home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
EnigmA Amiga Run 1998 July
/
EnigmA AMIGA RUN 29 (1998)(G.R. Edizioni)(IT)[!][issue 1998-07 & 08].iso
/
recent
/
iib122.lha
/
IIB
/
Threads
/
Star_Render
< prev
next >
Wrap
Internet Message Format
|
1998-04-05
|
6KB
Date: Sat, 28 Mar 1998 17:46:13 -0800
From: Drift Dennis <drift@NIGHTHAWK.COM>
Subject: Re: [IML] IFW: STARS RENDER
Back a few years ago Impulse added the star back drop, only problem was, it
acted like roughness, don't use in an anim, a cry went up from the Imagin
Nation and the Impulsive gods heard and said LET THERE BE CONTINUTIY, and
so the people had stars that could be panned and animatied to look more
then half way decent and the people said LO THIS IS COOL.
Now in IFW I render an anim with stary stary skies, and LO it looks like a
snow storm, very little camera movement, consistant and chaotic snow, a
blizzared no less.
So Now I rend my clothes and scatter ashes upon my head as I cry out WHA
HAPPENED?
(melodramaticly exiting stage left) woe betide the astronomical
----------------------------------
Date: Sat, 28 Mar 1998 23:33:06 -0600
From: Mike Bayona <mb@MB.SIMPLENET.COM>
Just a thought but is your "speed" setting in the stars dialog, set to
anything other than 0? If it is, its trying to make the stars move.
----------------------------------
Date: Sun, 29 Mar 1998 00:41:34 -0800
From: Drift Dennis <drift@NIGHTHAWK.COM>
I MUST BE GOING BLIND, Oh no, I swaer Officer I lthought I had my eyes open.
Thanx Mike, now I know just how stupid that stuff makes me (yadayadayada)
Drift... backing up in order to ram my head agin this here block wall.
----------------------------------
Date: Tue, 31 Mar 1998 09:17:11 -0600
From: Mike Bayona <mb@MB.SIMPLENET.COM>
>At that point I got disgusted with Imagine in
>general and haven't used it much since (among other things, I've been
>waiting 5 years for software which animates both the stars AND the sky
>correctly, and eventually had to move on).
I hope its not to late, but I have a tiny bit of info on what Iused to do
in Imagine [PC]. I never noticed any of the above problems you mentioned,
however I was never happy with the density of the starfield, it was always
my *opinion* that a setting of density = 1 should make the sky white with
stars, so that you could have the sky be as dense as you wanted. But, even
at 1, the stars were not near as dense as I liked, so I ended up one day
distorting the perspective until I was happy with the density, rendering a
stars only picture, and using this wrapped on a sphere for the sky, the
advantage this has, is you can take your star picture into photoshop and
make some "twilight" gradients in it or whatever you want, so its not just
plain black. Not only will this animate properly, but you will get
reflections of the sky in all your objects [the ones that reflect of
course], which is also nice.
I guess maybe this falls under "Tip of the Day"? :)
----------------------------------
Date: Tue, 31 Mar 1998 11:06:48 -0800
From: Stephen Bray <4dsol@UNIBASE.UNIBASE.COM>
Organization: 4D Solutions
....and of course if you're not interested in having the sky reflection
in your objects or need the sky to be animatable just render a starfield
and drop it in as a backdrop image. The backdrop image should at least
be the same size as the final rendered image..but resampled (smaller)
images work faster for testing purposes and provided they are the same
ratio they won't look distorted.
PS Make sure you load the correct image in the action dialog (or
quickrender dialog) when you're ready to do the final image tho.
----------------------------------
Date: Mon, 30 Mar 1998 19:28:05 -0700
From: Mike Grusin <michael.grusin@COLORADO.EDU>
Near as I can tell, here's the story of the stars from the IFA point of view
(it's been a long time so if the version numbers are wrong, I apologize)...
2.0: Starfield effect added, but all stars are the same brightness and don't
animate (they are reseeded for each frame).
3.0: Stars can now be animated (they are reseeded at begining of the
animation), but an axis error makes them move in the opposite direction from
the rest of the world.
3.3: Stars now move correctly, and additionally, they are now random
brightnesses. They look quite beautiful. However, for some reason the
motion of the sky is now broken; it is rendered upside down (screen-Y
inverted). This makes any animation using both the sky and the stars (such
as a twilight scene) useless.
4.0: The sky remains broken, despite bug reports.
5.0: The sky remains broken in the latest version, despite additional bug
reports. I don't know if the latest ACUP has fixed this bug, or if any of
this pertains to IFW.
Now for your specifc question: I've noticed the same thing, and am not sure
whether it's due to aliasing problems, an implementation problem, or a new
bug (all distinct possibilities). I haven't played with it extensively; I
tried a few workarounds (in 5.0) for the above sky problem, none of which
were ultimately acceptible. At that point I got disgusted with Imagine in
general and haven't used it much since (among other things, I've been
waiting 5 years for software which animates both the stars AND the sky
correctly, and eventually had to move on).
I suspect it's an aliasing problem; perhaps the stars are rendered so small
that the dimmer ones fall "between" pixels and get averaged out. I suspect
this is the problem because if you watch a star-animation carefully, you may
see the brighter stars moving correctly and the dimmer ones strobing. You
might try increasing the antialising level to the best possible setting and
seeing if this helps. HOWEVER, I'm not sure WHY this would be a problem,
since the stars SHOULD be a prerender effect; each star should simply fill
the nearest entire pixel with the correct color/brightness, no subpixel work
should then need to be done (which is why I say it might be an
implementation problem).
My advice: play with it, try a few things, figure out what's exactly what's
wrong, then submit bug reports repeatedly until you get a response (maybe
Impulse paying more attention these days).
----------------------------------