ftp.xmission.com/2014.06.ftp.xmission.com.tar/ftp.xmission.com/pub/lists/fractdev/archive/fractdev.200011

Internet Message Format | 2000-11-28 | 41KB

From: "=?ISO-8859-1?B?UHVza+FzIElzdHbhbiBqci4=?=" <[e-mail]>
Subject: I have code for improved cellular automata
Date: 04 Nov 2000 22:24:00 +0100

I have Borland Pascal code for 2 types of an improved cellular automata.

The first is another version of the linear cellular automata - a bit like
type=cellular already built in - but with improvements: the cell being
recalculated can be excluded from calculating the 'neighborhood', it works
without specifying a subtype: the radius can be specified, and the number of
possible cell numbers is 0 to 35.

The second is a 2D extension: it works in a plane, not a line and it draws
the last state of the plane. The improvements on the linear type are also
there.

I'm looking for someone who could integrate it into Fractint, because I'm
not familiar with the Fractint source. It would need new input routines: the
improved linear type works with true string values - can also contain
characters A..Z, and the range can maybe extended - not values decoded from
a float number, and the 2D version needs a 2D array editor for the 2D 'seed'
array.

If anyone is interested, I can post the source, together with the compiled
files - some of you may not have BP, - it's about 25 kB. I release it for
integrating into Fractint, _NOT_ for making commercial programs!

If you wish to post it to the whole list, please let me know.




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: Fractint bugs of the month
Date: 05 Nov 2000 07:54:55 -0600

Folks,

Here is a list of Fractint bugs that I know about and haven't been able to
fix:

1.  The fractal type Circle locks up the machine when used with
inside=startrail.

2.  Two parser optimization problems:
     (a)  Flipped image when a constant a=1 is used versus a=(1,0).

bad_image {; Jim Muth  Don't try this at home.
 a=1, z=(-1/0.95)^(-1/9.9):
 z=a*(10*(z^(-1.1))+0.95*(z^(-11)))+1/pixel,
 |z| < 500
}

     (b)  Large bailout values cause some areas to be incorrectly identified
as 'inside'.

frm:bailoutproblem {
;p1 is the bailout
bail=real(p1),
z=c=pixel:
z=z*fn1(z)+c
|z|<bail
}

bad_bailout { ; the effect is very strong
  reset=2000 type=formula
  formulaname=bailoutproblem center-mag=0/0/0.6666667
  params=10000000000/0 float=y maxiter=255 inside=0 sound=off
  }

3.  When restoring an incomplete image made with the formula type, if the
formula can't be found, the calculation continues anyway.  Which, obviously,
doesn't create the desired image.  The report I received stated that the
image was disposed of, and this problem occurs with ifs and lsystem also.  I
have not verified the problem with ifs and lsystem.

Please let me know if I've forgotten any obvious bugs.

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: Thore Berntsen <[e-mail]>
Subject: RE: I have code for improved cellular automata
Date: 06 Nov 2000 08:16:02 +0100

I'm interessted in Your BP source.

Thore Berntsen
Norway

-----Original Message-----
Sent: 4. november 2000 22:24


I have Borland Pascal code for 2 types of an improved cellular =
automata.

The first is another version of the linear cellular automata - a bit =
like
type=3Dcellular already built in - but with improvements: the cell =
being
recalculated can be excluded from calculating the 'neighborhood', it =
works
without specifying a subtype: the radius can be specified, and the =
number of
possible cell numbers is 0 to 35.

The second is a 2D extension: it works in a plane, not a line and it =
draws
the last state of the plane. The improvements on the linear type are =
also
there.

I'm looking for someone who could integrate it into Fractint, because =
I'm
not familiar with the Fractint source. It would need new input =
routines: the
improved linear type works with true string values - can also contain
characters A..Z, and the range can maybe extended - not values decoded =
from
a float number, and the 2D version needs a 2D array editor for the 2D =
'seed'
array.

If anyone is interested, I can post the source, together with the =
compiled
files - some of you may not have BP, - it's about 25 kB. I release it =
for
integrating into Fractint, _NOT_ for making commercial programs!

If you wish to post it to the whole list, please let me know.




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"


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

From: "=?iso-8859-1?B?UHVza+FzIElzdHbhbiBqci4=?=" <[e-mail]>
Subject: Re: Fractint bugs of the month
Date: 08 Nov 2000 18:16:20 +0100

I remember I saw this one long ago in v19.2 and it's there in v20.1.1 too.

If arbitrary precision is used with the Mandelbrot-set and the bailout is
higher than the default - like 20 or more - there are some garbage
filaments.


The false inside bug can appear also at lower bailout if a function argument
reaches about 1e6, so it's very likely to be a function failure (like I
suspected.)


This formula produces the effect with p1=2 p2=1000, p1=3 p2=100, p1=6 p2=10.
In all these cases, fn(z^p1) can reach 1e6.

bailoutproblem2 {
;p1 is the exponent, try 1..8
;p2 is the bailout
z=c=pixel:
z=z*fn1(z^p1)+c
|z|<p2
}




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: Fractint bugs of the month
Date: 09 Nov 2000 20:04:36 -0600

> If arbitrary precision is used with the Mandelbrot-set and the bailout is
> higher than the default - like 20 or more - there are some garbage
> filaments.

Yes, I see some problems there.

> The false inside bug can appear also at lower bailout if a function
argument
> reaches about 1e6, so it's very likely to be a function failure (like I
> suspected.)

Except that the bug goes away if you turn off the parser optimizations with
debug=322.

I've tracked it down to the optimization of the 'less than' comparison.  I
just have no clue as to how to fix it.  It would also be a good bet that the
other comparisons have the same problem.

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: "=?iso-8859-1?B?UHVza+FzIElzdHbhbiBqci4=?=" <[e-mail]>
Subject: New (old) Fractint bug
Date: 10 Nov 2000 21:17:35 +0100

I found a bug that is still there in version 20.1.1. The perturbation for
type=mandel is handled incorrectly if inside is set to bof60 or bof61. 1. If
the real part is negative, it is still treated as positive. 2. If it
contains only an imaginary part: (0,n) it is treated as (-|n|,0). This bug
is in present in versions 19.2, 20.0, 20.1.1 and occurs only in inside=bof60
of bof61, not the other inside options.



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: New (old) Fractint bug
Date: 10 Nov 2000 18:45:34 -0600

> I found a bug that is still there in version 20.1.1. The perturbation for
> type=mandel is handled incorrectly if inside is set to bof60 or bof61. 1.
If
> the real part is negative, it is still treated as positive. 2. If it
> contains only an imaginary part: (0,n) it is treated as (-|n|,0). This bug
> is in present in versions 19.2, 20.0, 20.1.1 and occurs only in
inside=bof60
> of bof61, not the other inside options.

Yes, that's a strange feature.  It works like that so that images in "The
Beauty of Fractals" can be reproduced using Fractint.

I have in mind a work around that would require a command line option.  This
would avoid having to add code for backwards compatibility.  And, it would
still allow reproducing the BOF images.

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: "=?windows-1250?B?UHVza+FzIElzdHbhbiBqci4=?=" <[e-mail]>
Subject: Another video hotkey bug
Date: 15 Nov 2000 18:04:45 +0100

I found another bug in the video mode hotkey assignments. I have some
'windowed' entries in the FRACTINT.CFG file, which didn't fit into the old
(removed) width = height = 160.. 2048 limitation. They work without problems
but they can't be assigned to a hotkey and if they are already assigned
(with a text editor) they can't be unassigned. I tried it with several
entries and it seems that it has to do with the old size limits.




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: "=?iso-8859-1?B?UHVza+FzIElzdHbhbiBqci4=?=" <[e-mail]>
Subject: Bug in the incremental redraw mode
Date: 17 Nov 2000 22:42:55 +0100

If the incremental redraw in 'passes=g' is activated and a GIF is saved -
either while it is in progress or after it is completed - the drawing mode
flag in the GIF will be 'passes=1' instead of 'passes=g'. If the image is
saved to a PAR, the drawing mode flag will be bad only if it is saved while
the incremental redraw is in progress.





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 the incremental redraw mode
Date: 17 Nov 2000 18:54:43 -0600

> If the incremental redraw in 'passes=g' is activated and a GIF is saved -
> either while it is in progress or after it is completed - the drawing mode
> flag in the GIF will be 'passes=1' instead of 'passes=g'. If the image is
> saved to a PAR, the drawing mode flag will be bad only if it is saved
while
> the incremental redraw is in progress.

When I implemented the incremental redraw I spent a considerable amount of
time trying to fix this.  As you can see, I was unsuccessful.

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: Another video hotkey bug
Date: 17 Nov 2000 19:41:01 -0600

> I found another bug in the video mode hotkey assignments. I have some
> 'windowed' entries in the FRACTINT.CFG file, which didn't fit into the old
> (removed) width = height = 160.. 2048 limitation. They work without
problems
> but they can't be assigned to a hotkey and if they are already assigned
> (with a text editor) they can't be unassigned. I tried it with several
> entries and it seems that it has to do with the old size limits.

Tell me more.  I tried this with a 10x10 mode, which is the limit on the
small size, and it worked.  I was able to unassign and then reassign a
hotkey.

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 the incremental redraw mode
Date: 18 Nov 2000 07:42:26 -0600

>> If the incremental redraw in 'passes=g' is activated and a GIF is saved -
>> either while it is in progress or after it is completed - the drawing
mode
>> flag in the GIF will be 'passes=1' instead of 'passes=g'. If the image is
>>saved to a PAR, the drawing mode flag will be bad only if it is saved
while
>> the incremental redraw is in progress.

> When I implemented the incremental redraw I spent a considerable amount of
> time trying to fix this.  As you can see, I was unsuccessful.

And, having just said that, I think I've fixed it.

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: Off my rocker
Date: 18 Nov 2000 18:16:47 -0600

I can't work all the time.

isosceles          { ; A triangle, even if artificially formed
                     ;  Jonathan Osuch [e-mail]
                     ;  Nov 18, 2000 at 17:03:53
                     ;  Version 2001 Patchlevel 2
  reset=2001 type=julia(fn||fn) function=exp/cos
  center-mag=+0.33030303030303000/+0.81725554106910180/641.0256/1/-82.5
  params=1.3/0/2.5 float=y maxiter=1000 bailout=100 inside=bof60
  outside=atan logmap=yes sound=off
  colors=000PNH<3>_C7b94cA4<14>r_HsaItcJ<2>whMxjNxiN<17>jNAiM9iK9hJ8gI7fG6\
  <15>lNJlOKlOL<3>nRPnSQnTR<18>mrkmslmtm<2>mxplzrlzp<12>`zI_zGZzD<3>Vy2<27\
  >QY5QX5PW5<2>PT5OS6MVE<3>Afi7iq4lz<4>HpaKpXNqS<2>VtDYu7Ys8<6>c_IdXJeUL<3\
  >iIR<13>sAAt99u97<2>w74x62w62<27>VRGUSHTSHSTIRUIPVJ<13>_VQ
  }



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: Fractint and Windows ME
Date: 18 Nov 2000 18:13:41 -0600

Does anybody have any experience using Fractint with the latest version of
Windows?  One of the latest comments on the Fractint bug list is about
problems with it.

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: Fractint version 20.1.03
Date: 19 Nov 2000 12:16:37 -0600

The latest patch and executable of Fractint version 20.1 are now available.

It fixes the incremental redraw problem with setting passes=1 after being
interrupted.  The range of bailout values that work with ap-math is now
larger.  Still nothing to write home about.  A new command line option has
been added, nobof=yes, which causes the inside=bof60 and bof61 options to
NOT recreate the images from the book of the same name.  This makes the
behavior of these two inside options the same as the other inside options.

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: Thore Berntsen <[e-mail]>
Subject: RE: Fractint and Windows ME
Date: 20 Nov 2000 08:01:23 +0100

I use Fractint with Windows ME and have had no problems. Mostly it is my
Fractint Screen Saver that runs Fractint for me these days, in other words
in Disk-Video mode.

-----Original Message-----
Sent: 19. november 2000 01:14


Does anybody have any experience using Fractint with the latest version of
Windows?  One of the latest comments on the Fractint bug list is about
problems with it.

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"

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: "Morgan L. Owens" <[e-mail]>
Subject: Re: Fractint and Windows ME
Date: 20 Nov 2000 20:22:19 +1300

At 18:13 18/11/2000 -0600, Jonathan Osuch wrote:
>Does anybody have any experience using Fractint with the latest version of
>Windows?  One of the latest comments on the Fractint bug list is about
>problems with it.
It's not ME, but I've had no problems running v20 on Win2000SR1 (yes, a
service release already - how many more, I wonder?). I'm actually using
the same sstools.ini file as my 95 machine (in particular,
textsafe=save), modulo the video mode; makefcfg doesn't know what to do
with my NVidia RIVA TNT, but that's okay, the default .cfg works fine
up to the card's maximum 1600x1200 res.

I haven't tried it with looser textsafe settings - I'm supposed to be
working, and it's not my machine anyway - but the delay is unnoticeable (it
takes longer for the monitor's relays to switch).

Like I say, 2000 is not ME (which is why there is a ME); perhaps the
latter's fun and flashy orientation makes it a scarier platform for
Fractint.

Morgan L. Owens
"I'm not even supposed to be installing software on it."



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: Formula parser
Date: 24 Nov 2000 12:07:33 -0600

Tim,

I tried to compile/assemble the compiler version of the formula parser and
it will no longer assemble.  The ALTER_RET_ADDR macro in parsera.asm tries
to push the address of past_loop, which isn't defined anywhere.

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: Updated float only and xfractint
Date: 24 Nov 2000 12:42:03 -0600

Tim,

I've brought both the float only and the Xfractint sources up to patch 3.

We need to think about how we want to archive these sources.  Either put
them in subdirectories of /ftp (public) or put them under /fractint
(private).  My preference would be for putting them under /fractint to keep
them private, but I don't have any good reason to do it that way.

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: Phil McRevis <[e-mail]>
Subject: Re: Updated float only and xfractint
Date: 24 Nov 2000 15:20:19 -0700


In article <000b01c05646$3ee0e740$0100a8c0@bananasenior>,
    "Jonathan Osuch" <[e-mail]>  writes:

> I've brought both the float only and the Xfractint sources up to patch 3.

Is this the code with my device driver stuff in it?  I poked at some
of the zip files on <http://www.fractint.org/ftp/> and didn't see
anything that looked like it had my changes.  What file do I look at?
--
Ask me about my upcoming book on Direct3D from Addison-Wesley!
<http://www.xmission.com/~legalize/>    Legalize Adulthood!
    ``Ain't it funny that they all fire the pistol,
      at the wrong end of the race?''--PDBT
[e-mail]    <http://www.xmission.com/~legalize/who/>

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: Updated float only and xfractint
Date: 25 Nov 2000 11:43:10 -0600

> Is this the code with my device driver stuff in it?  I poked at some
> of the zip files on <http://www.fractint.org/ftp/> and didn't see
> anything that looked like it had my changes.  What file do I look at?

No, currently we are keeping that code between Tim and myself.  It hasn't
changed much since September.  About that time an update to the Allegro WIP
source came out.  I was unable to get the new Allegro source to compile
using djgcc and then couldn't get the previous version to work either.
Under Linux I have a version of Xfractint working that doesn't yet save the
video modes to fractint.cfg so it has to cycle through the video modes each
time it starts up.  That's pretty ugly.

If you like, I can email my latest source to you.  Or, if you would prefer I
can put it at
<http://www.fractint.org/ftp/> with a cryptic file name and password
protected.

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: Proposal for restructuring of Xfractint's Makefile
Date: 26 Nov 2000 01:09:19 -0600

Jonathan and anyone else concerned,

I am finally getting back to Xfractint after a rather long absence (for
various reasons.) During that time, I read a very informative article in the
November issue of "Linux Journal". This article dealt with where
user-installed programs should be installed in the Linux directory structure.
The author stated that there are four standards, along with the UNIX
community in general, that have established the /opt directory for add-on
software.

I run Linux-Mandrake on my machine, and all the KDE & Gnome binaries are in
/usr/bin and/or /usr/X11R6/bin. But then again, technically they aren't
user-installed programs. X programs are in /usr/X11R6/bin. The only
user-installed program I have in my /opt directory is Adobe's Acrobat Reader,
which wasn't included in the installation CD. I believe RedHat's install
program puts KDE in /opt, but I'm not sure. (I think they use to in Redhat
Linux 6.2.)

My proposal is this:
While I am updating Xfractint on my computer, I would like to modify the
Makefile to reflect the changes mentioned above, so we can try to adhere to
the Linux Standard Base, and other UNIX standards. I propose making the
following changes:
The binary be placed in /opt/xfractint-20.x.xx/bin/
The .par, .map, .frm, etc files be placed in /opt/xfractint-20.x.xx/
Any configuration files be placed in /etc/opt/xfractint-20.x.xx/
Any files created by the user will be placed in /home/user/xfractint-20.x.xx/
The man file should remain where it is currently installed.

The Makefile could also be modified to check for and change the xfractint
version number in the dir. names during an upgrade.
I would also like to update the read.me file with new compiling/install
instructions.

Of course, I will thoroughly test all changes that I make before submitting
it as a patch.

Comments and questions are welcome.
Scott Boyd
--
[e-mail]
http://home.swbell.net/sdboyd56/
-----------------------
(Wanted: a new, clever tagline...)

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: Proposal for restructuring of Xfractint's Makefile
Date: 26 Nov 2000 08:04:51 -0600

Scott,

> I propose making the following changes:
> The binary be placed in /opt/xfractint-20.x.xx/bin/

That seems reasonable.  Would there be a problem with using /home/user/bin?
What about the source file location?  I know you are thinking about
distributing a pre-compiled executable, but while we are changing things...

> The .par, .map, .frm, etc files be placed in /opt/xfractint-20.x.xx/

That could get very messy.  How about putting them in
/opt/xfractint-20.x.xx/par,
/opt/xfractint-20.x.xx/map, /opt/xfractint-20.x.xx/frm, etc.

> Any configuration files be placed in /etc/opt/xfractint-20.x.xx/

Yes.  We could put an sstools.ini file here with the above directories
already set.  I assume that Xfractint will be able to find it here???

> Any files created by the user will be placed in
/home/user/xfractint-20.x.xx/

Perhaps we would want to place the par, map, frm, etc files under here also?

> The man file should remain where it is currently installed.

Yes.  It is in desperate need of an update, also.

> The Makefile could also be modified to check for and change the xfractint
> version number in the dir. names during an upgrade.

Maybe we don't want to do this.  There may be reason to have more than one
version loaded.  For example, if a feature (or bug) is removed in a newer
version and the user still wants to be able to generate fractals using it.
The question is, how do we get the desired version to execute?

> I would also like to update the read.me file with new compiling/install
> instructions.

Very much needed.

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]
Subject: RE: Re: Proposal for restructuring of Xfractint's Makefile
Date: 27 Nov 2000 0:18 0000


Hi Folks,

Sounds good so far, I've just got Debian 2.2 installed and got
the latest xfractint to compile OK with no problems, barely had
to tweak the makefile :-)

One very minor niggle...

>> Any configuration files be placed in /etc/opt/xfractint-20.x.xx/

>Yes.  We could put an sstools.ini file here with the above
>directories

Sstools.ini doesn't belong exclusively to FractInt ( though I'll
admiit that I can't think of any other stone soup tools ported to
linux ) and so perhaps should be in the users home directory.

As I said, a very minor niggle :-)

Cheers,
         Robin.


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: Re: Proposal for restructuring of Xfractint's Makefile
Date: 26 Nov 2000 21:14:15 -0600

On Sunday 26 November 2000 08:04, you wrote:
> Scott,
>
> > I propose making the following changes:
> > The binary be placed in /opt/xfractint-20.x.xx/bin/
>
> That seems reasonable.  Would there be a problem with using /home/user/bin?
>
Yes and/or no. In a one-user system, such as a home-computer, that would
work. But in a multi-user environment, such as at a school, then only that
user could use the binary. That's why presently it is put in /usr/X11R6/bin,
which is a dir that is readable/executable by everyone.

> What about the source file location?  I know you are thinking about
> distributing a pre-compiled executable, but while we are changing things...
>
When you unzip a source package, it normally goes in /usr/src/package-name/.
Then the makefile would copy all the files to their prescribed dirs. I was
also thinking along the lines of having the Makefile that goes in the
original source package reflect the changes I propose. In other words, I
would be submitting the modified makefile to you for inclusion into the
Xfractint source.

> > The .par, .map, .frm, etc files be placed in /opt/xfractint-20.x.xx/
>
> That could get very messy.  How about putting them in
> /opt/xfractint-20.x.xx/par,
> /opt/xfractint-20.x.xx/map, /opt/xfractint-20.x.xx/frm, etc.
>
As it is right now, all the .par, .map, .frm files are all dumped in one
directory anyway! But your suggestion sounds better to me. Would make them
easier to browse thru. But then we would have to make sure that Xfractint
could find them in the correct directories.

> > Any configuration files be placed in /etc/opt/xfractint-20.x.xx/
>
> Yes.  We could put an sstools.ini file here with the above directories
> already set.  I assume that Xfractint will be able to find it here???
>
That also would need to be set in the program and/or the Makefile. I'll have
to check and see where the setting is for Xfractint to look for it's config
files.

> > Any files created by the user will be placed in
>
> /home/user/xfractint-20.x.xx/
>
> Perhaps we would want to place the par, map, frm, etc files under here
> also?
>
My reply above about placing the binary under /home/user/ vs
/opt/package-name also applies here. Of course, any of those files that a
user creates could be placed under their home dir.

> > The man file should remain where it is currently installed.
>
> Yes.  It is in desperate need of an update, also.
I'll keep that in mind also.
>
> > The Makefile could also be modified to check for and change the xfractint
> > version number in the dir. names during an upgrade.
>
> Maybe we don't want to do this.  There may be reason to have more than one
> version loaded.  For example, if a feature (or bug) is removed in a newer
> version and the user still wants to be able to generate fractals using it.
>
Good point! Then it would be best to put the newer version in a seperate dir,
with the dir name containing the version number.

> The question is, how do we get the desired version to execute?
>
Simple - create a link ("shortcuts" in MS-speak) to each seperate binary. The
executable filename can be renamed with the version no. tacked on the end.

I haven't started any of the proposed changes yet, so there's stil lots of
room for ideas, changes, suggestions, discussions, etc. And remember, my
proposal is only to nudge Xfractint more in line with several standards and
the Linux/Unix community in general, and so new users wouldn't have to tweak
the Makefile or move files around after it's compiled/instaled.

Scott Boyd
--
[e-mail]
http://home.swbell.net/sdboyd56/


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]
Subject: RE: Re: Re: Proposal for restructuring of Xfractint's Makefile
Date: 27 Nov 2000 18:19 0000

Hi Scott,

>What? There's some other Stone Soup Group tools out there?

There's piclab IIRC but I don't know if it's available in linux,
probably not.(might be an easy port though

Though there is still the point that different users
will want different sstools.ini settings, so it should still go under
the users home somewhere.

Cheers,
         Robin
P.S. if you go to:
http://sources.redhat.com/autobook/
you'll find a link to an online guide to autoconf, automake and
libtool... good stuff!




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: Re: Re: Re: Proposal for restructuring of Xfractint's Makefile
Date: 27 Nov 2000 21:00:07 -0600

On Monday 27 November 2000 12:19, you wrote:
> Hi Scott,
>
> >What? There's some other Stone Soup Group tools out there?
>
> There's piclab IIRC but I don't know if it's available in linux,
> probably not.(might be an easy port though
>
Yeah - that's the name of it! I was thinking of that program using
sstools.ini, but I couldn't remember the name. I didn't know it was a Stone
Soup program though. I haven't even heard anything about it in a long time,
so you are more than likely right about it not being ported to Linux.

> Though there is still the point that different users
> will want different sstools.ini settings, so it should still go under
> the users home somewhere.
>
Good point. I also just thought of how a Linux installation will have a
global .bashrc and .bash_profile in the /etc, dir., but each user can also
have their custom .bashrc and .bash_profile in their home dir. The same could
be done with Xfractint config files....

> P.S. if you go to:
> http://sources.redhat.com/autobook/
> you'll find a link to an online guide to autoconf, automake and
> libtool... good stuff!

I'll look into it soon.

Regards,
Scott

--
[e-mail]
http://home.swbell.net/sdboyd56/
A computer without a Microsoft operating system
is like a dog without bricks tied to it's head.

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: Proposal for restructuring of Xfractint's Makefile
Date: 27 Nov 2000 21:26:48 -0600

Scott wrote:

> Yeah - that's the name of it! I was thinking of that program using
> sstools.ini, but I couldn't remember the name. I didn't know it was a Stone
> Soup program though. I haven't even heard anything about it in a long time,
> so you are more than likely right about it not being ported to Linux.

Piclab definitely is a Stone Soup program, and definitely hasn't
been (and won't be) ported to Linux. Strictly a DOS program, uses
assembler. Lee Crocker's web page doesn't even mention his
Piclab program (see www.piclab.com).

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: Tim Wegner <[e-mail]>
Subject: piclab at piclab.com
Date: 27 Nov 2000 21:32:14 -0600

You have to search, but Lee's piclab site *does* have piclab!!

http://www.piclab.com/pub/index.html

There's also some code that we used in implementing Newton in
Fractint.

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: Tim Wegner <[e-mail]>
Subject: Re: Piclab
Date: 27 Nov 2000 23:27:18 -0600

Scott asked (private email):

> I can't remember what Piclab did. Can you tell me in a nutshell? All I can
> remember hearing about it, was that it came out about the same time as
> Fractint did, ie: eons ago in computer-world terms.  8-)

Piclab is a wonderful command-line image processing program.
Let's you make command files that consistently and repeatably do
various image processing tasks. For example, I used it to increase
the contrast of all the images for my book.

It's obsolete because it is a DOS program with no GUI. Lee
stopped supporting it a long time ago. Bert Tyler and I actually did
a bit of maintenance a while ago - I don't think the version at Lee's
site is actually the latest one.

I wrote a lot about piclab in my Image Lab book.

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: "Thierry B." <[e-mail]>
Subject: Re: Piclab
Date: 28 Nov 2000 08:14:31 +0000

Tim Wegner wrote:
>
> It's obsolete because it is a DOS program with no GUI. Lee
> stopped supporting it a long time ago. Bert Tyler and I actually did
> a bit of maintenance a while ago - I don't think the version at Lee's
> site is actually the latest one.
>

    A long time ago, I've tried to port Piclab on Unix.
    I give up because I have not enough time and understanding
    the code (specially IO) was, hem, hard :)

    But I can make a new trial... Where are the sources ?


    not related: is this list archived somewhere ?

--
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"


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

From: "Paul N. Lee" <[e-mail]>
Subject: Re: Piclab
Date: 28 Nov 2000 19:48:45 -0600

Thierry B. wrote:
>
> not related: is this list archived somewhere ?
>

You might try the following:

     ftp://ftp.xmission.com/pub/lists/fractdev/archive/

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"