ziolkowski)
>> Render the non moving frame, and then in the Action editor,
>>add that frame in the globals as a background image. If the camera
>>doesn't move, then you won't have any problems with this.
Hey, here's an idea I've had a fair amount of success with, and that
is using moving background images. If you start out in your
stage, setup the *entire* animation with all the elements.
Then, make a copy of the entire project, rename it .background
or something. In the .background project delete all the foreground
objects and render. In the original project, delete all the
background objects. As long as the camera and lights are the
same, when you put the .background images into the backdrop,
there isn't a seam between the two. You can even run the
images through adPro or something to blur them to fake focal
length. This saves mondo memory, and if you set it up correctly,
you can even raytrace different layers to get the desired look,
while leaving others scanline. I've used up to four layers. The
major drawback to this is fog and transparent objects won't show
up in all but the first layer.
Good luck!
steveZ Rhythm & Hues, Inc.
celia!stevez@usc.edu
"That's not Art Linkletter, that's Mickey Mouse!"
-Art Linkletter, Disneyland opening day
##
Subject: Re: Volcanoes
Date: Thu, 18 Nov 1993 11:19:12 -0600 (CST)
From: Daniel Jr Murrell <djm2@Ra.MsState.Edu>
>
>
> > I wanted to create a volcanoe, so I presume the best way is to give it a
>
> I've found that its best to create such a thing in VistaPro 3, save in
> "Turbo Silver" format, and import the object to Imagine. Once in
> imagine, I turn on phong shading, and add some Essence textures to add
> bump maps and such. You can get pretty decent results that way.
>
> If you want to make your own volcano instead of use whatever comes out
> of your landscape program, try their (VRLI's) "Terraform" utility or
> the similar thing from Natural Graphics.
>
> - steve
>
Even easier than that, get Terrain, make a really high peak, and save it out.
Use InterChange Plus (John said it handles Terrain objects fine) or Turbo Silver
to save it back in an Imagine compatible format. Select points of the highest
peak, and drop them down inside the "volcano."
Dan
Terrain Advocate :)
##
Subject: RE: Long Render Times
Date: Thu, 18 Nov 93 10:07:00 PST
From: Stethem Ted 5721 <TedS@ms70.nuwes.sea06.navy.mil>
I agree completely! Maybe some of you guys that are calling Impulse
could express very deeply that 3.0 should be optimized for the '040. Or
maybe we could start a fax campaign here and inform Impulse of this
need/desire. I would think that if the Imagine code is written in C or some
high-level language, it would just be a matter of the right choice of
compiler to optimize the code for the '040. However, I don't know if
Imagine may not be written in Assembly or maybe parts of it. I suspect it
is written in C, though, seeing the port over to 80X86 MS-DOS machines. If
it were written in Assembly, I think it would be faster, too. But that
would probably be a one-way ticket to the funny farm to do that!
----------
From: imagine-relay
To: imagine; lopez
Subject: Re:Long Render Times
Date: Thursday, November 18, 1993 9:19AM
> > I do have an '040 with 32-bit RAM but maybe not enough. Of course,
based
> > on a past speed comparison between '040/33MHz and 486/66MHz, the
rendering
> > time could be half what it is with a 486/66 MHz! And then there is
those
>>> increase in speed with the extra ram. I dont know what I'm doing right,
>>> but my render times seem to hover around 1-2 minutes per frame. The
most
It would be nice if Impulse would optimize Imagine for running on an 040
machine. Slower rendering times now are partly because Imagine ISN'T
optimized
for an 040.
Jason K.
CYBERNETIC EXPRESSIONS
##
Subject: Re: 2.9 particle question
Date: Thu, 18 Nov 93 13:29:38 CST
From: setzer@comm.mot.com (Thomas Setzer)
Are the particles in the detail editor made of small triangles? Or are they
more like the primitive sphere? I would think the intersection of a true
sphere would be much quicker than the intersection of a sphere approximated
by many triangles.
What other particles are availible in the detail editor? Other than any object
you've created? ie are there special objects which were made specifically for
particles? If so, what are they?
Tom Setzer
setzer@ssd.comm.mot.com
"And of course, I'm a genius, so people are naturally drawn to my fiery
intellect. Their admiration overwhelms their envy!" - Calvin
##
Subject: Field Rendering
Date: Thu, 18 Nov 1993 11:37:09 -0800
From: kkalnasy@bvu-lads.loral.com (Kent Kalnasy)
At 5:27 PM 11/18/93 +0900, Jason S Birch wrote:
>OK, imagine we have two frames of an animation. Frame one looks like:
>
> aaaa1111
> aaaa2222
> aaaa3333
> aaaa4444
>
>Frame two:
>
> bbbb1111
> bbbb2222
> bbbb3333
> bbbb4444
>
>(Don't laugh at the "pictures" - it's the best way I could think of to
>get my point across, vis. "a" and "b" to distinguish frames, and
>"1".."4" to distinguish scanlines).
>
>Now, to play back the animation at 50fps (PAL), we need one field from
>frame "a" and one from frame "b" every 1/25 sec - ie:
>
> aaaa1111
> bbbb2222
> aaaa3333
> bbbb4444
>
>(Note that the "b" fields aren't displayed until *after* the "a"
>fields, and so it doesn't look "skewed" - ie. a moving object would
>*not* look like:
>
> ********
> ********
> ********
> ********
>
>because the first set of lines have faded by the time you see the
>second. This is how a video camera would record it - remember, the
>video camera records the image one field at a time, and it takes time
>to record each field. So one frame, consisting of two consecutive
>fields, does not represent a "snapshot" of time like with a photograph.
>In fact, even one field doesn't represent an instant of time either,
>because almost 1/50th of a second has elapsed from the time the field
>began being recorded to the time it ended.)
>
>So, since the even scanlines from "a" are never seen, and the odd
>scanlines for "b" are never shown, we don't need to render them.
>Instead, we render the scanlines that appear in the final "frame",
>composit the fields into frames, and play back those frames at 25fps on
>an interlaced display (which gives us 50 fields/sec, each field being
>from a different frame, hence an apparent 50 frames/sec).
I wish I could say that I followed your logic, but I didn't always.
It's simply a matter of timing. In the past, we've rendered static
images and then combined them into an animation. The assumption was
that both fields of each interlaced frame depicted the same point in
time, even though they were displayed sequentially on playback.
Consequently, moving objects tended to crawl, depending on the shape,
contrast, direction of motion and speed of motion.
Field rendering simply acknowledges the fact that each field of an
interlaced frame represents a different point in time. In the past
we rendered fields like this: and displayed them like this:
Frame 1 2 3 4 5 6 Frame 1 2 3 4 5 6
+---+---+---+---+---+-- +---+---+---+---+---+--
Field A A A A A A Field A A A A A A
Field B B B B B B Field B B B B B B
Field rendering allows us to and display them as always, but
render fields like this: without introducing artifacting:
Frame 1 2 3 4 5 6 Frame 1 2 3 4 5 6
+---+---+---+---+---+-- +---+---+---+---+---+--
Field A A A A A A Field A A A A A A
Field B B B B B B Field B B B B B B
Field "1B" shows the image as it would appear halfway between fields
1A and 2A. The animation looks strikingly more natural, as if it
had been videotaped rather than rendered. There is no savings in
the number of pixels rendered; it is simply a method of improving
the appearance of the animation.
------------------------------------------------------------------------------
Kent Kalnasy kkalnasy@bvu-lads.loral.com
Loral Advanced Distributed Simulation, Inc.
Bellevue, Washington (206) 957-3278
------------------------------------------------------------------------------
##
Subject: 3DS Vs. Imagine speed
Date: Thu, 18 Nov 93 18:02:49 EDT
From: "Breno A. Silva" <INF02@BRUFSE.BITNET>
>duplicate scenes in Imagine, Lightwave and 3D Studio I can speak very
Hey, great about the speed comparison LW X Imagine, IMHO the anti-alia-
sing of LW STINKS, but don't make me curious! Let me know the speed in
the 3D Studio Rick, after all, I want to use the fastest and the better,
be it PC or Amiga only software. Anyway, I know the 3DS makes reflections
and shadows WITHOUT ray-tracing, so faster render times won't surprise me.
Is it true that 2.9 makes reflection map like 3DS and Topas?
BTW, I hope Imagine 3.0 include Lens Flare. I think it was one of the
best (if not THE best) additions of LW 3. With the stage editor inter-
activity AND the screen refresh speed up I was waiting, LW will rest a bit,
since I prefer the Imagine interface (except the animation of 2.0 that's
now GREAT!).
Breno A. Silva (INF02@BRUFSE.BITNET)
##
Subject: Re: 2.9 particle question
Date: Thu, 18 Nov 93 16:21:04 -0500
From: mbc@po.CWRU.Edu (Michael B. Comet)
>
>
>Could you replace the inscribed spheres with, say, cat and dog objects? Or
>Snowflake objects? Or any object you have? Beethoven(sp?) busts falling
>from the sky? Cows, even better!(Note: I am not advocating cruelty to cows:)
>
>I get the impression that this is possible, but I haven't heard that anyone
>has done it yet.
It can be done...but I have not yet tried it. Basically, there is
a button under the choices for the pyramids, diamonds etc....that is labeld
file. Clicking on this brings up an object load requestor. This is
similar to the buttons for brushes and textures. it gets checked when in
use.
Anyhow, I guess you can use any object you want for this. I wonder
how it will handle morphs....maybe you could morph 2 particle objects with
different file name objs....maybe not. Hmmm.
I have heard that this just CHEWS up memory....so you only want to
use it for small particle objects...or simple objects.
Maybe I'll try this later. Right now I'm working on finishing this
cool robot since it looks like I will be able to actually animate it thanks
to states. No more of that cycle editor axis mangling crud.
As far as the rendering times....I have not gone crazy with
particles but they seem to render at a fair pace. If i recall though the
manual warns one about using to much since rendering 10,000 dodecahedraons
will have 10*10,000 faces etc....
--
+------------------------------------------------------------------------+
| Michael Comet, mbc@po.CWRU.Edu, CWRU Software Engineer/Graphics Artist |
+------------------------------------------------------------------------+
##
Subject: RE: Long Render Times
Date: Thu, 18 Nov 93 14:21:15 -0700
From: Steve Koren <koren@hpfcogv.fc.hp.com>
> could express very deeply that 3.0 should be optimized for the '040. Or
If they are using SAS/C (good chance they are, I'd guess), then before
version 6.50 you could optimize for the 68030 (which used 030 op codes),
but there was no specific instruction scheduler for the 040. As of
6.50, there is; it reorders instructions to try to get maximum
parallelism from the CPU. (Certain combinations of instructions can be
run in parallel, so it moves things around where possible to get those
combinations).
I've heard this can help quite a bit. Exactly how much, I dunno. I'd
wager it helps less for floating point intensive code than otherwise.
There is also the problem that the 040 has fewer FP instructions than
the 882 had, so you need to simulate some in software. However, these
are mostly transcendental operations which, oddly enough, 3D rendering
software doesn't tend to use much. (*)
Guess the best bet is to wait for Impulse to port to that new transputer
card. The last Amigaworld had a blurb about it. Supposedly a number of
number-crunching type apps are porting to it. You just keep adding CPUs
until you run out of money or it becomes fast enough for your tastes,
whichever comes first :-) (Well, I assume there is some smallish
limit, but its probably bigger than my budget will support anyway).
- steve
(*) or I should say, doesn't tend to use explicitly. Most of the
transcendental functions tend to be computed implicitly, such as taking
cross products of normalized vectors instead of cosines of included
angles, etc. Faster that way, and generally more useful.
PS - Speaking of transcendental math, I'm resisting putting in the joke
about the fellow who transcended dental medication.
##
Subject: Re(2): Sphere primitive
Date: Thu, 18 Nov 93 23:00 GMT0
From: Jacek Artymiak <jartymiak@cix.compulink.co.uk>
>The sphere primitive (as in 1.1) is a mathematical construct, not a set of
>discrete points. The wireframe you see is just an approximation. Use the
>Sphere listed with the other primitives, where you can define the point
>count, and you'll be able to edit it.
>------------------------------------------------------------------------------
>Doug Kelly Information Specialist First Consulting Group
>Danimal
>--Rick Rodriguez
Thanks a lot! I was unnecessarily worried then. Other than that I love
Imagine 2.9. The Agate texture is really cool.
Jacek
##
Subject: Re: Field Rendering
Date: Thu, 18 Nov 93 17:48:07 CST
From: setzer@comm.mot.com (Thomas Setzer)
> I wish I could say that I followed your logic, but I didn't always.
> It's simply a matter of timing. In the past, we've rendered static
> images and then combined them into an animation. The assumption was
> that both fields of each interlaced frame depicted the same point in
> time, even though they were displayed sequentially on playback.
> Consequently, moving objects tended to crawl, depending on the shape,
> contrast, direction of motion and speed of motion.
>
[nice drawing deleted]
> Field "1B" shows the image as it would appear halfway between fields
> 1A and 2A. The animation looks strikingly more natural, as if it
> had been videotaped rather than rendered. There is no savings in
> the number of pixels rendered; it is simply a method of improving
> the appearance of the animation.
The savings come into play when someone might have done this manually. Say
you wanted a 30 frame(60 field) animation. You would render 60 full frames
(120 fields) and then throw away the alternating odd scan lines eg 1,3,5,...
on frame one, 2,4,6,... on frame two, 1,3,5,... on frame three, etc. See?
60 fields/s has got to be a lot smoother than 30 frames(30 fields)/s. Nice.
The PAR handles this, doesn't it? Someone should try it, let us know the
difference.
Tom Setzer
setzer@ssd.comm.mot.com
"And of course, I'm a genius, so people are naturally drawn to my fiery
intellect. Their admiration overwhelms their envy!" - Calvin
##
Subject: Imagine 2.9/OpalVision
Date: Fri, 19 Nov 93 01:24:06 GMT
From: Tony Jones <nova@ibmpcug.co.uk>
Posted for Paul Rance (2:253/516.2@fidonet.org):
SJL> 4. OPALVISION BUG. when Opalvision is selected in preferences, the
SJL> machine will
SJL> lock up upon attempting to display to the card. This only applies to
SJL> the Amiga 4000 with Opal card. Most other software I've used with Opal
SJL> has had this bug, and is easily fixed.
This doesnt happen on my 4000/opal setup. renders fine, the opal screen is
drawn out in real time with the imagine screen superimposed ontop, when
its finished it appears by itself.
Paul
##
Subject: Re: Field Rendering
Date: Thu, 18 Nov 1993 18:00:38 -0800
From: kkalnasy@bvu-lads.loral.com (Kent Kalnasy)
At 5:48 PM 11/18/93 -0600, Thomas Setzer wrote:
>> There is no savings in the number of pixels rendered; it is simply
>> a method of improving the appearance of the animation.
>
>The savings come into play when someone might have done this manually. Say
>you wanted a 30 frame(60 field) animation. You would render 60 full frames
>(120 fields) and then throw away the alternating odd scan lines eg 1,3,5,...
>on frame one, 2,4,6,... on frame two, 1,3,5,... on frame three, etc. See?
Yes! Now I understand what you were talking about. Thanks.
------------------------------------------------------------------------------
Kent Kalnasy kkalnasy@bvu-lads.loral.com
Loral Advanced Distributed Simulation, Inc.
Bellevue, Washington (206) 957-3278
------------------------------------------------------------------------------
##
Subject: Re: Field Rendering
Date: Fri, 19 Nov 93 12:28:29 WST
From: Jason S Birch <jasonb@cs.uwa.edu.au>
> Field rendering simply acknowledges the fact that each field of an
> interlaced frame represents a different point in time. In the past
> we rendered fields like this: and displayed them like this:
>
> Frame 1 2 3 4 5 6 Frame 1 2 3 4 5 6
> +---+---+---+---+---+-- +---+---+---+---+---+--
> Field A A A A A A Field A A A A A A
> Field B B B B B B Field B B B B B B
>
> Field rendering allows us to and display them as always, but
> render fields like this: without introducing artifacting:
>
> Frame 1 2 3 4 5 6 Frame 1 2 3 4 5 6
> +---+---+---+---+---+-- +---+---+---+---+---+--
> Field A A A A A A Field A A A A A A
> Field B B B B B B Field B B B B B B
>
> Field "1B" shows the image as it would appear halfway between fields
> 1A and 2A. The animation looks strikingly more natural, as if it
> had been videotaped rather than rendered.
Ah, those are much better pictures for what I was trying to say. :-)
That's exactly right. (BTW - since the instant of time at the top of
field "1A" is almost 1/50th of a second before the instant of time at
the bottom of field "1A" - that is, the scene should have changed in
the meantime - presumeably taking that into account would give you an
even better looking animation. However, I doubt you could notice.)
> There is no savings in
> the number of pixels rendered; it is simply a method of improving
> the appearance of the animation.
The savings comes from recognizing that you really *do* want 50 frames
per second for the smoothest motion, but you don't have to render every
scanline in those frames - you just render one field's worth. If you
wanted to do achieve the same smoothness of motion with Imagine, you
would have to render each frame in it's entirety. Therefore it would
look like:
Frame 1 2 3 4 5 6
+---+---+---+---+---+--
Field A * A * A * A * A * A *
Field * B * B * B * B * B * B
where the "*"'s represent the fields that were rendered (since you must
render a complete frame) but never seen. Looking at it from this POV,
field rendering is simply a method for saving the number of pixels
rendered given you want the smoothest possible animation. :-)
> ------------------------------------------------------------------------------
> Kent Kalnasy kkalnasy@bvu-lads.loral.com
> Loral Advanced Distributed Simulation, Inc.
> Bellevue, Washington (206) 957-3278
> ------------------------------------------------------------------------------
--
Mr Jason Birch _--_|\ Internet: jasonb@cs.uwa.edu.au
Department of Computer Science / \ Tel (work): +61 9 380 1840
The University of Western Australia *_.--._/ Fax (work): +61 9 380 1126
Nedlands W. Australia 6009 v Tel (home): +61 9 386 8630
##
Subject: Essence II/Imagine 2.9
Date: Thu, 18 Nov 93 21:10:57 PST
From: RayTrace@cup.portal.com
What's the deal?! Is there anyway to get Imagine 2.9 to work with Essence II
textures?!?! Will Imagine 3.0 work with Essence II textures? I mean sure
it came with 50 textures but I want to use my Essence II textures which I
have used in countless projects before!! Has anyone talked to Impulse
about this? Or will there be an upgrade for Essence textures that work
with 2.9/3.0? Otherwise Imagine 2.9 is GREAT! I love those particles!
##
Subject: Re: 2.9 particle question (fwd)
Date: Thu, 18 Nov 1993 23:19:00 -0600 (CST)
From: Daniel Jr Murrell <djm2@Ra.MsState.Edu>
Forwarded message from Greg:
> From greg.tsadilas@hofbbs.com Thu Nov 18 22:30:34 1993
> Subject: Re: 2.9 particle question
>
>
> Dan, feel free to send this to the IML
>
> Rendering times using the Paricle system...it can increase dramatically
>
> A particle can be a FACE, a PRIMITIVE, or another OBJECT
>
> Depending which you choose, you will either add NO additional faces, to
> thousands of additional faces. Each face of an object that you apply the
> Particle system to becomes a particle. If you predefine the particle to be a
> 100 faced object, the number of faces becomes 100 x (number of faces in
> affected object).
>
> GreG
>
ouch! so does that mean that Imagine will be rendering several thousand particles
for a decently complex particle system, with the particles being a hundred or so
faces worth each? Does Imagine include any kind of sppedup algorithms to handle
these cases? Seems like that could take hours to render each frame.
Dan
##
Subject: RE: Long Render Times
Date: Fri, 19 Nov 1993 01:21:19 -0500
From: Jason B Koszarsky <kozarsky@cse.psu.edu>
>I've heard this can help quite a bit. Exactly how much, I dunno. I'd
>wager it helps less for floating point intensive code than otherwise.
Try running AIBB first using standard 68000 code and then use the 020+
optimized code. I did this on my 3000 and the speed increase is very
significant.
Jason K.
CYBERNETIC EXPRESSIONS
##
Subject: Re: Field Rendering
Date: 19 Nov 93 01:51:00 EST
From: "J_GEORGE" <J_GEORGE@vger.nsu.edu>
Thanks to those that posted that information on field renderings. I have a
couple of questions:
1] Although this can't be done within Imagine, has anyone experimented with
doing so through one of the image processing programs using arexx?
2] If so, since there would be no time saved during rendering, would the
savings be passed on to the actual memory used for an animation? Correct me if
I'm wrong, but that would seem to be a good way to economize on memory usage
for those creating animations to be played back in memory, thus longer anims
using less RAM (for those of us who are stuck being memory conscious). ;-)
3] Okay, so I had to add a third question. Is this a technique commonly used by
professional [computer] animators?
4] LAST QUESTION: Are there any books/texts that describe this process in
greater detail?
Thanks in advance..
I\/Iax I\Iomad
##
Subject: Re: Archive#43
Date: Fri, 19 Nov 93 07:41:05 MST
From: pringleg@cuugnet.cuug.ab.ca (Greg Pringle)
> I'm sure that I already posted this but can't remember it appearing on
> the list.
That's because you forgot to Cc: to the mailing list when
you replied to me..
> couldn't upload it to the usual spot on wuarchive. I'll upload it there
> when waurchive lets me.
> Nik.
Just came through today.. I sent a copy up to the Imagine Archive.
I also put up a mpeg animation done with imagine PC, and a short
text file describing how to make mpeg animations from imagine,
in case anyone's interested..
Greg
--
+------------------------------+---------------------------------------+
| Greg Pringle | Amiga VBBS - Multitasking, Windowed |
| pringle@cpsc.ucalgary.ca | BBS'ing! |
| pringleg@cuugnet.cuug.ab.ca | VBBS 14.4K: (403) 284-2048 & 284-5625 |
+------------------------------+---------------------------------------+
##
Subject: RE: Long Render Times
Date: Fri, 19 Nov 93 06:55:09 -0700
From: Steve Koren <koren@hpfcogv.fc.hp.com>
> Try running AIBB first using standard 68000 code and then use the 020+
> optimized code. I did this on my 3000 and the speed increase is very
> significant.
Yeah, but this is a different optimization. Using 020+ code makes the
compiler use the enhanced 020(30) opcodes which are faster for many
things than the 000. But there are two completely separate
optimizations possible for the 040 over the 030. One, which AIBB *does*
do, is the FP optimization. The other, which it doesn't do, is a
*reordering* of the instruction stream to achieve maximum
parallelization and pipelining - this is somewhat independent of whether
it uses 020/030 or just 000 opcodes. So basically, code which is
*really* optimized for specifically the 040 will get an even bigger
improvement than it appears from your AIBB tests. The 060 will get a
bigger improvement yet from this sort of instruction reordering. Risc
chips depend heavily on this also.
Now if we could just get C= to give us a real memory subsystem in the
4000 we'd get another big improvement in speed.
Well, this is straying away from IML topics, so I'll be quiet now...
- steve
##
Subject: 040 optimization
Date: Fri, 19 Nov 93 10:11:08 EST
From: mart4678@mach1.wlu.ca (Phil Martin u)
>
> need/desire. I would think that if the Imagine code is written in C or some
> high-level language, it would just be a matter of the right choice of
> compiler to optimize the code for the '040. However, I don't know if
> Imagine may not be written in Assembly or maybe parts of it. I suspect it
> is written in C, though, seeing the port over to 80X86 MS-DOS machines. If
> it were written in Assembly, I think it would be faster, too. But that
> would probably be a one-way ticket to the funny farm to do that!
I don't know, if you've got working C code it would mostly be a
mechanical type of job to translate it into ASM. PageStream 3.0 was
entirely re-written in ASM, and I'm sure it sees a big speed increase.
Just look at the speed difference between PD games written in C and
commercial games written in ASM, there's no comparison.
>
> It would be nice if Impulse would optimize Imagine for running on an 040
> machine. Slower rendering times now are partly because Imagine ISN'T
> optimized for an 040.
I wouldn't worry about this too much, from what I hear the speed increase
is generally under 10% in the best circumstances.
>
> Jason K.
> CYBERNETIC EXPRESSIONS
>
Phil Martin.
##
Subject: Re: Field Rendering
Date: Fri, 19 Nov 93 09:44:01 CST
From: setzer@comm.mot.com (Thomas Setzer)
Sorry for the length
>
> Thanks to those that posted that information on field renderings. I have a
> couple of questions:
>
> 1] Although this can't be done within Imagine, has anyone experimented with
> doing so through one of the image processing programs using arexx?
I thought 2.9 handled this? isn't that why this discussion came up? Did I
misread again?
> 2] If so, since there would be no time saved during rendering, would the
> savings be passed on to the actual memory used for an animation?
Whoops, you missed something. There is a time savings. I didn't notice till
today, but a previous post kinda misstated something
> we rendered fields like this: and displayed them like this:
>
> Frame 1 2 3 4 5 6 Frame 1 2 3 4 5 6
> +---+---+---+---+---+-- +---+---+---+---+---+--
> Field A A A A A A Field A A A A A A
> Field B B B B B B Field B B B B B B
>
> Field rendering allows us to and display them as always, but
> render fields like this: without introducing artifacting:
>
> Frame 1 2 3 4 5 6 Frame 1 2 3 4 5 6
> +---+---+---+---+---+-- +---+---+---+---+---+--
> Field A A A A A A Field A A A A A A
> Field B B B B B B Field B B B B B B
>
The above isn't really true, as another poster pointed out.
Its more like this(let me try my hand at ascii):
> we rendered fields like this: and displayed them like this:
>
> Frame 1 2 3 4 5 6 Frame 1 2 3 4 5 6
> +---+---+---+---+---+-- +---+---+---+---+---+--
> Field A B C D E F Field A B C D E F
> Field A B C D E F Field A B C D E F
>
> Field rendering allows us to
> render fields like this: And display them like this:
>
> Frame 1 2 3 4 5 6 Frame 1 2 3 4 5 6
> +---+---+---+---+---+-- +---+---+---+---+---+--
> Field A C E G I K Field A C E G I K
> Field B D F H J L Field B D F H J L
See, twice as many fields, for the same price! This adds smoothness to the
animation.
> 3 Okay, so I had to add a third question. Is this a technique commonly used by
> professional [computer] animators?
I believe it is for TV. For movies, it wouldn't make sense. As to how they
go from movies to TV, you got me.
Tom Setzer
setzer@ssd.comm.mot.com
"And of course, I'm a genius, so people are naturally drawn to my fiery
intellect. Their admiration overwhelms their envy!" - Calvin
##
Subject: Re[2]: Field Rendering
Date: 19 Nov 93 07:35:00 -0800
From: Ed_Totman@ucsdlibrary.ucsd.edu
<...stuff deleted...>
>3] Okay, so I had to add a third question. Is this a technique commonly used by
>professional [computer] animators?
....
>Thanks in advance..
>
> I\/Iax I\Iomad
The real reason I asked about field rendering in the first place
is that I know someone who produces animations for TV
commercials, and his animations look so much smoother than mine.
I'm sure that field rendering is not the only reason, though.
##
Subject: Re: Field Rendering
Date: Fri, 19 Nov 93 9:18:11 PST
From: jwalkup@sfsuvax1.sfsu.edu (Jeff Walkup)
Thomas Setzer writes:
>
> 60 fields/s has got to be a lot smoother than 30 frames(30 fields)/s. Nice.
> The PAR handles this, doesn't it? Someone should try it, let us know the
> difference.
Yes the PAR handles this nicely. I did a few tests with LW 3.0. I also
used motion blur. Since motion blur adds about 5 more units of motion
per frame (that's what it looks like, the manual is very vague),
I figure that this combo results in about 300 units of _perceived_
motion per second!
I'm curious though, about how the JPEG compression of the PAR affects
field rendering. Does the PAR store each field seperately, or does it
combine them into one compressed frame? Whatever, it looks great when
played back.
--
Jeff Walkup - jwalkup@sfsuvax1.sfsu.edu - Digital Animator / Videographer
##
Subject: Lens flare (was RE: 3DS Vs. Imagine speed)
Date: 19 Nov 1993 17:51:24 +0000
From: "Oxley David" <oxleyd@dodo.logica.co.uk>
In article <9311182130.AA13712@email.sp.paramax.com>, Breno A. Silva wrote:
>BTW, I hope Imagine 3.0 include Lens Flare.
They've put a lens flare *TEXTURE* in 2.9, that renders sort of
semi-transparent star shapes in place of the object to which the texture is
attached. I suppose if you made a lightsource object in the detail editor, you
could add a lens flare texture to it. I haven't tried this yet, and I wonder
if the flare would re-orient itself to face the camera as long as the light was
in the camera's field of view, regardless of how much camera/light movement
there was...if you see what I mean.
I hope the lens flare texture is just a stop-gap until they implement lens
flare properly in 3.0. By properly, I mean something like the way Lightwave
seems to implement it.
Regards, David Oxley. <oxleyd@logica.co.uk>
##
Subject: FW: Long Render Times
Date: Fri, 19 Nov 93 09:29:00 PST
From: Stethem Ted 5721 <TedS@ms70.nuwes.sea06.navy.mil>
----------
From: Rick Rodriguez
To: TedS
Subject: RE: Long Render Times
Date: Thursday, November 18, 1993 10:47PM
Imagine IS optimized for the '040. It has been for a very long time.
Please stop repeating this ridiculous assertion. For confirmation,
call Impulse tomorrow and speak directly to Mike Halvorson.
--Rick Rodriguez
##
Subject: Re: Motorola FTP site? (was RE: Long Render Times)
Date: Fri, 19 Nov 93 15:35:09 EDT
From: Eric Chet <echet@magnus.acs.ohio-state.edu>
>
> >I believe the FP optimization AIBB does for the 040 is inline in software any
> >exception traped instructions that are not included in the 040 FPU. This saves
> >in the exception trap time. I believe on the motorola ftp site it's possible
> >to get the FPSP source. It would be great if SAS had a option to inline all
> >040 FPU instructions that are exception trapped. I know in SAS/C 6.50 it does
> >instruction reordering for the 030/040/882.
>
> Where is this Motorola FTP site? What's there?
>
> --Andy Church
>
>
I have not found a site that mirrors the Motorola BBS totally. This is the
information I have.
Motorola BBS: 512-440-2725
FPSP source from the Motorola BBS and have converted it
to GNU GAS format. The source is available by anonymous ftp from:
ftp.usask.ca /pub/software/fpsp/fpsp_gas.tar.gz
Motorola mirrors
ernie.uvic.ca
suniams1.statistik.tu-muenchen.de /incoming/motorola /pub/motorola
bode.ee.ualberta.ca
calvin.stanford.edu
ee.utah.edu
red.cs.tcu.edu
68k sites
cabrales.cs.wisc.edu
lucy.ifi.unibas.ch
Hope this helps. Eric
##
Subject: Re: Field Rendering
Date: Fri, 19 Nov 93 16:06:04 EST
From: David Watters <watters@cranel.com>
> I'm curious though, about how the JPEG compression of the PAR affects
> field rendering. Does the PAR store each field seperately, or does it
> combine them into one compressed frame? Whatever, it looks great when
> played back.
The PAR saves each frame (2 fields). Field rendered animations look totally
spectacular on the PAR!
_ ___
David ~ |_|,--' |@,__
Watters ~ ( )-_______-()`-
--
David R. Watters (watters@cranel.com) Cranel Inc. Development & Engineering
"Porsche. The very name is, to many, the last word in sports cars. Any car
blessed with these magic seven letters is sure to be the very best. Period!"
- Car and Driver, January 1993
##
Subject: Re: Field Rendering
Date: Fri, 19 Nov 1993 12:02:24 -0800
From: kkalnasy@bvu-lads.loral.com (Kent Kalnasy)
At 9:44 AM 11/19/93 -0600, Thomas Setzer wrote:
>> 1] Although this can't be done within Imagine, has anyone experimented with
>> doing so through one of the image processing programs using arexx?
>
>I thought 2.9 handled this? isn't that why this discussion came up? Did I
>misread again?
3.0 is supposed to, but I don't think 2.9 does this yet. Or did I just
miss it? Regardless, it's one of the features advertised for 3.0.
>> 2] If so, since there would be no time saved during rendering, would the
>> savings be passed on to the actual memory used for an animation?
>
>Whoops, you missed something. There is a time savings. I didn't notice till
>today, but a previous post kinda misstated something
>
>> we rendered fields like this: and displayed them like this:
>>
>> Frame 1 2 3 4 5 6 Frame 1 2 3 4 5 6
>> +---+---+---+---+---+-- +---+---+---+---+---+--
>> Field A A A A A A Field A A A A A A
>> Field B B B B B B Field B B B B B B
>>
>> Field rendering allows us to and display them as always, but
>> render fields like this: without introducing artifacting:
>>
>> Frame 1 2 3 4 5 6 Frame 1 2 3 4 5 6
>> +---+---+---+---+---+-- +---+---+---+---+---+--
>> Field A A A A A A Field A A A A A A
>> Field B B B B B B Field B B B B B B
>>
>The above isn't really true, as another poster pointed out.
>Its more like this(let me try my hand at ascii):
I don't think I misstated anything here. It was more a matter of
semantics, I believe.
>> we rendered fields like this: and displayed them like this:
>>
>> Frame 1 2 3 4 5 6 Frame 1 2 3 4 5 6
>> +---+---+---+---+---+-- +---+---+---+---+---+--
>> Field A B C D E F Field A B C D E F
>> Field A B C D E F Field A B C D E F
This is confusing. How are you distinguishing between the first and
second field of each frame? Maybe we're miscommunicating the idea of
fields. One field is all of the odd display lines and the other is all
of the even display lines. On televisions and interlaced monitors these
are displayed alternately. On non-interlaced monitors the beam paints
each display line in sequence, but generally twice as fast. The idea
of fields does not apply to non-interlaced displays. On televisions
and other interlaced displays you get one entire frame every 1/25 (PAL)
or 1/30 (NTSC) of a second, and each field of that frame takes 1/50
(PAL) or 1/60 (NTSC) of a second to display.
Sorry if this is obvious and for the generalizations. I'm just trying
to figure out where our disconnect is.
>> Field rendering allows us to
>> render fields like this: And display them like this:
>>
>> Frame 1 2 3 4 5 6 Frame 1 2 3 4 5 6
>> +---+---+---+---+---+-- +---+---+---+---+---+--
>> Field A C E G I K Field A C E G I K
>> Field B D F H J L Field B D F H J L
>
>See, twice as many fields, for the same price! This adds smoothness to the
>animation.
Whether you call the first two fields 1A and 1B (as I did) or A and B
(as you do) there are still two fields per frame. The earlier discussion
resulted from different ideas being represented. I was trying to explain
field rendering in general while the other poster was describing how field
rendering saves you time over rendering a frame at a time, throwing away alternate fields and merging the remaining fields to achieve the same
effect as field rendering.
Field rendering doesn't give you any more fields. It's not a matter of
interlaced vs. Amiga non-interlaced display modes. As a matter of fact,
this entire discussion presupposes that we're talking about interlaced
display modes, and focuses on what image is displayed in the second
field of each frame.
You know, if we had non-interlaced displays as our target, the whole
issue of field rendering goes out the window. With televisions being
the prime target though, field rendering remains very important.
------------------------------------------------------------------------------
Kent Kalnasy kkalnasy@bvu-lads.loral.com
Loral Advanced Distributed Simulation, Inc.
Bellevue, Washington (206) 957-3278
------------------------------------------------------------------------------
##
Subject: Re: 3DS Vs. Imagine speed
Date: Fri, 19 Nov 1993 16:24:30 -0600 (CST)
From: kalb0003@gold.tc.umn.edu
On Thu, 18 Nov 1993, Breno A. Silva wrote:
> BTW, I hope Imagine 3.0 include Lens Flare. I think it was one of the
> best (if not THE best) additions of LW 3. With the stage editor inter-
> activity AND the screen refresh speed up I was waiting, LW will rest a bit,
> since I prefer the Imagine interface (except the animation of 2.0 that's
> now GREAT!).
>
> Breno A. Silva (INF02@BRUFSE.BITNET)
Here's what they did. One of the many new textures that they
included is a lens flare texture. In my opinion it is a lame attempt to
satisfy Light Wave admirerers. It must be applied to a disk and placed
behind the light. It is a little better than some of the Aladdin 3D pics
I've seen, but not by much. Not very imaginative.
Maybe the Apex magicians could find a solution?
##
Subject: good grain and knotholes in aged wooden timbers ?
Date: Fri, 19 Nov 1993 18:47:22 -0500 (EST)
From: "JOSEPH F. HART" <VISHART@ubvmsb.cc.buffalo.edu>
Greetings....
I am working on a project involving wood with knotholes in it. I
have noticed that the supplied wood texture generates a long grain
that is rather too smooth and straight for my purposes. I would also
like to try to get some realistic looking knotholes in what appears to
be aged and crudely cut timbers. Does anyone have any good
suggestions on perturbing the grain and adding knotholes ?
___________________________________________________________________
| Internet: VISHART@ubvms.cc.buffalo.edu
Joe Hart | /// Plink: OSS542
Niagara Falls, NY | \\\/// Ham call: WA2SND
| \XX/ AMIGA - Computers for REAL MEN
===================================================================
##
Subject: Re: 2.9 Particle Question
Date: Fri, 19 Nov 1993 18:23:04 -0600 (CST)
From: Schumacher Gordon C <gschumac@ux4.cso.uiuc.edu>
> As far as the rendering times....I have not gone crazy with
> particles but they seem to render at a fair pace. If i recall though the
> manual warns one about using to much since rendering 10,000 dodecahedraons
> will have 10*10,000 faces etc....
That's something important to keep in mind... BTW, anyone know how come
it goes quickly? This sort of thing in 2.0 would be *horrendously* slow...
[NitPicker v3.02] Actually, a dodecahedron has 12 faces...
Sorry for not changing the To: line on my first send, BTW... I forgot!
--
Gordon Schumacher
/-------------------------------------------------------------------\
| Champaign- "We apologize for the inconvenience." _@_ |
| Urbana -HHGTTG / \ |
| kilroy was here | o o | |
\-------------------------------------------------------U|--U--|U---/
##
Subject: Imagine Lens Flare
Date: Sat, 20 Nov 93 0:07:53 EST
From: woovis@jcnpc.cmhnet.org (William V. Swartz)
I gotta say I kind of like this implementation of the 'Lens Flare Texture' you
guys are describing in 2.9, but you guys seem to be knocking it. Seems to me
it will come in handy to be able to pick which light sources cause flaring as a
scene with many lights placed for effect would look rediculous with say, flares
crawling up a wall behind a table lamp! This may not be the best example of
what I mean and it may be a false assumption on my part that all lights in LW
will flare if that option is enabled but still I would not be so quick as to
trade this method for LW's.
Methinks I will go ahead and order so I can play with 2.9 as I am getting a
feeling of longer delays on 3.0! (Not really, I just want to see the new stuff
bad enough now from reading everyones comments!)
//
\X/ -BiL-
woovis@jcnpc.cmhnet.org (See my 'Imagine'-ary signature below)
##
Subject: 2.9 Startup
Date: Sat, 20 Nov 93 07:54:09 GMT
From: Andrew Nunn <apn@moby.demon.co.uk>
Remember the days of the long pause while the title screen loaded? In
2.9 if Imagine.pic is missing from the directory, the program jumps
straight to the project editor.
Andrew
##
Subject: Re: FW: Long Render Times
Date: Fri, 19 Nov 1993 23:59:47 -0800 (PST)
From: Steve Lopez <lopez@cyberspace.com>
On Fri, 19 Nov 1993, Stethem Ted 5721 wrote:
> From: Rick Rodriguez
> To: TedS
> Subject: RE: Long Render Times
> Date: Thursday, November 18, 1993 10:47PM
>
> Imagine IS optimized for the '040. It has been for a very long time.
> Please stop repeating this ridiculous assertion. For confirmation,
> call Impulse tomorrow and speak directly to Mike Halvorson.
> --Rick Rodriguez
I'm Inclined to agree...It really cooks on my 2000 w/GVP 040 '33 I have
averagespeeds of 1-2 mins per frame, and the longest I've been able to
slow it is about 15 mins per frame. Although, It's not nearly as fast on
my friends 4000, I wonder why? I think maybe the people having trouble
might try diddling with the prefs a bit...I changed the Octree settins a
bit when I first got it...just fooled with it, and now it's faster. Good
luck!
Only on AMIGA!!
/----------------------------------------------\
/ /// Steve Lopez -> lopez@cyberspace.com \
/ /// Student at -> Art Institute of Seattle \
/ \\\/// Program -> Audio & Video Production \
\ \XX/ `040' Computer -> Amiga 2000 GVP040'33 DCTV /
\ : DSS8+ 9megs Flicker fxr /
\ "I can do that..." : Midi DJ500c Supra 14.4k /
\----------------------------------------------/
##
Subject: Re: trace vs scanline
Date: Sat, 20 Nov 93 08:23:18 EST
From: marino@mindvox.phantom.com (Paul Marino)
Ted,
It sounds like your running into the same wall that we've all run into at one
time or another.
Lots of ambition, no speed.
Well, my solution would be to composite your animation.
What I mean by that is, if your camera doesn't move throughout the animation
you can generate ONE still image of your Atlantis scene, and then
scanline render only the animated objects.
This can be done easily by generating that first frame (Atlantis) and then use
it as your background frame within Imagine.
It sounds as if their are other ways around this prob as well...(Complex anims
often have complex solutions....I've been through this plenty of times)
Leave some "E" on how you made out...
Your fellow Imagineer,
-Paul Marino.
##
Subject: Re: Imagine 2.9 - It's HERE!
Date: Sun, 21 Nov 1993 01:41:10 +1100 (EST)
From: Nikola Vukovljak <nvukovlj@ucc.su.OZ.AU>
On Tue, 16 Nov 1993, Michael B. Comet wrote:
>
>
> >
> >You mention editing in Detail editor in perspective. Can you actually
> >move points and faces or not ? This would be way cool if possible.
> >
> >Nik.
> >
> >
> >
> >
>
> Yes, it draws the points faces etc in perspective, and you can
> select and move them in the perspective window with clicking, bounding box
> and lasso etc...
>
> Pretty neat.
Alright! I can't wait to get it!
Should be early next week hopefully.
Nik.
##
Subject: RE: Long Render Times
Date: Sun, 21 Nov 1993 02:12:45 +1100 (EST)
From: Nikola Vukovljak <nvukovlj@ucc.su.OZ.AU>
On Thu, 18 Nov 1993, Steve Koren wrote:
>
> > could express very deeply that 3.0 should be optimized for the '040. Or
>
> If they are using SAS/C (good chance they are, I'd guess), then before
> version 6.50 you could optimize for the 68030 (which used 030 op codes),
> but there was no specific instruction scheduler for the 040. As of
> 6.50, there is; it reorders instructions to try to get maximum
> parallelism from the CPU. (Certain combinations of instructions can be
>From what a friend has told me who looked at the code, Impulse is using
Manx C. Now, Manx C is no longer available for the Amiga. Also, I think
that Impulse have indicated that they are now doing the software writing
on the PC. So, this creates 2 problems these days as far as optimization
goes.
Nik.
##
Subject: 3D anim software rated
Date: 17 Nov 93 11:28:00 -0800
From: Ed_Totman@ucsdlibrary.ucsd.edu
<Thanks to all who replied about backgrounds and trace times.>
Get the Nov 93 special issue of NewMedia magazine. The following
software packages are compared. I would say that Imagine rates
favorably in the comparison:
3d reality adobe dimen. aladdin 4d animatrix mod. ark geom.
3d studio caligari 24 caligari brdc caligari wind cheetah 3d
crystal 3d crystal topas dynaperspectve electricimage envisage 3d
explore frontface gig 3dgo IMAGINE 3.0 infini-d
lightwave macromodel pwr animator presenter prof ray dream 5.0
real 3d v2 realize rend renderize live sculpt 3/4d sculptor
showplace sketch! softimage env stratavision studiopro
swivel 3d adv visualizr three-D upfront vertigo 9.5
walthrough win visual reality will vinton's playmation
One of the catagories is "alpha channel compositing", which is not
supported in Imagine. Is this related to "field rendering?".
What do these two terms mean and how do they relate to 3d
animation?
##
Subject: Re: Sphere primitive doesn't work
Date: Wed, 17 Nov 1993 16:12:13 -0600 (CST)
From: Daniel Jr Murrell <djm2@Ra.MsState.Edu>
>
>
>
> I've got my 2.9 PAL upgrade a couple of days
> ago. I noticed that it impossible to apply
> Deformations or different Pick/Add modes to
> the sphere primitive (Function/Add/Sphere).
> Anyone encountered similar problems?
No, you can't deform the sphere primitive. It doesn't have any real
points to deform around. Deformations only work when there is a point to
pivot/twist/whatever around. Try the faceted sphere instead.
A question:
I was going to wait until 3.0 is out to upgrade. Y'all think I ought
to go ahead and do it now? I'm getting really jealous listening to the
stories about the particle systems, deformations, real-time perspective
window updating, etc.
>Jacek
Danimal
##
Subject: Alpha Compositing
Date: Sat, 20 Nov 93 11:22:58 EST
From: marino@mindvox.phantom.com (Paul Marino)
Sorry to be so behind the times...(my email had Ted's first post in front
of all the responses)
Ted, (& all others who are interested...)
I'm sure by now you've tried this method of rendering the
still background image, you might want ot use an alpha channel render
to composite the animation (Jason Birch mentions this in his post)
Its actually quite simple, however you'll need ADPro to help you...(If
you don't have ADPro, I strongly suggest you RUN out and buy a copy right
away...)
In Imagine...
1) Render your animated submarine (or spaceship, or airplane,
or mother-in-law, etc.) by itself.(preferrably over a black background)
2) Now go into the detail editor and load the object(s) that are moving
in your animation.
3) Take off ALL of the attributes that you have applied to the object.
---
4) Make the object(s) 255 white and click on BRIGHT. (This makes your
object fully shading independent).
5) Save this object(s) as another name (I usually end with the suffix ".alf")
6) Exit the Detail editor.
So far, this might sound a bit unusual but stay with me its worth the trouble.
6) Using the CLI or WB (or any other directory util you have)
Copy the Project you rendered the animated moving object. Again,for
consistency and clarity label the new project "X_Alf.imp"
7) Go back into Imagine and open up your new "X_Alf" project.
8) Enter the Action Editor. Locate your moving object(s) and replace its
name with the new white object(s).
9) Render this white object animation.
Now, what you've just created is an animation of this white object(s) which,
when rendered, will be pixel-to-pixel to the animated colored object(s).
You may now ask "Why in the world do I need this?".
We created this animation so that we can composite (there's that word again)
our object on top of a scene that on its own takes hours(9 hours, TBE).
We'll use these frames in ADPro to as elements to composite with.
Now load up ADPro.
1) Load the background image into ADpro. (Atlantis?)
2) Click on the "Replc" button and make sure it switches to "Comp".
3) With the loader set to "Alpha", open the first picture of
your first animation.
4) After this picture loads in, a "composition control" requester will appear.
Click on "O.K.". (FYI - this requester allows you to manipulate the
compositing of images, we dont need to use it here however.)
5) After leaving the "composition control" requester, a SECOND file requester
will appear asking for an alpha channel file. This is where our white object
frames come in. Click on the first frame of that animation.
**** NOTE: the white object frames need to be first converted to ****
**** grayscale (8 bit) before they can be used by the alpha loader ****
ADPro then examines the white object frame. It lays in the color object
over the background image, using the white object frame as reference for
transparency. (full white->opaque, full black-> totally transparent)
6) After a few seconds... Hooray! You've just composited a frame!
This gives you quite a bit of control over rendering.
Question:
What happens if you have objects that don't move in front of
your moving animation? How do I get my object to go behind it?
Answer:
Easy. In your white object animation include those objects as well. Only
make those black - 0 black. When ADPro gets to the frame in which the white
object goes behind the black,it will only place the color object (your first
animation) where white exists. Thus creating "foreground" with
of your background image.
BTW, ADPro uses FRED to apply this technique over sequential frames.
Also, Imagine 3.0 will supposedly have alpha channel support. So having to copy
projects for alpha passes SHOULD be a thing of the past.
Hope this answers more questions than it creates....
Your friendly NYC Imagineer,
Paul Marino.
P.S. Sorry for eating up so much space...
##
Subject: Yellow Crystal Sphere
Date: Mon, 22 Nov 93 11:12:24 CST
From: drrogers@camelot.b24a.ingr.com (Dale R Rogers)
Hello,
I'm trying to figure out the lighting and attribute settings to
make a transparent crystal ball with a bright yellow tint.
I have a acheived the transparency effect fine by playing with the
filter value, however I'm have a hell of a time getting the color
to show as yellow. I have set the attribute values to the
following.
Color: R=255
G=255
B=0
Reflectance: R=1
G=1
B=1
Filter: R=255
G=255
B=0
Hardness is 255, Shininess=0, Specular is a little brighter than
color. I have a global brush map of sky colors for some metalic
elements that will come into the animation later.
The sphere tends to take on the colors of the global brush map as
opposed to the color of the sphere as defined in its attributes.
the specular highlights show as the color I need.
I thought perhaps I needed a light behind it to illuminate the
glass. So I added a cylinrical light, tracked it to the sphere,
and placed it behind and above the sphere. I'm still getting a
purpleish sphere, as opposed to a yellow sphere.
Any ideas?
The sphere looks great. The transparency is great. I just need
it to be yellow. What am I missing? I have practically memorized
Understanding Imagine 2.0 (page 36 I think) about understanding
the attribute settings. I still must be missing something.
Dale
____________________________^____________________________
dale r. rogers
Intergraph Corporation
Building Design & Management MailStop: LR24A4
drrogers@b24a.b24a.ingr.com Tel: (205) 730-8294
.
##
Subject: Re: Sphere primitive doesn't work
Date: Mon, 22 Nov 93 12:40:04 CST
From: drrogers@camelot.b24a.ingr.com (Dale R Rogers)
|The sphere primitive (as in 1.1) is a mathematical construct, not a set of
|discrete points. The wireframe you see is just an approximation. Use the
|Sphere listed with the other primitives, where you can define the point
|count, and you'll be able to edit it.
Is this function in 2.0? Also, would it take less time to render
due to the absnce of points?
I have a crystal ball that I am using in an animation. I need to
assign it various attributes, but will not morph it, or change
it's shape at all. It takes a long time to render because I
created it with a lot of faces in order to get the smooth surface.
Decreasing the number of points would speed things up
considerably.
Dale
##
Subject: Re: Imagine Lens Flare
Date: Mon, 22 Nov 93 11:22:15 EST
From: Mark Thompson <mark@westford.ccur.com>
> I gotta say I kind of like this implementation of the Lens Flare Texture you
> guys are describing in 2.9, but you guys seem to be knocking it. Seems to me
> it will come in handy to which light sources cause flaring
LightWvae gives you full control of flare parameters for every light source
individually. Only those lights you choose will have any flare effects. The
implementation and results are vastly superior to any other implementation
I have seen in other 3D programs (including 3D Studio). I have to say that
Imagine's implementation is rather lame.
%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~%
% ` ' Mark Thompson CONCURRENT COMPUTER %
% --==* RADIANT *==-- mark@westford.ccur.com Principal Graphics %
% ' Image ` ...!uunet!masscomp!mark Hardware Architect %
% Productions (508)392-2480 (603)424-1829 & General Nuisance %
% %
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
##
Subject: Re: Field Rendering
Date: Fri, 19 Nov 93 12:05:30 EST
From: Mark Thompson <mark@westford.ccur.com>
> The savings come into play when someone might have done this manually.
However, even when done automatically by the software, there is a render time
penalty for field rendering. Even though the same number of pixels are
being computed per frame, the presence of two time shifted fieds means that
all the initization calculations most be done twice for each frame. This
includes traversing the database, all coordinate transformations, computing
shadow maps, etc. However, generally this is not a huge impact.
> 60 fields/s has got to be a lot smoother than 30 frames(30 fields)/s.
It is MUCH smoother, especially when there is a lot of motion right in front
of the camera.
> The PAR handles this, doesn't it? Someone should try it, let us know the
> difference.
The PAR fully supports fields and frames. The difference can be tremendous
when there is a lot of frame to frame motion. The individual frames look
pretty crappy because of the time separation of the fields, but the resulting
animation very smooth. If you take a look at the individual frames of my
Creature animation I uploaded, you will note that they are field rendered
(and some are also motion blurred).
%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~%
% ` ' Mark Thompson CONCURRENT COMPUTER %
% --==* RADIANT *==-- mark@westford.ccur.com Principal Graphics %
% ' Image ` ...!uunet!masscomp!mark Hardware Architect %
% Productions (508)392-2480 (603)424-1829 & General Nuisance %
% %
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
##
Subject: Re: Sphere primitive doesn't work
Date: Mon, 22 Nov 93 15:36:14 CST
From: setzer@comm.mot.com (Thomas Setzer)
>
> |The sphere primitive (as in 1.1) is a mathematical construct, not a set of
> |discrete points. The wireframe you see is just an approximation. Use the
> |Sphere listed with the other primitives, where you can define the point
> |count, and you'll be able to edit it.
>
> Is this function in 2.0? Also, would it take less time to render
> due to the absnce of points?
>
Yes. Yes(actually there is probably a different algorithm for the intersection
of a ray(ray-tracing) for the math sphere which causes the speed up)
> I have a crystal ball that I am using in an animation. I need to
> assign it various attributes, but will not morph it, or change
> it's shape at all. It takes a long time to render because I
> created it with a lot of faces in order to get the smooth surface.
> Decreasing the number of points would speed things up
> considerably.
Yes, use the sphere if you can. I think it is under the add in the function
menu. Not under primitives but on the same level as primitives and axis etc.
Tom Setzer
setzer@ssd.comm.mot.com
"And of course, I'm a genius, so people are naturally drawn to my fiery
intellect. Their admiration overwhelms their envy!" - Calvin
##
Subject: Re: Yellow Crystal Sphere
Date: Mon, 22 Nov 93 15:32:18 CST
From: drrogers@camelot.b24a.ingr.com (Dale R Rogers)
|> Filter: R=255
|> G=255
|> B=0
|
|I *think* you have to set:
|
| Filter: R=0
| G=0
| B=255
|
|because you want to filter out the blue, and leave the yellow. You're
|filtering out the yellow and leaving the blue.
Interesting. I could have sworn the book said you are setting the
value that you want to want to pass through. It was talking about
a stained glass effect. Therefore, if you wanted a tinted glass,
you would set the value to the color that you wanted the color to
be. Maybe I just had a brain fart and misinterpreted the
paragraph. I'll double check.
Thanks,
Dale
____________________________^____________________________
dale r. rogers
Intergraph Corporation
Building Design & Management MailStop: LR24A4
drrogers@b24a.b24a.ingr.com Tel: (205) 730-8294
.
##
Subject: Re: Sphere primitive doesn't work
Date: Mon, 22 Nov 1993 16:26:00 -0600 (CST)
From: Schumacher Gordon C <gschumac@ux4.cso.uiuc.edu>
> I have a crystal ball that I am using in an animation. I need to
> assign it various attributes, but will not morph it, or change
> it's shape at all. It takes a long time to render because I
> created it with a lot of faces in order to get the smooth surface.
If I am not mistaken, though, it would be impossible to morph the sphere
primitive - for the exact reason that it *has* no points.
Anyone tried this?
--
Gordon Schumacher
/-------------------------------------------------------------------\
| Champaign- "We apologize for the inconvenience." _@_ |
| Urbana -HHGTTG / \ |
| kilroy was here | o o | |
\-------------------------------------------------------U|--U--|U---/
##
Subject: Work area size
Date: Mon, 22 Nov 93 15:02:40
From: greg.tsadilas@hofbbs.com
About a week I said that Imagine 3.0 will allow you to define a larger
work area. This was based on a conversation I had with Mike Halvorson.
Since then I found that I misunderstood. Apparently the ACTION editor
will have more room in it. The editors will remain as-is size-wise.
GreG
##
Subject: screen size bytes
Date: Mon, 22 Nov 1993 20:20:02 -0500
From: "Mr. Scott Krehbiel" <scott@umbc.edu>
Did any9one post the locations of the numbers to change the
screen size? Can someone mail me that information please??
I'm talking about the hack to let Imagine run on other graphics
boards.
thanks
Scott Krehbiel
scott@umbc4.umbc.edu
##
Subject: RE: imagine 2.9
Date: Mon, 22 Nov 93 21:53:13 GMT
From: Waland J F <walaj@essex.ac.uk>
Hi, I've just joined the list, have imagine 2.0 (from Amiga Format, here in
the UK) and am waiting for my copy of Imagine 3.0 (#95 not #75 as stated in the
magazine !!! -should cheer all those who are upset about the pricing structure)
anyway, I have seen references to imagine version 2.9 - what is it and what
does it do (especialy what is new compared to 2.0)
thanks in advance,
jon
##
Subject: Envisage 3D Competive Upgrade Offer from Imagine
Date: Mon, 22 Nov 93 21:08:12 -0600
From: "Tech Support" <wk00725@worldlink.com>
TO: IML Imagineers
FROM: Byte by Byte Corporation, a leading provider of professional 3D
modeling, rendering, and animation software. Developers of the
Sculpt 3D and Sculpt-Animate 4D software for the Amiga as well as the
Sculpt series of 3D software for the Apple Macintosh. Recently
released Envisage 3D for the MS-DOS platform.
PRODUCT: Envisage 3D is a professional 3D modeling, rendering, and
animation software package running on MS-DOS based computers.
The product lists for $995 and has been shipping since August 1993.
SPECIAL OFFER: Limited offer to acquire Envisage 3D for $500, a 50% savings.
30 day money back guarantee if not satisfied.
WHO QUALIFIES: First 50 Imagine owners with proof of Imagine ownership.
Any owner of Sculpt-Animate 4D for the Amiga (unlimited)
VERSION: The commercial 1.1 version of Envisage 3D complete with in depth
documentation.
HOW TO ORDER: Contact Byte by Byte Corporation
Voice: (512) 795-0150
Fax: (512) 795-0021
BBS: (512) 795-0032
Internet: wk00725@worldlink.com
PAYMENT METHODS: Mastercard, VISA, Cashier's Check, Company Check
Personal checks may require ten days to clear.
ENVISAGE 3D FEATURE LIST
Modeler: Workstation grade, hybrid spline/vertex editor with high speed redraw
Four view using plan, elevation and interactive perspective view
Individual models up to 7 million polygons, edges, or faces
Automatic construction of models from plan and elevation profiles
Deformation modeling with Shaper and Magnet tools
Boolean subtractive modeling and intersection
Lofting of non-parallel cross-sections with different vertice counts
Welding: seam and crease specification and control
Built-in procedural spirals and helices
Path extrusion with size changes
Type in ATM fonts
Automatic fill of enclosed polygons or to a specified point
Automatic beveling and roundovers
Height fields using image maps to create models, terrains, etc.
Mathematical surface generation
Various explodes and shatters
Equal facility with organic or geometric forms
Interactive real time rotation and scaling
Selection by any attribute including colors, textures, etc.
Tracing of GIF images
Hierarchical naming and linking of models
Rendering: High speed Phong rendering - render 1 to 3 minutes per frame
Image mapping with up to 32 maps per model
Bump mapping
Reflection and environment mapping
Glass, mirrors, and reflections
Procedural textures with preview
Automatic Shadow casting
Unlimited number of lights
Glows
Ambient, point, spot - lights selectable with or without shadows
Hidden line wireframe with adjustable detail resolution
Rich high resolution renderings up to 5,000 x 5,000 pixels
Five selectable levels of anti-aliasing
Generation of FLI and FLC files
Optional rendering modules for SGI, Sun, DECStation, and VAX workstations
Animation: Professional timeline editor
Interactive real time frame control
Real time preview without need for pre-calculation
Skeletonal animation
Comprehensive morphing
Dynamic fluid surfaces (ripple and oceanic wave generation)
Animated ground plane
Animated FLI and FLC textures for movies within movies
Graphic velocity and position controls for ease-in and ease-out
Interactive spotlight and camera cone tracking
Multiple cameras
User customizable bounding boxes
Automated special effects (modulation, springs, shaking, etc)
Open architecture for 3rd party plug-in effects and textures
Imports: DXF, ATM, Imagine 2.0, 3D Studio, GIF, TGA, TIFF, FLI, FLC
Exports: DXF, 3D Studio, GIF, TGA, TIFF, FLI, FLC
##
Subject: Re: Sphere primitive doesn't work
Date: Tue, 23 Nov 93 02:57:22 PST
From: 23-Nov-1993 0540 <leimberger@marbls.enet.dec.com>
Dale,
I was never happy with the sphere primitive. If you get close
you will see it is not as detailed as the other. It renders faster
and there may be times you can take advantage of this. I recently
played with an anim that simply took the title I use for my
hobby work(never have time to do real work) and morphed it from
a sphere into three lines of test. Well when I conformed the text
to a sphere it was fairly ragged. I encased this inside another
sphere for the first frame. Then I dropped the outer sphere, and
started the explode in the second frame. Because of the speed of
it happening you are hard pressed to tell there are seperate
objects involved. In a case like this I may get away with the
primitive sphere(but I diden't bother for 1 frame). However if
your close up and locked on your sphere you probably can't.
BTW I recently modeled a crystal ball. It sits in a fancy
gold ring with three prongs to support it. The ring sets on a
wooden base Imade with the sweep function. It's not to bad at
all. I scaned in the parts used for the prongs as opposed to
trying to model them. As for color, Your always going to pick
up the global map. If you turn down some of the filter settings
you loose tranparency. Is it possible to wrap the sphere in a
yellow brush and use essence to filter the brush?
bill
/*
bill leimberger nashus n.h.
*/
##
Subject: Re: Yellow Crystal Sphere
Date: Tue, 23 Nov 93 8:19:13 CST
From: drrogers@camelot.b24a.ingr.com (Dale R Rogers)
|Hey Dale,
| About your yellow crystal sphere, you didn't mention how
|much reflectivity you had, but you may have your specular
|settings a little high. That's probably why you're picking up
|your global colors, instead of showing your yellow sphere.
| I've also found a less intense yellow seems to show a
|warmer color. Maybe you want that cold yellow, but just for a
|lark, try easing your red back to say 220, and your green down to
|190 or 180. It really makes a golden yellow.
|
I thought I had mentioned my reflectivity. It is set to 1,1,1.
That way it would reflect the entire spectrum; but just slightly.
As far as the specularity, I thought the settings merely indicate
the color of the highlights. If the sphere is yellow, I would
want the highlights to be a slightly brighter yellow. Is this not
how the specular setting works? Why would the specular setting
influence the global parameters?
BTW, I did back off on the yellow a bit and it is warmer looking;
thanks :-)
Dale
____________________________^____________________________
dale r. rogers
Intergraph Corporation
Building Design & Management MailStop: LR24A4
drrogers@b24a.b24a.ingr.com Tel: (205) 730-8294
.
##
Subject: Statue of Liberty object
Date: Tue, 23 Nov 1993 16:25:44 +0100 (MET)
From: Lorenzo Buonomo <buonotto@cli.di.unipi.it>
Does exists an object of Statue of Liberty?
Someone can UUENCODE-EMAIL me it (if it's a PD object)
thanks in advance
Lorenzo Buonomo
EMAIL: buonotto@cli.di.unipi.it
##
Subject: script for rain effects
Date: Tue, 23 Nov 93 08:52:21 -0700
From: Steve Koren <koren@hpfcogv.fc.hp.com>
Hi,
Here is a little ARexx script I wrote to accomplish adding rain effects
to rendered images (from anything; doesn't have to be Imagine). The
script requires ImageFX (which no self respecting 3D artist should be
without anyway :-) ). It works best if your scene has the WaterDrops
texture from EssenceII applied to surfaces where appropriate. The final
rain ends up looking similar to that in Mark Thompson's (cool) "Store"
image. Its a decent effect for post-processing, although naturally the
rain won't end up reflecting the lighting conditions in your scene
unless you use an alpha channel to make it do so. This is also not
useful for animation. If someone wants to enhance the script to do
animated rain, feel free. Shouldn't be too hard. You can also use an
alpha channel in ImageFX to apply rain to only certain parts of your
picture.
You get to choose about 12 parameters of the rain. The script will pick
reasonable defaults based on your image width and height.
BTW, as long as I'm on the topic of ARexx, why oh why won't Impulse add
an ARexx port to Imagine? It would be just *incredibly* useful. They
don't have one in 2.9/3.0, do they? The things you could do combining
Imagine & other graphics tools with ARexx would be incredible, if only
Impulse would do a little work on an ARexx port. We *are* in the 90's,
after all.
- steve
~~~ cut here ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/*sys:rexxc/rx
*****************************************************************************
* *
* ARexx program to simulate rain effects in the current ImageFX buffer. *
* *
* Author: Steve Koren *
* Date : 20-Nov-93 *
* *
* This code prompts for the following parameters: *
* *
* length = length of rain, in pixels *
* vary = rain length variation, in pixels *
* width = rain width, in pixels, 1-8 *
* blend = rain blending with background (0=transparent, 100=opaque) *
* skew_pc = percent slant variation in individual rain streaks *
* (this is best small) *
* density = rain density (0=none, 100=a whole lot) *
* slant = slant bias of all rain streaks (-40 to 40 or so) *
* vwidth = set to vary width of rain streaks *
* vlength = set to vary length of rain streaks *
* r = rain color, r, 0-255 *
* g = rain color, g, 0-255 *
* b = rain color, b, 0-255 *
* *
* Defaults are chosen for all values. The following values are chosen *
* based on the size of the image: *
* *
* length *
* vary *
* width *
* *
*****************************************************************************/
OPTIONS RESULTS
/* -- if an ImageFX is not running, complain ------------------------------ */
if pos('IMAGEFX.1', (show(ports))) = 0 then do
say "ImageFX server not running."
exit 5
end
/* -- Get info about current buffer --------------------------------------- */
GetMain
if result = "" then do
RequestNotify "Operation requires a buffer."
exit 0
end
PARSE VAR result buf.name buf.width buf.height buf.depth .
/* -- set defaults -------------------------------------------------------- */
rain.length = buf.height / 14
rain.width = buf.width / 160
rain.blend = 45
rain.vary = buf.height / 20
rain.skew_pc = 2
rain.density = 20
rain.slant = 10
rain.vwidth = 0
rain.vlength = 1
rain.r = 115
rain.g = 109
rain.b = 102
/* -- request parameters from user ---------------------------------------- */
RainReq.1 = 'I/120/16/Rain Length :/'||rain.length
RainReq.2 = 'I/310/16/Rain Width :/'||rain.width
RainReq.3 = 'I/120/30/Rain Blend :/'||rain.blend
RainReq.4 = 'I/310/30/Rain Vary :/'||rain.vary
RainReq.5 = 'I/120/45/Rain Skew Pc :/'||rain.skew_pc
RainReq.6 = 'I/310/45/Rain Density :/'||rain.density
RainReq.7 = 'I/120/60/Rain Slant Pc:/'||rain.slant
RainReq.8 = 'X/200/62/Vary Width/'||rain.vwidth
RainReq.9 = 'X/315/62/Length/'||rain.vlength
RainReq.10 = 'I/120/80/Rain R:/'||rain.r
RainReq.11 = 'I/220/80/G:/109:/'||rain.g
RainReq.12 = 'I/320/80/B:/102:/'||rain.b
ComplexRequest '"Rain Effect Parameters"' 12 RainReq 430 115
if rc ~= 0 then do
RequestNotify "Cancelled."
exit 0
end
rain.length = max(result.1, 5)
rain.width = min(result.2, 8)
rain.blend = max(min(result.3, 100), 0)
rain.vary = max(result.4, rain.length)
rain.skew_pc = min(result.5, 100)
rain.density = max(min(result.6, 100), 0)
rain.slant = max(min(result.7, 100), 0)
rain.vwidth = result.8
rain.vlength = result.9
rain.r = max(min(result.10, 255), 0)
rain.g = max(min(result.11, 255), 0)
rain.b = max(min(result.12, 255), 0)
rain.numlines = buf.height * buf.width / rain.length / rain.width
rain.numlines = trunc(rain.numlines * rain.density / 1000)
/* -- get old state info -------------------------------------------------- */
LockInput
Redraw Off
LockGui
DrawMode Normal; old_drawmode = result
Blend rain.blend; old_blend = result
Pen 0 rain.width; PARSE VAR result old_pen.type old_pen.size
EdgeMode FeatherIn rain.width; PARSE VAR result old_e.m old_e.r
GetPalette 0; PARSE VAR result oldp.r oldp.g oldp.b
SetPalette 0 rain.r rain.g rain.b
UnLockGui
/* -- draw a bunch of lines ----------------------------------------------- */
randu(time(seconds)) /* set seed */
BeginBar "Rain" rain.numlines
do i = 1 to rain.numlines
line.x0 = trunc(randu() * buf.width)
line.dy = rain.length + trunc((randu() - .5) * 2 * rain.vary)
line.y0 = trunc(randu() * buf.height) - line.dy
line.x1 = line.x0
if rain.vlength ~= 0 then do
line.x1 = line.x0 + line.dy * rain.skew_pc / 100 * (randu() - .5) * 2
end
line.x1 = line.x1 + line.dy * rain.slant / 100
line.y1 = line.y0 + line.dy
LockGui
if rain.vwidth ~= 0 then do
newwidth = trunc(randu() * rain.width) + 1
Pen 0 newwidth
EdgeMode FeatherIn newwidth
end
Line line.x0 line.y0 line.x1 line.y1
UnLockGui
Bar i
end
EndBar
/* -- restore state info -------------------------------------------------- */
DrawMode old_drawmode
Blend old_blend
Pen old_pen.type old_pen.size
SetPalette 0 oldp.r oldp.g oldp.b
EdgeMode old_e.m old_e.r
randu(time(seconds))
UnlockInput
Redraw On
Redraw
exit 0
##
Subject: Re: Work area size
Date: Tue, 23 Nov 1993 11:53:45 -0500
From: Jason B Koszarsky <kozarsky@cse.psu.edu>
> Since then I found that I misunderstood. Apparently the ACTION editor
>will have more room in it. The editors will remain as-is size-wise.
How can various graphics boards promote Imagine to larger work areas but this
can't be done with the stock chipsets?
Jason K
##
Subject: 2.9 and DCTV
Date: 23 Nov 93 09:36:00 -0800
From: Ed_Totman@ucsdlibrary.ucsd.edu
I've noticed that DCTV does not display properly from 2.9,
renders and animations included. I have to display the
images from other programs. Anyone else have this problem?
BTW, the new crumpled texture makes an excellent
bottom-of-the-pool pic/anim!
Ed
##
Subject: NTSC anim questions
Date: Tue, 23 Nov 1993 10:40:02 PDT
From: scott.pack@aldus.com (Scott Pack,SQA)
Hi Imagineers!
I have a opportunity to produce a animation for a local company for use in
a sales video. The animation they want is pretty simple, but I've never
gone to tape with a animation. So I have a few questions.
1. What is the aspect ratio of a NTSC screen?
2. What are some general rules about colors? (i.e. avoid saturation)
3. How can I field render in Imagine 2.0? (This hasn't been covered, has
it?)
4. What's the best way for me to provide my work to the video producer who
will be using a Toaster? (i.e. framestores, IFF24's, complete anim -
format?)
Thanks.
-Scott
scott.pack@aldus.com
##
Subject: Re: 2.9 and DCTV
Date: Tue, 23 Nov 1993 14:18:40 -0500
From: changc9@rpi.edu (Cedric Georges Chang)
On Nov 23, 9:36am, Ed_Totman@UCSDLIBRARY.UCSD.EDU wrote:
> Subject: 2.9 and DCTV
> I've noticed that DCTV does not display properly from 2.9,
> renders and animations included. I have to display the
> images from other programs. Anyone else have this problem?
> Ed
>-- End of excerpt from Ed_Totman@UCSDLIBRARY.UCSD.EDU
I have called Impulse about this and the bug has already been reported.
Imagine 3.0 will have this problem fixed.
Cedric
--
---------------------------------------------------------------------------
Cedric Chang Mechanical Engineer // Commodore
changc9@rpi.edu Rennselaer Polytechnic Institute \X/ Amiga 3000
---------------------------------------------------------------------------
##
Subject: Re: Field Rendering
Date: Tue, 23 Nov 93 14:03:51 EST
From: Mark Thompson <mark@westford.ccur.com>
Tom sent me this in private email, but I thought I would post to the list
in case there was anyone else who misunderstood what I was refefring to.
Thomas Setzer <setzer@comm.mot.com> writes:
> > even when done automatically by the software, there is a render time
> > penalty for field rendering. Even though the same number of pixels are
> > being computed per frame
> I'm not sure this is true. Or perhaps you misunderstood what I tried to say
> therefore causing me to misunderstand your misunderstanding.
I think you are misunderstanding what I am saying :-) I am not refering to
render speed comparisons between automatic and manual field rendering tasks.
I am saying that field rendering imparts a speed penalty over not doing any
field rendering. This is due to the doubling of the initilization calculations
which now must be done twice per frame rather than once.
> So there is a savings in field rendering, if you had previously done field
> rendering manually(rendering full frames and throwing away the appropriate
> fields).
Right, but that is not what I was refering to. That savings is much more
obvious.
> I'm not sure I see the penalty.
The penalty I am refering to is more subtle, and it is a penalty going from
NO field rendering to automatic, program generated field rendering (which
Imagine doesn't currently do). LW has been doing autmatic field rendering for
a while now, and with the status info it belches out, you can easily see how
this penalty effects overall render time. It is not unusual to see a 10-20%
performance hit with it enabled. This is the additional time spent doing all
the object and camera positional calulations a second time for the same frame
(but time shifted 1/60 of a second). However, the number of pixel calculations
remain essentially the same. Hope this clears things up.
%~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~%
% ` ' Mark Thompson CONCURRENT COMPUTER %
% --==* RADIANT *==-- mark@westford.ccur.com Principal Graphics %
% ' Image ` ...!uunet!masscomp!mark Hardware Architect %
% Productions (508)392-2480 (603)424-1829 & General Nuisance %
% %
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
##
Subject: Re: 2.9 and DCTV
Date: Tue, 23 Nov 93 13:00:47 CST
From: drrogers@camelot.b24a.ingr.com (Dale R Rogers)
| I've noticed that DCTV does not display properly from 2.9,
| renders and animations included. I have to display the
| images from other programs. Anyone else have this problem?
Yips! Please let us know if others are having the problem. DCTV
is how I preview my pics. So I need that to work before I
upgrade.
Thanks,
Dale
##
Subject: Re: 2.9 and DCTV
Date: Tue, 23 Nov 93 12:21:48
From: scotte@137.110.11.73 (Scott Ellis)
Ed_Totman said on Nov 23 :
> I've noticed that DCTV does not display properly from 2.9,
> renders and animations included. I have to display the
> images from other programs. Anyone else have this problem?
Yes, I noticed this same thing, but since I hardly ever render to DCTV format,
it's not a big problem. ;-) I always tend to be on another screen when Imagine
decides to show the Quickrender, so I get to see 1/2 second of it, then I get the
"Delete Quickrender" requester. So I write a ToolManager macro to display the
Quickrender file. ;-)
It seems Impulse has rewritten their IFF output routine, as save DCTV format
files now display properly, in hires-lace, whereas Imagine 2.0 IFF files always
wanted to display in Lores-Nonlace. Go figure. It's probably a ViewTek problem,
but Impulse DID change something.
/////////////////////////////////////////////////////////////////////
// Scott Ellis // IRC: ScottE // WARNING: This signature warps //
// sellis@ucssun1.sdsu.edu // time and space in its vicinity //
/////////////////////////////////////////////////////////////////////
##
Subject: Re: Envisage 3D Competive Upgrade Offer from Imagine
Date: Tue, 23 Nov 93 14:20:48 CST
From: setzer@comm.mot.com (Thomas Setzer)
I hate seeing ads here.:(
> FROM: Byte by Byte Corporation, a leading provider of professional 3D
> modeling, rendering, and animation software. Developers of the
> Sculpt 3D and Sculpt-Animate 4D software for the Amiga as well as the
Byte my tongue.
> Sculpt series of 3D software for the Apple Macintosh. Recently
You left out an adjective, expensive:)
> released Envisage 3D for the MS-DOS platform.
>
> PRODUCT: Envisage 3D is a professional 3D modeling, rendering, and
> animation software package running on MS-DOS based computers.
> The product lists for $995 and has been shipping since August 1993.
^^^ wow!
how much does 3DS cost?
>
> SPECIAL OFFER: Limited offer to acquire Envisage 3D for $500, a 50% savings.
^^^ closer
> 30 day money back guarantee if not satisfied.
[stuff gone]
>
>
> ENVISAGE 3D FEATURE LIST
>
> Modeler: Workstation grade, hybrid spline/vertex editor with high speed redraw
^^^^^^^^^^^^^^^^^ what is this? better that personal computer
grade, yet not as good as super computer grade? Seriously,
what is a hybrid spline/vertex object? Can you describe this
more? Are these splines rendered or do they get transformed
into vertices and edges for rendering?
> Four view using plan, elevation and interactive perspective view
> Individual models up to 7 million polygons, edges, or faces
^^how do splines fit into this count?
> Automatic construction of models from plan and elevation profiles
> Deformation modeling with Shaper and Magnet tools
> Boolean subtractive modeling and intersection
> Lofting of non-parallel cross-sections with different vertice counts
^^if this is what I think it is...cool
> Welding: seam and crease specification and control
> Built-in procedural spirals and helices
> Path extrusion with size changes
^^cool, how about rotations?
> Type in ATM fonts
^ are these postscript?
> Automatic fill of enclosed polygons or to a specified point
^fill with what? is this a triangle based renderer like
Imagine? Does it allow for polygons?
> Automatic beveling and roundovers
> Height fields using image maps to create models, terrains, etc.
> Mathematical surface generation
> Various explodes and shatters
> Equal facility with organic or geometric forms
^^whats this? Equal what?
> Interactive real time rotation and scaling
> Selection by any attribute including colors, textures, etc.
> Tracing of GIF images
> Hierarchical naming and linking of models
>
> Rendering: High speed Phong rendering - render 1 to 3 minutes per frame
> Image mapping with up to 32 maps per model
> Bump mapping
> Reflection and environment mapping
> Glass, mirrors, and reflections
> Procedural textures with preview
> Automatic Shadow casting
> Unlimited number of lights
> Glows
^^ whats this?
How about visible light sources?
> Ambient, point, spot - lights selectable with or without shadows
> Hidden line wireframe with adjustable detail resolution
> Rich high resolution renderings up to 5,000 x 5,000 pixels
> Five selectable levels of anti-aliasing
> Generation of FLI and FLC files
> Optional rendering modules for SGI, Sun, DECStation, and VAX workstations
>
> Animation: Professional timeline editor
> Interactive real time frame control
How about field rendering?
> Real time preview without need for pre-calculation
> Skeletonal animation
> Comprehensive morphing
^^^ what is comprehensive about it? Can you go from/to 2
completely different objects(unlike Imagine?)
> Dynamic fluid surfaces (ripple and oceanic wave generation)
> Animated ground plane
> Animated FLI and FLC textures for movies within movies
> Graphic velocity and position controls for ease-in and ease-out
> Interactive spotlight and camera cone tracking
> Multiple cameras
> User customizable bounding boxes
> Automated special effects (modulation, springs, shaking, etc)
> Open architecture for 3rd party plug-in effects and textures
>
Tom Setzer
setzer@ssd.comm.mot.com
"And of course, I'm a genius, so people are naturally drawn to my fiery
intellect. Their admiration overwhelms their envy!" - Calvin
##
Subject: Re: Work area size
Date: Tue, 23 Nov 93 14:01:27 -0700
From: Steve Koren <koren@hpfcogv.fc.hp.com>
> How can various graphics boards promote Imagine to larger work areas but this
> can't be done with the stock chipsets?
Good question. Related topic: a while back I noted that if you use a
mode promotion utility, Imagine gets flaky when promoting to bigger
resolutions. However, I discovered last night that if you patch the
binary to increase the resolution, you can run in anything you want and
it doesn't exhibit the same flakyness. I tried up to 1024x768, and it
works fine (assuming you are doing some generic mode promotion such as
everything->SUPER72 or everything->EGS). I think somebody else
discovered this a long time ago, but, being a slowpoke, I just got
around to trying it it.
Having the extra resolution is *really* nice, esp for complex objects in
the detail editor. The standard tiny little 640x400 just doesn't
compare. There was a note about the bytes to patch in an IML mailing
from a number of months ago. Its probably in the archives somewhere.
Grep for "bytes".
You'd figure Impulse could just modify their code and let us use the
display database. Its a trivial code change. Bet it could be done in
10 minutes tops.
Nah. That'd make too much sense. Kinda like using the actual ISL
requester instead of inventing your own for no apparent reason.
- steve
##
Subject: Imagine morphing
Date: Tue, 23 Nov 1993 21:55:00 -0500
From: roy.park@canrem.com (Roy Park)
> ^^^ what is comprehensive about it? Can you go from/to 2
> completely different objects(unlike Imagine?)
Wait a sec... I thought Imagine's morph require two objects with same number
of points and surfaces. Isn't it? Or, is this new in 2.9?
----
Roy Park
roy.park@canrem.com
##
Subject: Re: Re: Long Render Times
Date: Wed, 24 Nov 93 15:04:59 EDT
From: "Breno A. Silva" <INF02@BRUFSE.BITNET>
>slow it is about 15 mins per frame. Although, it's not nearly as fast on
>my friend's 400, I wonder why?
Hey, I thought just the 33X25 Mhz would be enough to make your render much
faster, but with the burst mode of the A4K disabled... it should be TWICE
or THRICE :) faster!
Maybe his A4k has a G-Force 040 33Mhz for the 3000 installed on his cpu
slot... Anybody already confirmed if that's possible?
Was the 68040II/66Mhz already released? The 486DX2/66 are indecently faster,
and I'm starting to urge speed... Will the WARPs ever come to us?
Lots of question... quite few answers...
Breno A. Silva (INF02@BRUFSE.BITNET)
##
Subject: Re: Imagine morphing
Date: Wed, 24 Nov 1993 02:43:00 -0500
From: j#d#.moore@canrem.com (J. Moore)
Cr> > ^^^ what is comprehensive about it? Can you go from/to 2
Cr> > completely different objects(unlike Imagine?)
Cr> Wait a sec... I thought Imagine's morph require two objects with same
Cr> number of points and surfaces. Isn't it? Or, is this new in 2.9?
No, he said, "unlike Imagine". In other words, Imagine needs the same
number of points, but does this other program?
* Q-Blue 0.93 [NR] *
##
Subject: Morphing improvement!!
Date: Wed, 24 Nov 1993 19:02:00 -0500
From: roy.park@canrem.com (Roy Park)
Someone oughtta tell Impulse that the morphying feature should be improved
in next release (post 3.0) of Imagine. I've just had it with equal-point,
equal-face object morphing!
Maybe a third party company can do this instead?
----
Roy Park
roy.park@canrem.com
##
Subject: Particle Questions
Date: Thu, 25 Nov 93 09:28:59 PST
From: dedwards@scs.unr.edu (Daniel T. Edwards)
Has anyone had any success with the particle F/X?
Remember "Particle Dreams" (an animation seen in "Beyond the Mind's Eye")
I'd like to do something like that...
I just want the particles to fall to a z plane then disappear.
I also (at the same time) want them to move in x.
Is it true that this thing (giant requestor) doesn't do parabolas?
I guess it would be easier to have a new effect: "Launch"
... where you set the Z(direction of launch) and POWER(apparent
power of launch) and LANDINGZ(a Global,or local, Z parameter for
a plane at which the particles stop) it would be nice if this
dream F/X had some dispersion and some bouncing too.
Anyway, particles look like a lot of fun if I could just write my own
F/X.
Happily using Imagine 2.9 on a 700x462 screen without special hardware...
Email me for details on how you can too.
____________________________________________________________
/ \
| Amiga 2000 James R. Walker |
| 2MB Chip dedwards@unssun.scs.unr.edu |
| 18MB Fast ______________________________________________|
| 130 MB Hard / |
| 68040 33Mhz |Heinlein,Rand,Clarke,Adams,Asimov,Niven,Worley|
\____________________________________________________________/
##
Subject: Re: Morphing improvement!!
Date: Thu, 25 Nov 1993 12:03:47 -0600 (CST)
From: Daniel Jr Murrell <djm2@ra.msstate.edu>
>
> Someone oughtta tell Impulse that the morphying feature should be improved
> in next release (post 3.0) of Imagine. I've just had it with equal-point,
> equal-face object morphing!
Hehe. Calm down Roy. Non-similar object morphing is not that easy to do. As
far as I know, not even Wavefront can morph dissimilar objects.
>
> Maybe a third party company can do this instead?
Yes, that seems more along the lines of Interchange Plus or something similar.
Having such a feature built in to Imagine would probably make the program a lot
bigger. It's not just a matter of making the face and point count match up, but
you also have to make the points and faces reasonably correspond to one another.
You might have faces passing through the object, turning inside out, etc. if say
face 123 on one object is in a totally different relative area than face 123 of
the target object. You'd have to have some kind of algorithm that would resort
the faces to be in the same general area of both objects. I'd hate to have to
program something like that, and since Imagine calculates the morphs per frame,
your rendering times would take a serious hit. I think Morphus maybe can do
something similar, but I don't know how good the results are, and I think it
makes a lot of in-between objects as well, defeating the purpose. There's
simpler things that need work in Imagine than that.
Ok, a question:
Has anyone tried that lens flare texture with 2.9 yet? How does it look, render
etc.? Do you have to trace to get the right effect? And do you attach the
texture to a light? Does it look cheesy, like the example I saw in an ad for
Aladdin? Could someone upload a few example pictures? Just wondering, and thanks.
Dan
djm2@ra.msstate.edu
##
Subject: Re: Particle Questions
Date: Thu, 25 Nov 1993 19:42:09 -0600 (CST)
From: kalb0003@gold.tc.umn.edu
On Thu, 25 Nov 1993, Daniel T. Edwards wrote:
> I just want the particles to fall to a z plane then disappear.
Can be done
> I also (at the same time) want them to move in x.
They do that too
> Is it true that this thing (giant requestor) doesn't do parabolas?
The way the particles disperse is very artificial. They have a
tendancy to form up into regular rows that look like Lisajous(sp?) curves
if used on a sphere.
> I guess it would be easier to have a new effect: "Launch"
> ... where you set the Z(direction of launch) and POWER(apparent
> power of launch) and LANDINGZ(a Global,or local, Z parameter for
> a plane at which the particles stop) it would be nice if this
> dream F/X had some dispersion and some bouncing too.
You can orient the object axis to point upwards so that the
particles disperse up and out before falling back down and bouncing.
While not nearly as powerful as Real 3D, the new particles are
still relatively diverse and usefull. The randomizing of how the
particles disperse really needs to be fixed, or a work-around found. If
particles simply fall and bounce, without wind, it looks similar to waves
of sand in the desert, or a Japanes rock garden.
##
Subject: Re: Morphing improvement!!
Date: Thu, 25 Nov 1993 20:35:52 -0600 (CST)
From: kalb0003@gold.tc.umn.edu
On Thu, 25 Nov 1993, Daniel Jr Murrell wrote:
> Ok, a question:
> Has anyone tried that lens flare texture with 2.9 yet? How does it look, render
> etc.? Do you have to trace to get the right effect? And do you attach the
> texture to a light? Does it look cheesy, like the example I saw in an ad for
> Aladdin? Could someone upload a few example pictures? Just wondering, and thanks.
I have put a copy of "flare.default.jpg" in the Imagine section of
wuarchive. It is the default lense flare settings placed on a flat, white
disk. I'm not too impressed myself.
##
Subject: PS fonts for Imagine 2.9
Date: Sat, 27 Nov 93 14:58:17 PST
From: dedwards@scs.unr.edu (Daniel T. Edwards)
Does anyone know of a good place to find PostScript type 1 fonts?
Imagine 2.9 supposedly loads and interprets them.
The pamphlet (if you can call it that) says that the proper
type of font will have an extension of ".pfb"
I've looked on wuarchive for some, but didn't find any.
Should I be looking in the Mac or Pc directories?
It looks like the spline editoe will be of great use once I
can feed it.
Happily using Imagine 2.9 on a 700 x 462 screen without special
hardware... Email me to learn how you can too.
____________________________________________________________
/ \
| Amiga 2000 James R. Walker |
| 2MB Chip dedwards@unssun.scs.unr.edu |
| 18MB Fast ______________________________________________|
| 130 MB Hard / |
| 68040 33Mhz |Heinlein,Rand,Clarke,Adams,Asimov,Niven,Worley|
\____________________________________________________________/
##
Subject: Re: Field Rendering
Date: Sat, 27 Nov 1993 19:43:00 -0500
From: charles.blaquiere@canrem.com (Charles Blaquiere)
>>1] Although this can't be done within Imagine, has anyone experimented
with >>doing so through one of the image processing programs using
arexx?
Imagemaster has a couple of functions used to combine and extract
separate fields into one interlaced image; these were created with
X-Specs 3-D glasses in mind, but they allow any program to do field
rendering.
1) Halve the vertical resolution of your renderer.
2) Double the total number of frames in the animation.
3) Start rendering.
4) Start an ARexx macro which will wait for **pairs** of frames to be
rendered, then load each one and combine them. The composite image is
what gets recorded to tape of compiled into your animation.
Unfortunately, Imagemaster has no ARexx call for its "X-Specs
Interleave" function. A call to Black Belt Systems may be useful,
although Imagine users can either use 2.9 or wait for 3.0, depending
whether field rendering is in 2.9 or not.
>>2] If so, since there would be no time saved during rendering, would
the >>savings be passed on to the actual memory used for an animation?
Correct me >>if I'm wrong, but that would seem to be a good way to
economize on memory >>usage for those creating animations to be played
back in memory, thus longer >>anims using less RAM (for those of us who
are stuck being memory conscious). >>;-)
Since you're rendering twice as many frames, but with only half the
vertical resolution, the effects nearly balance each other out. Of
course, there's additional loading time for objects in the second frame,
plus interleaving time if you found a way (or reason) to use an external
image processing program.
##
Subject: Lens flare (was RE: 3DS Vs. Imagine speed)
Date: Sat, 27 Nov 1993 19:46:00 -0500
From: charles.blaquiere@canrem.com (Charles Blaquiere)
>> I wonder if the flare would re-orient itself to face the camera as
>> long as the light was in the camera's field of view, regardless of
>> how much camera/light movement there was?
Just add an Align bar to the flare object, and track it to the camera.
That, and making the object Bright to shield it from any lighting
problems, should do the trick.
##
Subject: Lens flare (was RE: 3DS Vs. Imagine speed)
Date: Sat, 27 Nov 93 20:18:37 -0700
From: Steve Koren <koren@hpfcogv.fc.hp.com>
[ stuff about the new lens flare in 2.9 ]
It always seemed to me that one ought to be able to add decent lens
flare as a post-processing step. Since lens flare happens inside the
"camera" it can't appear in back of another object, so you don't have to
worry about that. Seems like you could just point out the light sources
to some sort of post-processor and let it put the lens flare effect in
for you. Getting the post processor to do a reasonable job might take
some effort, though. But I bet it can be done.
Seems like the perfect job for an ImageFX ARexx script. Wouldn't slow
down your rendering that way. Maybe I'll give it a shot sometime.
It would only be useful for still frames of course; for an animation
it'd be easier to let the renderer take care of it (assuming you have
one which can).
Speaking of ARexx scripts, there was a bug in the one I posted here
before (Rain.ifx). I was doing bounds checking on the rain slant
parameter incorrectly, which means you couldn't make the rain fall to
the left with a negative slant parameter. Replace line 100 with this:
rain.slant = max(min(result.7, 40), -40)
That'll fix it. Sorry.
- steve
##
Subject: Re: Lens flare (was RE: 3
Date: Sun, 28 Nov 1993 03:24:00 -0500
From: j#d#.moore@canrem.com (J. Moore)
Ko> It always seemed to me that one ought to be able to add decent lens
Ko> flare as a post-processing step. Since lens flare happens inside the
Ko> "camera" it can't appear in back of another object, so you don't have to
Ko> worry about that. Seems like you could just point out the light sources
Ko> to some sort of post-processor and let it put the lens flare effect in
Ko> for you. Getting the post processor to do a reasonable job might take
Ko> some effort, though. But I bet it can be done.
That IS how 3D Studio's lens flare is done -- it's part of an add-on
module and it does it post-process.
* Q-Blue 0.93 [NR] *
##
Date: Mon, 29 Nov 93 07:07:52 -0700
From: Steve Koren <koren@hpfcogv.fc.hp.com>
> 4) Start an ARexx macro which will wait for **pairs** of frames to be
> rendered, then load each one and combine them. The composite image is
> Unfortunately, Imagemaster has no ARexx call for its "X-Specs
> Interleave" function.
I bet you could do it the brute force way with any of the "Big 3".
Instead of having a dedicated ARexx call to interleave buffers, write
your own in ARexx using a whole bunch of little rectangular cut and
pastes. It'd be a little slow, but you might be able to make it work.
Alternatively, I think most of the common image processing programs
these days allow you to write an external "hook" program in C, from
which you can modify the program's image buffer(s). That would give you
extra speed over ARexx. (At least ImageFX allows this; I recall the
others do also).
- steve
##
Subject: Re: PS fonts for Imagine 2.9
Date: 29 Nov 93 08:04:00 -0800
From: Ed_Totman@ucsdlibrary.ucsd.edu
PS fonts...
Happily using Imagine 2.9 on a 700 x 462 screen without special
hardware... Email me to learn how you can too.
Yes please tell us how!
---------
Location for fonts: wcarchive.cdrom.com
/pub/aminet/text/font/fonts_1.lha 600k
fonts_2.lha
fonts_3.lha
These are type 1, but you may have to rename them with the proper
extension (.pfb?). There are other type 1 fonts here as well.
Also, check out your local bbs's.
I noticed that not all fonts will load. Sometimes there's an
error message, sometimes just an axis appears with no text.
Anyone report this yet?
##
Subject: Re: Field Rendering
Date: Mon, 29 Nov 93 10:42:06 CST
From: setzer@comm.mot.com (Thomas Setzer)
> >>1] Although this can't be done within Imagine, has anyone experimented
> with >>doing so through one of the image processing programs using
> arexx?
>
> Imagemaster has a couple of functions used to combine and extract
> separate fields into one interlaced image; these were created with
> X-Specs 3-D glasses in mind, but they allow any program to do field
> rendering.
>
> 1) Halve the vertical resolution of your renderer.
>
You don't want to do this. You want to render an interlaced image, eliminate
even scan lines on odd frames and odd scan lines on even frames. So you need
software that not only combines two images, but separates them as well. Your
solution would cause an up and down jittery look, because you have in effect,
shifted the second image rendered down by one scan line. And if you are
trying to create field rendered animations you are probably looking for
quality, and the jitters would not be acceptable.
[stuff deleted]
>
> >>2] If so, since there would be no time saved during rendering, would
> the >>savings be passed on to the actual memory used for an animation?
> Correct me >>if I'm wrong, but that would seem to be a good way to
> economize on memory >>usage for those creating animations to be played
> back in memory, thus longer >>anims using less RAM (for those of us who
> are stuck being memory conscious). >>;-)
>
> Since you're rendering twice as many frames, but with only half the
> vertical resolution, the effects nearly balance each other out. Of
Since as above, you must render the full interlaced frames, and twice as
many of them, it will take around TWICE as long. THAT is what field rendering
buys you! Field rendering will allow you to achieve this effect in about
the half the time it takes to do it(field rendering) manually.
So, is there any software that can take an interlaced image and seperate it
into 2 images, even scanlines and odd scanlines? And then add 'em back?
Tom Setzer
setzer@ssd.comm.mot.com
"And of course, I'm a genius, so people are naturally drawn to my fiery
intellect. Their admiration overwhelms their envy!" - Calvin
##
Subject: Re: Lens flare (was RE: 3DS Vs. Imagine speed)
Date: Mon, 29 Nov 93 10:54:37 CST
From: setzer@comm.mot.com (Thomas Setzer)
>
> [ stuff about the new lens flare in 2.9 ]
>
> It always seemed to me that one ought to be able to add decent lens
> flare as a post-processing step. Since lens flare happens inside the
> "camera" it can't appear in back of another object, so you don't have to
> worry about that. Seems like you could just point out the light sources
> to some sort of post-processor and let it put the lens flare effect in
> for you. Getting the post processor to do a reasonable job might take
> some effort, though. But I bet it can be done.
Yeah, I thought about that. You'll notice on Lightwave renderings, the
flare seems to change based on the angle of the camera to the light souce
(or the position of the light source in the frame). It seems Imagines
lens flair would have a hard time simulating this. And if the light source in
Imagine is slightly off the screen, do you still see the lens flair? I would
imagine you would. Yuck. Yeah, post processing would probably be the best,
calculating the flair based on the position of the light source. Perhapes one
could use Imagines staging files to determine the true position of the light
source. Let us know when you are through coding:)
Tom Setzer
setzer@ssd.comm.mot.com
"And of course, I'm a genius, so people are naturally drawn to my fiery
intellect. Their admiration overwhelms their envy!" - Calvin
##
Subject: Re: Field Rendering
Date: Wed, 1 Dec 93 20:18:06 GMT
From: Andrew Nunn <apn@moby.demon.co.uk>
Hi Thomas,
On Nov 29 you wrote:
>
> So, is there any software that can take an interlaced image and seperate it
> into 2 images, even scanlines and odd scanlines? And then add 'em back?
>
ADPro has two Operators called Deinterlace and Interlace. The Deinterlace
operator separates the odd and even lines into two pictures in the upper
and lower half of the screen. The Crop Image Op could then be used save
the top or lower half. This could probably be done through Arexx.
Interlace does the reverse!
Its likely that the other Image processing packages probably have similar
functions.
Andrew
##
Subject: REACH OUT N TOUCH SOMEONE
Date: Wed, 1 Dec 93 13:39:46 CST
From: dave@flip.sp.paramax.com (Dave Wickard)
Salutations Renderoids! :-)
Hope Turkey Day was as enjoyable for all of you as it was
for me. I had enough tryptophan in me for a 3 day coma.
ZZZZzzzzzzzzz
------=======--------=====------=====-----======-----======-----
WHERE'S MY IML??? Perhaps you've been asking yourself this
over the last day or two. Mailings have been sporadic
at best, and non-existant at worst. After a bunch of 15-20
post days, it was kind of a letdown to see trouble arise
in the mailing system. What happened was this...
we are moving the files all over to another system.
The new system will be keeping the same name, so you
don't have to change anything at your end. However, in the
transition period, they were trying to have TWO systems
running transparently with the same name. This meant there
were aliases and temporary files and directories strung
out all over the place and it wound up getting a little
hairy. There was also a small matter of installing a new
version of Solaris (OS) on the new system. New hardware,
new software, new mailer software.... guess I can see why
we had a slight glitch. I *think* things are about ready
to go back on keel now, so if you mailed something to the
list within the last day or two and it never showed up...
chances are it may have been banished to the electronic
trashcan. The mailer should be ready again shortly. Thanks
for your patience.
------==========--------==========-------=========--------========
I thought I should clear something up, in case it confused
anyone else. If you subscribe, unsubscribe, or change addresses
at the usual-
imagine-request@email.sp.paramax.com
account, it will take effect everywhere. Sometimes there are
things still in the queue, or one site might be a day or
two behind the others... but everything works off a
master list I keep here, so don't worry, things will
straighten themselves out. Any slowdown in the few days
was caused by the above system shuffling. You don't need to worry
about additional UNSUBSCRIBE messages to the
imagine-request@shell.portal.com account.
-----------==========----------=========--------=========-----
I am forwarding a copy of all the IML postings since
Impulse dropped off the list to Mike H., and he is going to
read the posts so he will be up to speed when we get them
(Impulse) back online here (soon). I will let you know
shortly before we are rejoined by Impulse.
----------=====-----------======----------======-------------
Regarding Byte by Bytes's recent offer to members of the
IML for a reduced price on the 1.1 version of Envisage 3D...
I would encourage PC-based rendermaniacs to check it out.
I am very happy with the included documentation. It is
refreshing to immediately get into the meat of a rendering
program without all the floundering in a steep learning
curve. It has a fairly nice interface (it seems to me that
virtually all rendering package user interfaces could be
improved), it's rendering speed is quick, and it's output
is quite good.
It's a full-featured package that I got a chance to monkey
around with for a while. It has a nice feel to it and
I would encourage you to speak with Doug Kelly (dakelly@class.org)
for his more extensive experiements with this product and
be sure to ask him about his upcoming review of the package
in an upcoming issue of a commercial magazine.
While the $995 list price sets it above the budget of many
hobbyist 3D users, the reduced offer of $500 to IML readers
makes it a realistic alternative or even a supplementary
product with ImaginePC to some PC rendering hobbyists, and
is a comparative bargain for the professional PC-based 3D user.
I would like to thank Byte by Byte for making this offer
available to the rendering fans of the IML. Naturally, any
special pricing from any relevant manufacturers for IML
subscribers is appreciated during this...
THE SEASON OF BROTHERHOOD AND GIVING ;-) heh heh
------====-----======----=====----====----====-----====-----
Shortly after the mailer is up and running smoothly,
I will be sending out the IML directory again for the
first time in a while. I think we are now back up to
420 sites or so.
-------=========------=======-------========----========-----
Anyways, that's about it from here in the beautiful
and stately halls of Unisys. I made it through the
November layoff massacre... and now I will begin
looking for a safe hiding spot to try and avoid the
next butchery which is expected to be in January or March.
Hmmmmm, maybe this suspended ceiling tile panel can be
lifted out...... shhhhhhh
;-)
Dave Wickard (612) 456-2783 "Hey, these almonds sure are good!"
dave@flip.sp.paramax.com "Yeah... they were even better
dave@email.sp.paramax.com until Grandma sucked all the
Sam_Malone@cup.portal.com chocolate off of em."
dave@shell.portal.com
dwickard.EAGAN@mhs.sp.paramax.com
(in order of preference)