ftp.xmission.com/2014.06.ftp.xmission.com.tar/ftp.xmission.com/pub/lists/fractdev/archive/v01.n039

Internet Message Format | 2001-01-20 | 41KB

From: [e-mail] (fractdev-digest)
To: [e-mail]
Subject: fractdev-digest V1 #39
Reply-To: fractdev-digest
Sender: [e-mail]
Errors-To: [e-mail]
Precedence: bulk


fractdev-digest        Sunday, January 21 2001        Volume 01 : Number 039




----------------------------------------------------------------------

Date: Sat, 30 Dec 2000 23:03:51 -0600
From: Tim Wegner <[e-mail]>
Subject: Off topic - trip north

Jonathan,

I zoomed through your neck of the woodswith my family without
incident for the annual holiday trip. (Houston to Minneapolis and
back).

We had no trouble at all, but did encounter strong winds and
blowing snow in Iowa. The road conditions the two days we came
through (21st and 29th) were otherwise nearly perfect. However, I
swear we saw at least 200 cars stuck in the snow - not
exaggerating. They would usually be in groups of six or seven,
probably all having swerved to avoid some initial out-of-control car.
Most were only stuck, not damaged. I used to think it was only
Texas drivers who couldn't handle winter driving <grin!> Also saw
four or fix overturned semi-trailers. This seemed very strange
because driving conditrions were excellent when we came through.
A few hours earlier conditions had probably been very bad.

Today in northern Oklahoma we drove through a surreal scene. On
the previous day there must have been a freezing rainstorm that
had surely caused tremendous damage (downed power lines).
What we saw was several hundred miles of rural Oklahoma where
every bush, tree, fence, wire, and blade of grass was coated with
ice. It was extraordinarily beautiful.

Once again driving conditions were perfect. No wind, the road was
dry and clear, the temperature was above freezing, the sun was
shining. But the landscape was magically transformed into crystal.
I have seen similar effects before, but the ice had melted quickly.
This ice was so thick (maybe half an inch or more in diameter) that
the effect was undiminished despite rising temperatures. This
scene continued for several continuous hours. At one point we
stopped and pulled off the road to see the crystall bushes up close.

Crystal fractals!

Tim

- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Sun, 31 Dec 2000 07:09:03 -0600
From: "Jonathan Osuch" <[e-mail]>
Subject: Re: Off topic - trip north

Tim,

> I zoomed through your neck of the woodswith my family without
> incident for the annual holiday trip. (Houston to Minneapolis and
> back).

> We had no trouble at all, but did encounter strong winds and
> blowing snow in Iowa. The road conditions the two days we came
> through (21st and 29th) were otherwise nearly perfect. However, I
> swear we saw at least 200 cars stuck in the snow - not
> exaggerating.

The storms you missed on both the 21st and 29th were on the East side of
Iowa when I was driving home both those days.  Not pleasant.  You were very
lucky.

Yesterday Rachel put one of our cars into the median strip.  Hit an icy spot
and used the snow to slow the car down.  Nobody was hurt, no apparent damage
to the car.  It looks like she just pulled over and parked it.  But, it's
still there.  The weather got progressively worse during the day.  It's
supposed to be towed this morning, but it's snowing again.

Jonathan



- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Sun, 31 Dec 2000 07:35:40 -0600
From: "Jonathan Osuch" <[e-mail]>
Subject: Re: Bug of the Month

Tim,

> I do not believe that the medium model Fractint can ever support
> PNG except via an external program - e.g. save to a flat file and
> execl to an external program that transforms the external file to a
> PNG.

My thoughts were to get PNG into Xfractint.  The Allegro package lets you
save images in several different formats (not PNG).  Are we going to have to
go through the same contortions we do now for GIF between the different
ports?  This is regarding the fractal specific data that gets saved big
endian or little endian to the file and we have to swap it with Xfractint
and not Fractint.

Have you had any luck/time/patience getting the Allegro package installed?
That rpm takes forever to install with no visible signs of it working.  If
you switch to a different virtual terminal you'll see that the cpu is very
busy.

On a different subject, take a look at lodimagmul in parsera.asm.  And then
look at lodrealmul just above it.  The two are identical.  This doesn't seem
right.  I think the lodimagmul needs fxch and fchs added at the end to make
it correct.

Jonathan



- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Sun, 31 Dec 2000 11:03:37 -0600
From: Tim Wegner <[e-mail]>
Subject: Re: Off topic - trip north

(added list back in to Scott's response)

Scott wrote:

> I *was* going to ask you if you saw any
> fractal patterns in the ice crystals. Then I read the last comment...
> Looks like you did!  8-)

No doubt at all. Crystal fractals, a feast for the eyes.

> Oh - they said on the news Friday nite (I believe), that the Governor of
> Oklahoma had declared the whole state a disaster area.

Though I didn't see any damage myself otrher than a few utility
trucks, I am sure freezing rain is a very serious problem. Very
strange that a phenomenon that causes such a disaster is also
such a beautiful visual feast.

Tim


- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Sun, 31 Dec 2000 11:03:37 -0600
From: Tim Wegner <[e-mail]>
Subject: Re: Bug of the Month

Jonathan wrote:

> My thoughts were to get PNG into Xfractint.  The Allegro package lets you
> save images in several different formats (not PNG).

This is no big deal. PNG support can be added in a day or so. We
can implement support directly using Libpng, and Xfractint and the
allegro port.

> Are we going to have to
> go through the same contortions we do now for GIF between the different
> ports?  This is regarding the fractal specific data that gets saved big
> endian or little endian to the file and we have to swap it with Xfractint
> and not Fractint.

You have zeroed right in on the one hard issue.

There are two approaches. When PNG was first created, I thought
we were going to instantly adopt PNG. So I lobbied hard for some
special PNG chunks for saving GIF extension blocks. Using these
would be the easiest, because effectively we'd just have the
existing GIF extension blocks inside PNG, and essentially the
same code would be used for saving/reading data.

But it's certainly not a good idea to do this.

For one thing, I seriously doubt GIF-to-PNG utilities support these
chunks. (If we had done our PNG implementation immediately, we
probably could have influenced everyone to have this support, but
it's too late now.)

Secondly, the GIF extension blocks in PNG were only envisioned
as a temporary transition issue. The concern at the time was to
facilitate a massive move from GIF to PNG. This is a non-issue for
use because we can write our own GIF-to-PNG utility that correctly
saves

I raise the GIF extension blocks just for historical interest. There is
NO point to doing this.

I have reserved a chunk name fRAc for the purpose of saving fractal
data in PNG, but have never defined it.  Designing this chunk
properly will take effort. It can store numbers in text to eliminate all
endian issues. It should NOT use version numbers, but should use
a well-thought-through subchunks scheme that never changes the
definition of a subchunk. This is PNG's philosophy. If you need
something new, you define a new chunk. Ideally the fRAc definition
should not be fractint-specific, but this may be moot now. A team
design effort would be a mess. We could publish a proposed spec
and accept feedback from interested parties. The subchunk format
is great because others can just add their own subchunks if we
leave something out.

We can also use an internal chunk name. We could just use the
PAR format using a private chunk name. This would give us some
time to figure out fRAc.

Since we are making major mods to fractint, I would prefer to take
the opportunity to devise new solutions so we don't extend legacy
problems further.

> Have you had any luck/time/patience getting the Allegro package installed?
> That rpm takes forever to install with no visible signs of it working.  If
> you switch to a different virtual terminal you'll see that the cpu is very
> busy.

Nothing new since I attempted this before the holidays. I believe I
did not install Allegro properly. I'll have to try again.

Tim


- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Sun, 31 Dec 2000 12:53:04 -0500
From: "Damien M. Jones" <[e-mail]>
Subject: Re: Bug of the Month

Tim,

 - I have reserved a chunk name fRAc for the purpose of saving
 - fractal data in PNG, but have never defined it.  Designing
 - this chunk properly will take effort. It can store numbers
 - in text to eliminate all endian issues.

However if you store numbers as text you have precision issues, unless you
allow variable-size fields to store variable-precision numbers.

Would it be appropriate to simply include the information in plain text
form, as it is stored in a PAR file? The difference being that if it's in a
fRAc chunk, FractInt would know it's parameter information and would
attempt to interpret it, rather than the image simply being a PNG from an
undetermined source. But, this might go against the PNG philosophy (I
haven't examined the PNG format specification in detail, it's been too long
since I wrote image file format code).

I guess I'm just curious as to why you want a binary format for the chunk.
:)

Damien M. Jones   \\
[e-mail]  \\  Fractalus Galleries & Info:
                    \\  http://www.fractalus.com/

Please do not post my e-mail address on a web site or
in a newsgroup.  Thank you.


- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Sun, 31 Dec 2000 12:20:13 -0600
From: Tim Wegner <[e-mail]>
Subject: Re: Off topic - trip north

Jonathan wrote:

> The storms you missed on both the 21st and 29th were on the East side of
> Iowa when I was driving home both those days.  Not pleasant.  You were very
> lucky.

I have no doubt we were fortunate!

I do have respect for the weather, and watched it carefully. If it had
looked like we would be heading into a storm, I'd have tried to
change our schedule. Of course one can't always tell ahead ...

> Yesterday Rachel put one of our cars into the median strip.  Hit an icy spot
> and used the snow to slow the car down.  Nobody was hurt, no apparent damage
> to the car.  It looks like she just pulled over and parked it.

I'm glad Rachel was not hurt. This appeared to me to be the case
for most of the stranded cars, with a few unfortunate exceptions.

Tim


- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Sun, 31 Dec 2000 12:19:01 -0600
From: Tim Wegner <[e-mail]>
Subject: Re: Bug of the Month

Damien wrote:

> However if you store numbers as text you have precision issues, unless you
> allow variable-size fields to store variable-precision numbers.

Exactly - variable size. The format could start with a number that
says how long the field is.

And also wrote:

> Would it be appropriate to simply include the information in plain text
> form, as it is stored in a PAR file?

That's roughly the idea.

> I guess I'm just curious as to why you want a binary format for the chunk.
> :)

I don't. Binary format in a chunk is bad. I'd prefer everything in the
chunk to be plain text. However, if there are numbers (e.g. field
length) they would be in network byte order.

Tim


- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Mon, 1 Jan 2001 07:49:28 -0600
From: "Jonathan Osuch" <[e-mail]>
Subject: Version 20.1 patch 6

Patch 6 and all the executables are on the developer's web site.  This is a
minor patch that only affects saving key assignments to fractint.cfg.  This
time it should work, honest.

Jonathan



- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Sat, 13 Jan 2001 07:35:26 -0600
From: "Jonathan Osuch" <[e-mail]>
Subject: Float only Allegro port

For those intrepid souls out there that have tried to use the float only
Allegro port of Xfractint:

1)  Either copy default.map to the directory you are compiling in, or remark
out the colors= statement in sstools.ini.  This will allow Xfractint to run.

2)  Comment out the textcolors statement in sstools.ini since this is in
there so I can see some of the text on my laptop.

I will address these problems the next time I upload the source.

Jonathan



- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Mon, 15 Jan 2001 20:10:55 -0600
From: Tim Wegner <[e-mail]>
Subject: Welcome Jean-Pierre!

Folks, I'd like to introduce Jean-Pierre.Demailly, who just joined
this list. Jean-Pierre sent us a modified unixscr.c for Xfractint which
works when X is set for a truecolor mode. This goes a long way
toward fixing a long-time irritant, which is that Xfractint barely
functioned unless in a 256-color mode.

I will merge this in to an official patch, but (most likely) not until the
weekend. Meanwhile, I have put the file at:

http://www.fractint.org/ftp/experimental/uniscr.c

in case anyone wants to experiment.

Just for Jean-Pierre's information... Our most active developer,
Jonathan Osuch, is generally off line except for weekends (he
commutes between Cedar Rapids Iowa and Chicago) so the list is
often quiet. Then again you never know :-)

Tim


- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Tue, 16 Jan 2001 08:46:39 +0000
From: "Thierry B." <[e-mail]>
Subject: Re: Welcome Jean-Pierre!

Tim Wegner wrote:
>
> http://www.fractint.org/ftp/experimental/uniscr.c
>
> in case anyone wants to experiment.
>

    http://www.fractint.org/ftp/experimental/unixscr.c


- --
Thierry Boudet                                   http://la.buvette.org/

- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Tue, 16 Jan 2001 10:55:50 -0700
From: "Rich" (aka "Phil") <[e-mail]>
Subject: Re: Welcome Jean-Pierre!

In article <3A63594F.19805.7F6B46@localhost>,
    Tim Wegner <[e-mail]>  writes:

> Folks, I'd like to introduce Jean-Pierre.Demailly, who just joined
> this list. Jean-Pierre sent us a modified unixscr.c for Xfractint which
> works when X is set for a truecolor mode. This goes a long way
> toward fixing a long-time irritant, which is that Xfractint barely
> functioned unless in a 256-color mode.

Umm... I'm fairly certain I sent in a fix for that about two years
ago...
- --
Ask me about my upcoming book on Direct3D from Addison-Wesley!
Direct3D Book <http://www.xmission.com/~legalize/book/>
         Home <http://www.xmission.com/~legalize/>
    Fractals! <http://www.xmission.com/~legalize/fractals/>

- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Tue, 16 Jan 2001 17:11:33 -0600
From: Tim Wegner <[e-mail]>
Subject: Re: Welcome Jean-Pierre!

Rich wrote:

> Umm... I'm fairly certain I sent in a fix for that about two years
> ago...

Yes I recall. Jonathan did his best to merge your version with the
main source tree. Maybe when he gets back he can explain to us
his recollection of what happened. My understanding is that the
current Xfractint sources have your changes, but maybe the
unixscr.c changes either didn't merge correctly or aren't there.

Tim



- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Tue, 16 Jan 2001 17:15:29 -0700
From: "Rich" (aka "Phil") <[e-mail]>
Subject: Re: Welcome Jean-Pierre!

OK, so what's the reason why my name and email address still don't
appear in the credits screen?  At least the last time I downloaded
Fractint 20.x, it didn't appear.

Got money.

Want recognition ;-).
- --
Ask me about my upcoming book on Direct3D from Addison-Wesley!
Direct3D Book <http://www.xmission.com/~legalize/book/>
         Home <http://www.xmission.com/~legalize/>
    Fractals! <http://www.xmission.com/~legalize/fractals/>

- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Wed, 17 Jan 2001 00:08:49 -0600
From: "Scott D. Boyd" <[e-mail]>
Subject: Testing results of the 16-bit Xfractint enhancement

Tim, and any other interested developers:

OK - I've finished running tests of the 16-bit color fix for Xfractint. I
applied this "fix" to Xfractint v20.1.04, and gave it a different binary
name, and put it in a different directory from my stock 20.1.04. I decided
to use Jim Muth's Fractal of the Day for Jan. 17th as my test, since I
know what it is supposed to look like.

I first created the FOTD in an 8-bit Xserver, with Xfractint-20.1.04, and
saved it to a file. I then switched to a 16-bit Xserver, and loaded the
image into both Xfractint-20.1.04, and the modified Xfractint. It appeared
all blue with black bkgrnd in Xfractint-20.1.04 (as expected), and colors
were different in the modified Xfractint. The image appears correctly
though, when I view it with any other image viewer, or with Netscape. It
even has the same colors as the same image on Paul N. Lee's Website.

I then created the same image using the .par file in the modified
Xfractint. The resulting image had the wrong colors. When viewing that
image with another
image viewer, the colors were even worse - no - make that *uglier*!

I uploaded them to my Website, and just to make sure it wasn't something
wrong with the colors on my Linux box, I fired up a Win95 (gasp!)
partition on an old P100 machine. I checked the color-depth that this
Win95 box was using, and it is set to "High Color (16 bit)". When I viewed
the images on my Website, they were as I described them in the last two
paragraphs.

Here are the URLs if any one would like to take a look at what I have been
describing:

the image created in 8-bit color and Xfractint-20.1.04:
http://home/swbell.net/sdboyd56/test/fotd0117-8bit.gif

The image created in 16-bit color and the modified Xfractint-20.1.04:
http://home/swbell.net/sdboyd56/test/fotd0117-16bit.gif

(Both of these files are approx 215KB in size.)

Any comments are welcome...
Scott Boyd

- --
[e-mail]
http://home.swbell.net/sdboyd56/
- -------------------------------------------------------
"Microsoft is a cross between the Borg and the Ferengi.
Unfortunately, they use Borg to do their marketing and
Ferengi to do their programming."

                 --- Simon Slavin in asr

- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Wed, 17 Jan 2001 00:35:41 -0600
From: "Scott D. Boyd" <[e-mail]>
Subject: Re: Testing results of the 16-bit Xfractint enhancement

On Wednesday 17 January 2001 00:08, Scott D. Boyd wrote:

Oops! I put in a forward-slash where a dot should be....

The correct URLs are:

> the image created in 8-bit color and Xfractint-20.1.04:
> http://home.swbell.net/sdboyd56/test/fotd0117-8bit.gif
>
> The image created in 16-bit color and the modified Xfractint-20.1.04:
> http://home.swbell.net/sdboyd56/test/fotd0117-16bit.gif
>
> (Both of these files are approx 215KB in size.)
>
Scott

- --
[e-mail]
http://home.swbell.net/sdboyd56/
- -------------------------------------------------------
"Microsoft is a cross between the Borg and the Ferengi.
Unfortunately, they use Borg to do their marketing and
Ferengi to do their programming."

                 --- Simon Slavin in asr

- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Wed, 17 Jan 2001 11:37:43 -0300
From: "Humberto R. Baptista" <[e-mail]>
Subject: RE: Testing results of the 16-bit Xfractint enhancement

    Hi,

    I'm sure this must have been through this sometime in the list but I don=
't
recall the conclusion:

    GIFs are bounded to 8 bit depth, how the 16 bit image is supposed to be
saved as a GIF?

    Other thing: both images seem ok, except for the pallete. In the "16bit"
one the pallet is the default Fractint pallete and in the 8 bit appears a
different one (probably like the one in the original FOD).

    Best regards to all.

    HB

- -----Mensaje original-----
De: [e-mail]
[mailto:[e-mail]]En nombre de Scott D. Boyd
Enviado el: Mi=E9rcoles, 17 de Enero de 2001 03:36
Para: [e-mail]
Asunto: Re: Testing results of the 16-bit Xfractint enhancement


On Wednesday 17 January 2001 00:08, Scott D. Boyd wrote:

Oops! I put in a forward-slash where a dot should be....

The correct URLs are:

> the image created in 8-bit color and Xfractint-20.1.04:
> http://home.swbell.net/sdboyd56/test/fotd0117-8bit.gif
>
> The image created in 16-bit color and the modified Xfractint-20.1.04:
> http://home.swbell.net/sdboyd56/test/fotd0117-16bit.gif
>
> (Both of these files are approx 215KB in size.)
>
Scott

- --
[e-mail]
http://home.swbell.net/sdboyd56/
- -------------------------------------------------------
"Microsoft is a cross between the Borg and the Ferengi.
Unfortunately, they use Borg to do their marketing and
Ferengi to do their programming."

                 --- Simon Slavin in asr

- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"


- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Wed, 17 Jan 2001 09:01:34 -0600
From: "Scott D. Boyd" <[e-mail]>
Subject: Re: Testing results of the 16-bit Xfractint enhancement

On Wednesday 17 January 2001 08:37, Humberto R. Baptista wrote:

Actually, they are both 8-bit (256-color) GIF images. One was created
while XFree86/KDE were running in 8-bit color depth, the other, running in
16-bit color depth. This is pertaining to the color depth of the Xserver
on my display (or monitor). The current version of Xfractint will only
work properly while XFree86 is running in 8-bit color.

Scott

> Hi,
>
>     I'm sure this must have been through this sometime in the list but I
> don't recall the conclusion:
>
>     GIFs are bounded to 8 bit depth, how the 16 bit image is supposed to be
> saved as a GIF?
>
>     Other thing: both images seem ok, except for the pallete. In the
> "16bit" one the pallet is the default Fractint pallete and in the 8 bit
> appears a different one (probably like the one in the original FOD).
>

- --
[e-mail]
http://home.swbell.net/sdboyd56/
- -------------------------------------------------------
"Microsoft is a cross between the Borg and the Ferengi.
Unfortunately, they use Borg to do their marketing and
Ferengi to do their programming."

                 --- Simon Slavin in asr

- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Wed, 17 Jan 2001 13:15:38 -0300
From: "Humberto R. Baptista" <[e-mail]>
Subject: RE: Testing results of the 16-bit Xfractint enhancement

    I see (dumb me),

    Is there any link between the video routines and the pallete saving in t=
he
GIFs? I recall vaguely that the GIF saving routine read the pallete from =
the
video DAC, it may be reading garbage or the old default fractint pallete
instead of the true values (which are probably in the pallete data strutu=
re
in RAM not in the DAC).

    Just my $0,02 :-)

    Best regards,

    HB

- -----Mensaje original-----
De: [e-mail]
[mailto:[e-mail]]En nombre de Scott D. Boyd
Enviado el: Mi=E9rcoles, 17 de Enero de 2001 12:02
Para: [e-mail]
Asunto: Re: Testing results of the 16-bit Xfractint enhancement


On Wednesday 17 January 2001 08:37, Humberto R. Baptista wrote:

Actually, they are both 8-bit (256-color) GIF images. One was created
while XFree86/KDE were running in 8-bit color depth, the other, running i=
n
16-bit color depth. This is pertaining to the color depth of the Xserver
on my display (or monitor). The current version of Xfractint will only
work properly while XFree86 is running in 8-bit color.

Scott


- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Thu, 18 Jan 2001 20:12:07 -0600
From: Tim Wegner <[e-mail]>
Subject: Re: Testing results of the 16-bit Xfractint enhancement

Scott wrote, concerning testing the unixscr.c changes:

> Any comments are welcome...

The links to your test images didn't work, so I will make some
general comments, unencumbered by any knowlege of Xfractint or
the latest unixscr.c changes :-)

There are two ways that GIF's made under two environments might
differ:

1. The 8-bit color indices might be different.
2. The color palettes might be different.

In addition, even if neither #1 nor #2 are true, we might have:

3. The images look different under the two environments despite
the GIFs being the same.

If item #1 is not true (i.e. if the same color indices exist in both
images), then we are nearly home free, because we just need to
check all the palette handling routines. If item #1 is true, then we
have big problems which need to be debugged.

In many ways operating Xfractint in a truecolor environment is
easier, because a 256-color GUI environment has complications
due to colors needing to be preserved.

At the moment I can only test in a 16 bit hi-color environment,
because I have to jump through some hoops to get the proper
drivers for my Nvidia Vanta board on my HP. Hopefully uasing a 16
bit mode is not a problem. The board supports 24 bit color but the
drivers don't come with linux, and I would also have to reinstall
Mandrake in order to get XFree86 4.0.x.

I'd like to dig out Rich's two-year-old patch which apparently didn't
get in to our main sources and compare the unixscr.c changes
with Jean-Pierre's.

BTW I can't use Xfractint for more than a little bit before it loses
track of the keystrokes and exits if I try to use a cursor key.

Tim


- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Thu, 18 Jan 2001 20:40:41 -0600
From: "Scott D. Boyd" <[e-mail]>
Subject: Re: Testing results of the 16-bit Xfractint enhancement

On Thursday 18 January 2001 20:12, Tim Wegner wrote:
> Scott wrote, concerning testing the unixscr.c changes:
> > Any comments are welcome...
>
> The links to your test images didn't work, so I will make some
> general comments, unencumbered by any knowlege of Xfractint or
> the latest unixscr.c changes :-)
>
I just checked, and it seems to be working. I had typos in the origin=
al=20
email, perhaps that's the one you're referring to. Here's the correct=
=20
URL's:

>=A0the image created in 8-bit color and Xfractint-20.1.04:
>=A0http://home.swbell.net/sdboyd56/test/fotd0117-8bit.gif
>
>=A0The image created in 16-bit color and the modified Xfractint-20.1=
.04:
>=A0http://home.swbell.net/sdboyd56/test/fotd0117-16bit.gif
>
> BTW I can't use Xfractint for more than a little bit before it lose=
s
> track of the keystrokes and exits if I try to use a cursor key.
>
I know that you told me several months ago whether you were running=
=20
Xfractint in xterm or KDE's Konsole, but I don't remember now. I have=
=20
tried running Xfractint in Konsole, but it won't properly read the cu=
rsor=20
keys (arrow keys). Whenever I press a cursor key, Xfractint jumps to =
the=20
"exit Xfractint (are you sure?)" screen. I have found that it works b=
est=20
with xterm, where all keys are read correctly. Besides, xterm uses le=
ss=20
memory. That's important on a 64MB Linux box running KDE 2.0. (Gotta =
get=20
more RAM soon...)

Scott
- --=20
[e-mail]
http://home.swbell.net/sdboyd56/
- -------------------------------------------------------
"Microsoft is a cross between the Borg and the Ferengi.=20
Unfortunately, they use Borg to do their marketing and=20
Ferengi to do their programming."=20

                 --- Simon Slavin in asr=20


- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Fri, 19 Jan 2001 19:07:11 -0600
From: "Jonathan Osuch" <[e-mail]>
Subject: Re: Welcome Jean-Pierre!

Rich wrote:

> Umm... I'm fairly certain I sent in a fix for that about two years
> ago...

Unless it was included with the experimental source, I suspect that fix fell
victim to the numerous reinstalls of Linux that have occurred in the last
two years.  And, two years is way beyond the four day limit on my memory
bank capacity.  8-))

> OK, so what's the reason why my name and email address still don't
> appear in the credits screen?  At least the last time I downloaded
> Fractint 20.x, it didn't appear.

Which name and email address would you like on the credits screen?

Jonathan



- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Sat, 20 Jan 2001 11:50:28 +0100 (MET)
From: Jean-Pierre Demailly <[e-mail]>
Subject: Re: unixscr.c

> Rich wrote:
>
> > Umm... I'm fairly certain I sent in a fix for that about two years
> > ago...
>

Is this two year old fix available somewhere? I'd be interested to see
what was in it and (possibly) merge with the changes I proposed. In fact,
my fix is more than just a fix, it also includes a new algorithm for
generating palettes in truecolor modes** -- which still would need to be
integrated with the gif output routine (probably easy). At first glance,
the palettes generated by the new algorithm produce better looking
colors than what one gets in pseudocolor mode.

Best wishes,

Jean-Pierre Demailly

** Including color cycling.

- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Sat, 20 Jan 2001 07:56:19 -0600
From: "Jonathan Osuch" <[e-mail]>
Subject: Re: Testing results of the 16-bit Xfractint enhancement

Tim,

> BTW I can't use Xfractint for more than a little bit before it loses
> track of the keystrokes and exits if I try to use a cursor key.

Because I dislike the 'web' view when I'm trying to look at files, I've
started using XWinCommander.  This program calls up Xterm instead of using
Konsole.  It is possible to change the configuration of Konqueror so that it
calls Xterm.

My initial problem with this patch is that 256 color mode is now broken.
I'll have a look at the Allegro port, since it started from Rich's base
experimental source and it handles the color maps for all color depths.

Jonathan



- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Sun, 21 Jan 2001 10:51:36 -0700
From: Rich <[e-mail]>
Subject: Re: Welcome Jean-Pierre!

In article <001a01c0827f$a0b8d8a0$0100a8c0@bananasenior>,
    "Jonathan Osuch" <[e-mail]>  writes:

> Unless it was included with the experimental source, I suspect that fix fell
> victim to the numerous reinstalls of Linux that have occurred in the last
> two years.  And, two years is way beyond the four day limit on my memory
> bank capacity.  8-))

It may be buried inside the "experimental" code that I have at
<ftp://ftp.xmission.com/pub/users/l/legalize/fractals/>.

I think that code incorporates the device driver abstraction and I
massaged the relevant xfractint code into the X11 driver.
- --
Ask me about my upcoming book on Direct3D from Addison-Wesley!
Direct3D Book <http://www.xmission.com/~legalize/book/>
         Home <http://www.xmission.com/~legalize/>
    Fractals! <http://www.xmission.com/~legalize/fractals/>

- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

Date: Sun, 21 Jan 2001 09:59:42 -0700
From: Rich <[e-mail]>
Subject: Re: unixscr.c

In article <[e-mail]>,
    Jean-Pierre Demailly <[e-mail]>  writes:

> Is this two year old fix available somewhere?

I did a quick search for it and couldn't find it.  I'd have to look
deeper to find if I still have it as an isolated patch somewhere.  I
am virtually 100% certain I mailed it to Tim a long time ago, but that
was before I started keeping a permanent record of all my outgoing
mail.
- --
Ask me about my upcoming book on Direct3D from Addison-Wesley!
Direct3D Book <http://www.xmission.com/~legalize/book/>
         Home <http://www.xmission.com/~legalize/>
    Fractals! <http://www.xmission.com/~legalize/fractals/>

- --------------------------------------------------------------
Thanks for using Fractdev, The Fractint Developer's Discussion List
Post Message:   [e-mail]
Get Commands:   [e-mail] "help"
Administrator:  [e-mail]
Unsubscribe:    [e-mail] "unsubscribe fractdev"

------------------------------

End of fractdev-digest V1 #39
*****************************