ftp.xmission.com/2014.06.ftp.xmission.com.tar/ftp.xmission.com/pub/lists/fractdev/archive/fractdev.200102

Internet Message Format | 2001-02-27 | 23KB

From: "Scott D. Boyd" <[e-mail]>
Subject: Colormap differences between Fractint and Xfractint
Date: 01 Feb 2001 19:56:58 -0600

I apologize if this email gets posted to the list twice....

It seems that there is a difference in the way Fractint 20.0 and Xfractint
20.1.04 interpret a colormap. This was discovered when Paul N. Lee and
myself both created Jim Muth's FOTD for 1-31-01. Paul's and Jim's versions
of the image were identical, mine, created with Xfractint, was different.
I downloaded and installed Fractint 20.0 on a Win95 machine I have, and
created the image, using Jim's .par file and colormap file. As expected,
it turned out looking like Paul's and Jim's images.

I have put up a page that shows both images, with text explaining which
one's which. The URL is:
http://home.swbell.net/sdboyd56/fotd/comparison.html

By the way: using disk video mode, Xfractint created the same image in
both 8-bit Xserver and 16-bit Xserver. Just so you'll know that isn't
causing the discrepancy.

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"


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

From: "Damien M. Jones" <[e-mail]>
Subject: Re: Colormap differences between Fractint and Xfractint
Date: 01 Feb 2001 21:33:04 -0500

Scott,

Were I to guess I'd say it's not the colormap that's the issue, it's the
formula parser itself. There are probably differences in overflow between
the two versions, DOS and X, and that's causing things to behave
erratically. This image uses bof60 inside coloring, and just about
everything you see in the image comes from that.

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"


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

From: "Paul N. Lee" <[e-mail]>
Subject: Re: Colormap differences between Fractint and Xfractint
Date: 01 Feb 2001 20:43:11 -0600

Scott D. Boyd wrote:
>
> It seems that there is a difference in the way
> Fractint 20.0 and Xfractint 20.1.04 interpret a
> colormap.

For informational purposes, I have tried this under Win98SE in a DOS
window using versions 20.1.0.5 and 20.1.0.6 with the same results.

It appears there is also a problem when saving the colors within the PAR
file, as mentioned by Jim in his posting.

[e-mail] wrote:
>
> The parameter file renders in 8-1/2 minutes, though I strongly
> advise setting the color palette by using the color map file
> attached to the end of this letter.  Due to a technical limitation,
> the colors as recorded in the parameter file are slightly but
> significantly incorrect.  This is the first time this limitation has
> actually caused me to record the colors in the more accurate
> *.map format.
>

Sincerely,
P.N.L.
http://www.fractalus.com/cgi-bin/theway?ring=fractals&id=43&go

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"


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

From: "Jonathan Osuch" <[e-mail]>
Subject: Re: Colormap differences between Fractint and Xfractint
Date: 01 Feb 2001 21:06:34 -0600

Scott,

> It seems that there is a difference in the way Fractint 20.0 and Xfractint
> 20.1.04 interpret a colormap.

One of these days, we'll be able to correctly read in that bailout value.
Then the **** will hit the fan.

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"


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

From: Tim Wegner <[e-mail]>
Subject: Rich's X changes
Date: 02 Feb 2001 21:20:56 -0600

Thank goodness I found the email with Rich Thomson's changes.
You can find it at:

http://www.fractint.org/ftp/experimental/rich.txt

I'll have a look this weekend and see if I can convert it to a diff
against the current version. The email tells the version Rich started
with.

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"


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

From: "Gerald K. Dobiasovsky" <[e-mail]>
Subject: Bug in julia_inverse
Date: 03 Feb 2001 23:22:22 +0100

(Considering who's producing all those patches,
 this goes to Jonathan Osuch, I guess...)

Hi,

Fractal type julia_inverse stops after plotting
a few dots when tree-traversal is set to breadth
(random and depth left and depth right seem to
be ok).

This happens not only in version 20.1.6, but in
the original 20.0, too. Version 19.6 works ok.

My apologies if this is already known...


Gerald



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"


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

From: Tim Wegner <[e-mail]>
Subject: Re: Bug in julia_inverse
Date: 03 Feb 2001 17:51:00 -0600

Gerald wrote:

> Fractal type julia_inverse stops after plotting
> a few dots when tree-traversal is set to breadth
> (random and depth left and depth right seem to
> be ok).
>
> This happens not only in version 20.1.6, but in
> the original 20.0, too. Version 19.6 works ok.

Thanks, we'll look into it. The fact that i8t used to work should help
track it down.

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"


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

From: "Jonathan Osuch" <[e-mail]>
Subject: Re: Bug in julia_inverse
Date: 03 Feb 2001 20:45:20 -0600

Gerald,

> This happens not only in version 20.1.6, but in
> the original 20.0, too. Version 19.6 works ok.

Version 20.0 works okay for me.  This problem appears with patch 1 to
version 20.0.  I don't see any obvious reason for this to happen.

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"


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

From: "Jonathan Osuch" <[e-mail]>
Subject: Re: Bug in julia_inverse
Date: 04 Feb 2001 07:55:05 -0600

Gerald,

> Fractal type julia_inverse stops after plotting
> a few dots when tree-traversal is set to breadth
> (random and depth left and depth right seem to
> be ok).

I found it.  The fix will be in my next patch.

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"


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

From: "Gerald K. Dobiasovsky" <[e-mail]>
Subject: Re: Bug in julia_inverse
Date: 04 Feb 2001 22:57:57 +0100

Jonathan,

Thanks to you (and Tim Wegner) for the immediate
response.

Today I stumbled upon another (tiny) bug.

Setting the background color for true color .TGA
output from the command line ("background=...")
results in "Oops. I couldn't understand..." for
legal values of red/green/blue with the
exception of 0/0/0.

In CMDFILES.C (line 2733):

   if (far_strcmp(variable,s_background) == 0) {     /* background=?/?
*/
      if (totparms != 3 || intparms != 3) goto badarg;
                for (i=0;i<3;i++)
                        if (intval[i] & 0xff)
                                goto badarg;

The bitmask 0xff should read ~0xff, I'd say.


Gerald


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"


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

From: "Jonathan Osuch" <[e-mail]>
Subject: Re: Bug in julia_inverse
Date: 04 Feb 2001 18:35:17 -0600

Gerald,

> The bitmask 0xff should read ~0xff, I'd say.

Yes, good catch.  I'll put it in the next patch.  Thanks.

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"


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

From: [e-mail]
Date: 10 Feb 2001 11:00:41 -0700

2001
10:34:05 -0700
[193.54.241.89])
    by zambeze.ujf-grenoble.fr (Pro-8.9.3/8.9.3/Configured  by
AD & JE 25/10/1999) with ESMTP id SAA00934
    for <[e-mail]>; Sat, 10 Feb 2001
18:34:00 +0100 (MET)
Message-Id: <[e-mail]>
Cc: [e-mail] (Jean-Pierre Demailly)
X-Mailer: ELM [version 2.4ME+ PL47 (25)]
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary=ELM981826445-23293-0_
Content-Transfer-Encoding: 7bit
Sender: [e-mail]
Precedence: bulk
Reply-To: fractdev

(Note: this message bounced because the attached patch was too
big. I removed the diff from this message and have posted it using
my list admin super powers :-). I have the patch and and will look
into it this weekend and will post it at fractint.org. Thanks Jean-
Pierre! Tim)

Hi:

Please find in attachment a patch to apply against xfractint-
20.1.06.
I've merged some of my changes suggested last January with Rich's patch
(as proposed in April 1999). Although Rich's patch was against v19.61,
most of his changes could still be applied to v20.1.06.

In case you would like to get the whole modified package, it is available
on our FTP server, at the URL
ftp://ftp.ac-grenoble.fr/ge/mathematics/fractals/xfractint-20.1.06exp.tgz
(by the way, you can find there a lot of open source software,
especially in relation with education or sciences).

A brief explanation of what the patch should do:

- Incorporate Rich's fixes, so that xfractint now properly handles
  TrueColor/DirectColor visuals (what I suggested in January also
  produced more or less the same results, but I didn't realized at
  the time that the dacbox[] data in unixscr.c  were the only data
  needed for handling the color palettes; so I introduced special
  unnecessary extra data for that purpose, which were redundant
  with the dacbox...)

- introduce some built-in palettes, actually 50 different ones
  that can be interactively changed through six new (hopefully not
  otherwise used?) keys:
     0     (return to default)
     !     (invert colors)
     < >   (cycle by -1 or +1 step)
     ( )   (cycle by -6 or +6 steps)
  This is done by adding a few lines to the initdacbox() routine
  and by slightly extending the menu choices in the ev_key_press()
  routine.

  In relation to this, maybe a point where fractint could be improved:
  rather than just loading palettes once for all as lists of RGB values
  (which by the way takes a substantial disk or memory space), there
  should also be a way of programming the initdacbox() routine from
  within xfractint, via mathematical algorithms (a list of which could
  be hardcoded as I did). I think it's nice to be able to change the
  palette of colors with just one click or one key press...

- Fix in encoder.c, so that saving a *.gif image incorporates the dacbox()
  data selection, exactly as it appears on screen.

- Some additions to unixscr.c, so that resizing the main window results
  in the image being updated with the new size.

- Bug fixes:
  * replaced "typedef char S8;" by "typedef signed char S8;"
    in the #ifdef unix case of "port.h"
    Otherwise xfractint didn't compile properly under LinuxPPC -- and
    (probably) wouldn't either on other bigendian architectures
    (as a result, the ncurses interface was broken...). That fix
    doesn't seem to affect smallendian machines.
  * removed an "XFreeColors()" call that caused xfractint to hang up
    when changing the palette
  * removed  free(Ximage->data)  which was redundant with the subsequent
    XDestroyImage(Ximage), and caused xfractint to segfault when
    resizing the window.
  * in unixscr.c again, replaced
       font_pixmap = XCreatePixmap(Xdp, Xw, 64, 8, 1);
    which caused the "File saved..." message to fail in TrueColor visuals,
    by
       font_pixmap = XCreatePixmap(Xdp, Xw, 64, 8, Xdepth);

Jean-Pierre Demailly


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"


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

From: "Lee H. Skinner" <[e-mail]>
Subject: FOTD for Feb. 10
Date: 10 Feb 2001 17:37:54 -0700

Jonathan,

   Today's FOTD seems to run fine with 20.01.2, but only generates a
blue screen with 20.01.6.  It must be broken.

Lee

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"


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

From: "Lee H. Skinner" <[e-mail]>
Subject: Re: FOTD for Feb. 10
Date: 10 Feb 2001 18:26:33 -0700

Jonathan,

>>   Today's FOTD seems to run fine with 20.01.2, but only generates a
blue screen with 20.01.6.  It must be broken.  <<

I goofed!  I accidentally miscopied part of the par file when I tried
running it
with 20.01.6.    It's OK after all.  - Sorry!

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"


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

From: "Jonathan Osuch" <[e-mail]>
Subject: Re: FOTD for Feb. 10
Date: 10 Feb 2001 21:06:32 -0600

Lee,

>>
>>   Today's FOTD seems to run fine with 20.01.2, but only generates a
blue screen with 20.01.6.  It must be broken.  <<

I goofed!  I accidentally miscopied part of the par file when I tried
running it
with 20.01.6.    It's OK after all.  - Sorry!
<<

That's okay.  I've been taking a break from Fractint for the last week or
so.  I almost have another patch ready, but I haven't been able to
concentrate lately.

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"


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

From: "Scott D. Boyd" <[e-mail]>
Subject: Comments on the truecolor enhancements to Xfractint
Date: 16 Feb 2001 13:23:51 -0600

I have finally found the time to compile Xfractint 20.1.06 with the
truecolor patch from Jean-Pierre Demailly. (Credit also goes to Rich for
his mods submitted a long time ago.)

My opinion on this latest patch: FANTASTIC!  IT WORKS GREAT!!

I created two recent FOTD's, and they looked identical to the images
created with Xfractint 20.1.04 with my Xserver in 8-bit color mode. I also
loaded a colorful fractal I had created back in Dec., and zoomed in
various areas, it all worked well.

I am using XFree86 4.0.1 in 16-bit color mode, and KDE 2.0. I have found
that I no longer have to use the "-private" option, as Xfractint works
fine either way. Also, the annoying color-switching of other visuals on
the screen, when swtching between the two windows that Xfractint uses, has
now been eliminated.

I haven't yet tested the enhancement while running X in 8-bit color. (I'd
just as soon forget running in 8-bit color mode.  KDE 2.0 is designed for
16-bit or better color modes.)

I'd like to recommend that this patch be incorporated ASAP!

Tim - I also compiled in the patch that I had sent to you back in December.

Scott Boyd
--
[e-mail]
http://sdboyd.penguinpowered.com/~sdboyd/
"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"


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

From: Rich <[e-mail]>
Subject: Re: Comments on the truecolor enhancements to Xfractint
Date: 16 Feb 2001 13:54:33 -0700


In article <01021613235100.01655@earth>,
    "Scott D. Boyd" <[e-mail]>  writes:

> I am using XFree86 4.0.1 in 16-bit color mode, and KDE 2.0. I have found
> that I no longer have to use the "-private" option, as Xfractint works
> fine either way. Also, the annoying color-switching of other visuals on
> the screen, when swtching between the two windows that Xfractint uses, has
> now been eliminated.

In 16-bit mode, there is no colormap so -private essentially does
nothing and that's also why there's no colormap flashing.

> I haven't yet tested the enhancement while running X in 8-bit color. (I'd
> just as soon forget running in 8-bit color mode.  KDE 2.0 is designed for
> 16-bit or better color modes.)

In 8-bit mode you'll probably need to use -private and again you'll
see colormap flashing.
--
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"


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

From: Tim Wegner <[e-mail]>
Subject: Re: Comments on the truecolor enhancements to Xfractint
Date: 16 Feb 2001 21:26:56 -0600

Scott wrote:

> My opinion on this latest patch: FANTASTIC!  IT WORKS GREAT!!
> I'd like to recommend that this patch be incorporated ASAP!
>
> Tim - I also compiled in the patch that I had sent to you back in December.

OK, I'll publish both together this weekend.

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"


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

From: "Jonathan Osuch" <[e-mail]>
Subject: Re: Comments on the truecolor enhancements to Xfractint
Date: 17 Feb 2001 06:58:24 -0600

>> My opinion on this latest patch: FANTASTIC!  IT WORKS GREAT!!
>> I'd like to recommend that this patch be incorporated ASAP!
> >
> >Tim - I also compiled in the patch that I had sent to you back in
December.

> OK, I'll publish both together this weekend.

I would prefer that we fix the problem with 8-bit graphics modes first.

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"


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

From: "Jonathan Osuch" <[e-mail]>
Subject: Patch 7 coming soon
Date: 27 Feb 2001 21:04:14 -0600

I have the truecolor support in Xfractint working in 8, 15, 16, and 24-bit
color depths, and will try to put a patch together this weekend.  I suspect
32-bit also works but am unable to test it.

One remaining problem/question is how the images get saved.  Or restored.  I
haven't looked at this yet.

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"