ftp.xmission.com/2014.06.ftp.xmission.com.tar/ftp.xmission.com/pub/lists/fractdev/archive/fractdev.9807

Internet Message Format | 1998-07-31 | 97KB

From: "Peter Gavin" <[e-mail]>
Subject: RE: (fractdev) Visual C/C++ 5
Date: 13 Jul 1998 22:01:38 -0400

About a month ago you said you got fractint to compile with visual C++...  I
was wondering if you could tell me how you did it.  Does it run faster when
compiled for 32 bit?

Thanks in advance,

Pete
<[e-mail]>



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: "Peter Gavin" <[e-mail]>
Subject: RE: (fractdev) Visual C/C++ 5
Date: 13 Jul 1998 22:08:22 -0400

Sorry, that last message was meant for Tim... <g>

Pete
<[e-mail]>

//  -----Original Message-----
//  From: [e-mail]
//  [mailto:[e-mail]]On Behalf Of Peter Gavin
//  Sent: Monday, July 13, 1998 10:02 PM
//  To: [e-mail]
//  Subject: RE: (fractdev) Visual C/C++ 5
//
//
//  About a month ago you said you got fractint to compile with
//  visual C++...  I
//  was wondering if you could tell me how you did it.  Does it run
//  faster when
//  compiled for 32 bit?
//
//  Thanks in advance,
//
//  Pete
//  <[e-mail]>
//
//
//
//  --------------------------------------------------------------
//  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: "Tim Wegner" <[e-mail]>
Subject: RE: (fractdev) Visual C/C++ 5
Date: 15 Jul 1998 18:00:39 -0600

Pete asked:

> About a month ago you said you got fractint to compile with visual C++...  I
> was wondering if you could tell me how you did it.  Does it run faster when
> compiled for 32 bit?

This wasn't me. Fractint can't be compiled under Visual C/C++ without massive
changes. There is someone trying it, I haven't heard lately how it's going. I will
be making a Visual C/C++ version at my own glacial speed :-).

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: [e-mail] (Kragen)
Subject: (fractdev) new L-system computation method
Date: 15 Jul 1998 20:46:04 -0400 (EDT)

I've come up with a better way of computing L-system strings.  On my
5x86-133, I can compute the order-21 string for the 'Dragon' L-system
from fractint.l in 17 seconds.  The string is 25 megabytes.

The biggest advantage of this, I think, is that the beginning of the
string is available for plotting almost instantly -- you don't have to
wait for the calculation to complete.

I haven't tried actually plotting the output from it.

The source (in convenient standalone program form -- and I think it
should run OK on platforms other than Linux) is at
<URL:http://www.pobox.com/~kragen/sw/fastlsys.c>.

Does anyone think this would be nice to have in Fractint 20?

I'm also going to try getting it into LPARSER.

Kragen


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] (Kragen)
Subject: (fractdev) ping?
Date: 17 Jul 1998 17:32:46 -0400 (EDT)

Well, I sent out a message to this list a few days ago.  No one
responded, and no one even visited the URL I listed.  Is anyone
actually on the list, or were they all bored?

Kragen


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: (fractdev) ping?
Date: 17 Jul 1998 20:16:37 -0600


In article <[e-mail]>,
    [e-mail] (Kragen)  writes:
> Well, I sent out a message to this list a few days ago.  No one
> responded, and no one even visited the URL I listed.  Is anyone
> actually on the list, or were they all bored?

Looks like a good idea to me; I've thougth the same thing about
fractint's L-system handling.  One disadvantage of preorder traversal
is that you must compute the bounding box of the resulting endstring
by an analysis of the rules.  I think this is why breadth first
traversal was done so that the bounding box could be computed easily.
--
http://www.xmission.com/~legalize/    Legalize Adulthood!
[e-mail]
``Ain't it funny that they all fire the pistol,     <URL: http://
  at the wrong end of the race?''--PDBT     www.eden.com/~thewho>

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] (Kragen)
Subject: Re: (fractdev) ping?
Date: 17 Jul 1998 23:34:47 -0400 (EDT)

On Fri, 17 Jul 1998, Phil McRevis wrote:
> In article <[e-mail]>,
>     [e-mail] (Kragen)  writes:
> > Well, I sent out a message to this list a few days ago.  No one
> > responded, and no one even visited the URL I listed.  Is anyone
> > actually on the list, or were they all bored?
>
> Looks like a good idea to me; I've thougth the same thing about
> fractint's L-system handling.  One disadvantage of preorder traversal
> is that you must compute the bounding box of the resulting endstring
> by an analysis of the rules.  I think this is why breadth first
> traversal was done so that the bounding box could be computed easily.

I've thought a lot about that.  I don't really think it's possible to
compute the bounding box by analysis of the rules, at least, not any
more easily than by just running the L-system.

I'll have to hack it into xfractint and see what the profiling says; it
may be that it's feasible to resize the bounding box on the fly, or to
compute the L-system twice in order to know the bounding box for
plotting the second time.

btw, it runs in almost-constant space; the traversal stack is never
deeper than your order.  So you can generate 25MB of L-system string
while taking up 236K of RAM.

On today's machines, it might actually be faster to recalculate the
string rather than fetching it from memory -- but I'll have to try it
before I assert that's really true.

Kragen


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: (fractdev) ping?
Date: 18 Jul 1998 10:11:48 -0600


In article <[e-mail]>,
    [e-mail] (Kragen)  writes:
> I've thought a lot about that.  I don't really think it's possible to
> compute the bounding box by analysis of the rules, at least, not any
> more easily than by just running the L-system.

Suppose you have a rule like F => F+F-F-F+F, which looks roughly like
this:


    +----+  => +----+    +----+
                    |     |
                    |     |
                    +----+

So this rule takes a 1x1 unit bounding box and transforms it into a
3x1 unit bounding box.  This means every time you replace an 'F' in
the string you're going to essentially be expanding the bounding box
by a factor of 3 in the orientation of the 'F' you're replacing.  You
can build up appropriate scale factors for every rule in your set and
use induction to show what the final bounding box will be for any
complete expansion to a certain level.
--
http://www.xmission.com/~legalize/    Legalize Adulthood!
[e-mail]
``Ain't it funny that they all fire the pistol,     <URL: http://
  at the wrong end of the race?''--PDBT     www.eden.com/~thewho>

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] (Kragen)
Subject: Re: (fractdev) ping?
Date: 18 Jul 1998 23:37:10 -0400 (EDT)

On Sat, 18 Jul 1998, Phil McRevis wrote:
> In article <[e-mail]>,
>     [e-mail] (Kragen)  writes:
> > I've thought a lot about that.  I don't really think it's possible to
> > compute the bounding box by analysis of the rules, at least, not any
> > more easily than by just running the L-system.
>
> Suppose you have a rule like F => F+F-F-F+F, which looks roughly like
> this:
>
>
>     +----+  => +----+    +----+
>                     |     |
>                     |     |
>                     +----+
>
> So this rule takes a 1x1 unit bounding box and transforms it into a
> 3x1 unit bounding box.  This means every time you replace an 'F' in
> the string you're going to essentially be expanding the bounding box
> by a factor of 3 in the orientation of the 'F' you're replacing.

It seems to me that knowing where every 'F' you're replacing is, and
what its orientation and scale is, is as difficult as running the
L-system.  Maybe I haven't understood correctly?

> You
> can build up appropriate scale factors for every rule in your set and
> use induction to show what the final bounding box will be for any
> complete expansion to a certain level.

As long as you know where each of those rules gets applied, no?

Please! prove me wrong! A simple and quick way to go from a general
(or, heck, even fairly restricted) L-system to a bounding box for order
N would be a godsend.

Kragen


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: "Peter Gavin" <[e-mail]>
Subject: (fractdev) The PAR & FRM repository
Date: 21 Jul 1998 01:20:00 -0400

Thanks to anyone who sent comments referring me to where I can get their
PAR/FRM files; however, it would be easiest for me if you could email a
direct link to the file if it is on the WWW or available by FTP (as in
http://www.fractalus.com/); or for you email the file(s) to me.  If I have
to go all over the internet searching for the files, even if it just
involves going to a web site and clicking a link, it still means I have to
wait for the site to load.  Once or twice isn't so bad, but several times
is.

Thanks!

Peter Gavin
<[e-mail]>

// End transmission


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: "Peter Gavin" <[e-mail]>
Subject: RE: (fractdev) The PAR & FRM repository
Date: 21 Jul 1998 01:28:19 -0400

Sorry about that last (irrelevant) post, but I accidentally clicked
"Fractdev" in mail address book rather than "Fractint". <g>



Pete
<[e-mail]>


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: "Peter Gavin" <[e-mail]>
Subject: (fractdev) RE: (fractint) Ultra Fractal 2.0 beta 5
Date: 23 Jul 1998 03:15:06 -0400

//  Ah.. seems the fractint developers have something to catch up with... :)
//
//  Christian
//
//  :* MDI (multiple documents) and multi-threading
//  :* Multiple layers, with all Photoshop merge modes and
//  :  alpha channel merging
.
.
.
[advertisement spam snipped]

I'll be impressed when this thing does deep zooms.  Then you can say
FractInt has catching up to do. ;)


BTW, if someone is planning on/has begun developing a Win32 version of
FractInt, I've been working on an arbitrary precision class (after debating
the pros and cons of FFTs and NTTs for some time) and if you are interested,
I would be willing to share it with you after I've got enough completed on
it.  I plan on sticking it into am MFC extension DLL.  Eventually I'll have
some inline assembly (once I think I'm good enough at it) but just for now
it'll be straight C++.  I hope to be able to get alot done within the next
month or so, but FFTs/NTTs are fairly new to me, and I've just recently got
the whole thing straightened out in my head.

Oh, does anyone know where I can find a freely available assembler that
creates Visual C++ compatible .obj files?  The only ones I've found create
a.out files, and all the converters I've seen only are for changing .objs to
a.outs, unfortunately.  (Besides, converters aren't known for being
reliable...)  I'd appreciate a few pointers :)  I don't really feel like
going to the store and buying Turbo/Macro Assembler, since I'm not exactly
rich <g> (I'm leaving for college in the fall and I gotta save at least a
couple cents just in case... :)  )

Thanks!

Pete
<[e-mail]>



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: (fractdev) RE: (fractint) Ultra Fractal 2.0 beta 5
Date: 23 Jul 1998 10:18:53 -0500

Peter,

 - Oh, does anyone know where I can find a freely available assembler that
 - creates Visual C++ compatible .obj files?

Someone posted on comp.lang.asm.x86 not too long ago that you can download
MASM 6.11d as part of a driver kit or something, for free, legitimately,
from Microsoft's web site.  I don't have the URL handy, but if you search
Deja News (www.dejanews.com) for it you should find it.

Damien M. Jones   \\
[e-mail]  \\  http://www.icd.com/tsd/ (temporary sanity designs)
                    \\  http://www.fractalus.com/ (fractals are my hobby)

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: "Peter Gavin" <[e-mail]>
Subject: RE: (fractdev) RE: (fractint) Ultra Fractal 2.0 beta 5
Date: 23 Jul 1998 11:41:28 -0400

Cool, thanks.

Pete
<[e-mail]>

//  -----Original Message-----
//  From: [e-mail]
//  [mailto:[e-mail]]On Behalf Of Damien M. Jones
//  Sent: Thursday, July 23, 1998 11:19 AM
//  To: [e-mail]
//  Subject: Re: (fractdev) RE: (fractint) Ultra Fractal 2.0 beta 5
//
//
//  Peter,
//
//   - Oh, does anyone know where I can find a freely available
//  assembler that
//   - creates Visual C++ compatible .obj files?
//
//  Someone posted on comp.lang.asm.x86 not too long ago that you
//  can download
//  MASM 6.11d as part of a driver kit or something, for free, legitimately,
//  from Microsoft's web site.  I don't have the URL handy, but if
//  you search
//  Deja News (www.dejanews.com) for it you should find it.
//
//  Damien M. Jones   \\
//  [e-mail]  \\  http://www.icd.com/tsd/ (temporary
//  sanity designs)
//                      \\  http://www.fractalus.com/ (fractals are
//  my hobby)
//
//  --------------------------------------------------------------
//  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: "Peter Gavin" <[e-mail]>
Subject: RE: (fractdev) RE: (fractint) Ultra Fractal 2.0 beta 5
Date: 23 Jul 1998 12:03:14 -0400

Great!  You can get MASM 6.0 as part of the Win98 DDK at
http://www.microsoft.com/hwdev/ddk/ddk98.htm
Its 19 megs but that includes some docs and stuff.  You don't _need_ win98,
just win95 or NT4, but it does require Visual C++ 4 or better.



Pete
<[e-mail]>

//  -----Original Message-----
//  From: [e-mail]
//  [mailto:[e-mail]]On Behalf Of Damien M. Jones
//  Sent: Thursday, July 23, 1998 11:19 AM
//  To: [e-mail]
//  Subject: Re: (fractdev) RE: (fractint) Ultra Fractal 2.0 beta 5
//
//
//  Peter,
//
//   - Oh, does anyone know where I can find a freely available
//  assembler that
//   - creates Visual C++ compatible .obj files?
//
//  Someone posted on comp.lang.asm.x86 not too long ago that you
//  can download
//  MASM 6.11d as part of a driver kit or something, for free, legitimately,
//  from Microsoft's web site.  I don't have the URL handy, but if
//  you search
//  Deja News (www.dejanews.com) for it you should find it.
//
//  Damien M. Jones   \\
//  [e-mail]  \\  http://www.icd.com/tsd/ (temporary
//  sanity designs)
//                      \\  http://www.fractalus.com/ (fractals are
//  my hobby)
//
//  --------------------------------------------------------------
//  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: "Damien M. Jones" <[e-mail]>
Subject: RE: (fractdev) RE: (fractint) Ultra Fractal 2.0 beta 5
Date: 23 Jul 1998 11:40:45 -0500

Peter,

 - Great!  You can get MASM 6.0 as part of the Win98 DDK

Yep.  It's gems like that I read the newsgroup for.  Most of it is pretty
lame, kids looking for people to help them with homework or something, but
once in a while there's something really good.

BTW, make sure you pick up the latest copy of Agner Fog's Pentium
optimization guide.  Even if you don't use it, it contains a lot of
interesting material about how things work inside the Pentium and Pentium-II.

Damien M. Jones   \\
[e-mail]  \\  http://www.icd.com/tsd/ (temporary sanity designs)
                    \\  http://www.fractalus.com/ (fractals are my hobby)

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: "Peter Gavin" <[e-mail]>
Subject: RE: (fractdev) RE: (fractint) Ultra Fractal 2.0 beta 5
Date: 23 Jul 1998 12:50:46 -0400

Kragen just recommended to me the GNU-Win32 tools from Cygnus, and I was
thinking...  any luck using these to compile the Linux version of fractint
for win32?  Has anyone tried? :)

Pete
<[e-mail]>

//  -----Original Message-----
//  From: [e-mail]
//  [mailto:[e-mail]]On Behalf Of Damien M. Jones
//  Sent: Thursday, July 23, 1998 12:41 PM
//  To: [e-mail]
//  Subject: RE: (fractdev) RE: (fractint) Ultra Fractal 2.0 beta 5
//
//
//  Peter,
//
//   - Great!  You can get MASM 6.0 as part of the Win98 DDK
//
//  Yep.  It's gems like that I read the newsgroup for.  Most of it
//  is pretty
//  lame, kids looking for people to help them with homework or
//  something, but
//  once in a while there's something really good.
//
//  BTW, make sure you pick up the latest copy of Agner Fog's Pentium
//  optimization guide.  Even if you don't use it, it contains a lot of
//  interesting material about how things work inside the Pentium
//  and Pentium-II.
//
//  Damien M. Jones   \\
//  [e-mail]  \\  http://www.icd.com/tsd/ (temporary
//  sanity designs)
//                      \\  http://www.fractalus.com/ (fractals are
//  my hobby)
//
//  --------------------------------------------------------------
//  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: [e-mail] (Kragen)
Subject: (fractdev) Fractint in dosemu
Date: 23 Jul 1998 12:59:44 -0400 (EDT)

I'm running Fractint 19.5 in dosemu 0.66.7 (under Caldera OpenDOS 7.01)
in XFree86 3.3.2 on Linux 2.0.30 on an AMD 5x86-133.  I'm liking it
very much!  Now I can do fractals in a window while I'm analyzing data,
reading my email, or whatever.

But there are some problems with it.  I think most of these are
problems with dosemu; I wonder if some of them could be worked around
in Fractint without any difficulty?

I hope this message is interesting to some of the fractint developers.
I'm sure it will be to the dosemu developers.

- When I select 800x600x256 from the DEL screen, it works properly, and
dosemu says: vesa_set_SVGA_mode(): mode=0x5e.  But if I TAB to the info
screen, then hit TAB again to get back to the picture, it doesn't print
that message, and it also doesn't redisplay the picture.  It looks like
it's displaying the first few pixels of the image as text, though -- my
screen fills with multicolored blocks of characters.  (Screenshots upon
request.)

Switching back and forth works fine at 320x200x256.

I have the same problem with some other VESA modes, too.  Maybe it's
universal to all of them.

The blocks of characters also appear momentarily switching from the
image to a text screen -- for example, when I hit 'T'.

- Sometimes Fractint seems to hang.  If I hit Esc a couple of times, I
can get to a menu and a sensible state, but if I'm in a VESA mode, of
course I've lost my fractal.  I suspect Fractint is trying to do
something, but I'm not seeing what it's doing due to some flaw in
dosemu.

- Sometimes, in VESA mode, solid-guessing doesn't fill in the solids on
some lines.  I'll see a fractal with big black cracks across it, with
dots scattered in the cracks.  The cracks go away on later iterations;
I mention this because it might be of help with other problems.

- color-cycling doesn't work.  plasma displays with the default VGA
colormap, and thus looks terrible.  This is obviously not fractint's
fault -- right?

- Sometimes, when zooming in (even in 320x200x256), my zoom box gets
filled with black.  This makes it hard to zoom.  Also, when TABbing
back and forth on some bifurcation fractals, what was previously drawn
gets erased -- it's displayed for a brief moment when returning to the
picture, then the whole screen becomes black, then the fractal
continues drawing from the right edge of what was previously drawn.

This happens, for example, during the first zoom in basic.key.

- Drawing an "icons" fractal after a "mandel" fractal at 320x200x256,
the top and bottom of the "mandel" fractal (above and below where the
"icons" fractal appears) remain on the screen.  The same is true of
running "diffusion" or "lsystem", or running "ant" after "plasma", or
even "ant" after "ant".  I suspect it's a general problem erasing the
screen.

(In fact, when zooming in on a "mandel" in solid-guessing mode, the old
image is progressively erased as the new image is drawn.  I initially
thought this was a clever new feature.)

- When I try the 1280x1024x256 VESA mode, Fractint says it's not
supported with my adapter.

- lsystem doesn't seem to work well at all in the VESA modes.  It looks
like fractint first switches to the VESA video mode, then switches back
to text-mode to display the L-System thinking message, then fails the
same way coming back from TAB or F1 fails.

OTOH, it works pretty well with the MCGA mode.  *sigh*

- When saving an image (in MCGA mode), the text message that normally
appears in the upper-left-hand corner of the screen when it finishes is
only one pixel high, and the rest of that scan line turns to black.

- Fractint consistently crashed dosemu during the demo:
ERROR: unexpected CPU exception 0x06 errorcode: 0x00000000 while in vm86 (DOS)

Program=sigsegv.c, Line=230
EIP: 0000:0000008d ESP: 4363:0000f032  VFLAGS(b): 00000 00110000 10000110
EAX: 00004201 EBX: 000099fe ECX: 000007cc EDX: 00000000 VFLAGS(h): 00003086
ESI: 00000012 EDI: 0000f52c EBP: 0000f44a DS: 4363 ES: 4363 FS: 0000 GS: 0000
FLAGS: PF SF IF RF VM  IOPL: 3
STACK: 00 00 f8 bf 00 00 00 00 00 00 -> 00 42 91 0c f1 0c 46 32 c2 15
OPS  : 01 39 10 00 01 26 01 d2 0a ca -> 26 8e 37 1c 01 d2 0a 41 10 00

Just after successfully loading new19001.gif.

(in new19.key.)

To reproduce: run demo.bat, select option 2, and wait.

I think this happens every time you try to browse images (with "L").  I
seem to remember that this used to crash the computer in real MS-DOS as
well.  Running fractint, hitting F3, and typing "L" makes it crash, too
(in dosemu, that is).  Sometimes it crashes with different register
values.


Anyway, I'm looking forward to looking at all the fractint list
formulas and pars with my nifty new toy.  :)

Kragen




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: "Peter Gavin" <[e-mail]>
Subject: RE: (fractdev) Fractint in dosemu
Date: 23 Jul 1998 13:15:34 -0400

I suspect the video problems have alot to do with the fact that your running
it in a window.  Correct me if I'm wrong, but I think VESA uses direct
hardware access to display graphics, which makes it fast, but windows 95
doesn't like that, so I figure X/Dosemu doesn't either.  In fact, if I
switch out of fractint while it's in a VESA graphics mode, I can't go back
and windows tells me it must close the program.  I _can_ use 320x200 or
320x240 but only the former works in a window.  All other graphics modes are
unusable.  I suppose part of the problem is caused by fractint trying to
write straight to the screen, interfering with X in some way.  And dosemu
looks like it's still in beta, from its version number <g>.


Pete
<[e-mail]>

//  -----Original Message-----
//  From: [e-mail]
//  [mailto:[e-mail]]On Behalf Of Kragen
//  Sent: Thursday, July 23, 1998 1:00 PM
//  To: [e-mail]
//  Cc: [e-mail]
//  Subject: (fractdev) Fractint in dosemu
//
//
//  I'm running Fractint 19.5 in dosemu 0.66.7 (under Caldera OpenDOS 7.01)
//  in XFree86 3.3.2 on Linux 2.0.30 on an AMD 5x86-133.  I'm liking it
//  very much!  Now I can do fractals in a window while I'm analyzing data,
//  reading my email, or whatever.
//
//  But there are some problems with it.  I think most of these are
//  problems with dosemu; I wonder if some of them could be worked around
//  in Fractint without any difficulty?
//
//  I hope this message is interesting to some of the fractint developers.
//  I'm sure it will be to the dosemu developers.
//
//  - When I select 800x600x256 from the DEL screen, it works properly, and
//  dosemu says: vesa_set_SVGA_mode(): mode=0x5e.  But if I TAB to the info
//  screen, then hit TAB again to get back to the picture, it doesn't print
//  that message, and it also doesn't redisplay the picture.  It looks like
//  it's displaying the first few pixels of the image as text, though -- my
//  screen fills with multicolored blocks of characters.  (Screenshots upon
//  request.)
//
//  Switching back and forth works fine at 320x200x256.
//
//  I have the same problem with some other VESA modes, too.  Maybe it's
//  universal to all of them.
//
//  The blocks of characters also appear momentarily switching from the
//  image to a text screen -- for example, when I hit 'T'.
//
//  - Sometimes Fractint seems to hang.  If I hit Esc a couple of times, I
//  can get to a menu and a sensible state, but if I'm in a VESA mode, of
//  course I've lost my fractal.  I suspect Fractint is trying to do
//  something, but I'm not seeing what it's doing due to some flaw in
//  dosemu.
//
//  - Sometimes, in VESA mode, solid-guessing doesn't fill in the solids on
//  some lines.  I'll see a fractal with big black cracks across it, with
//  dots scattered in the cracks.  The cracks go away on later iterations;
//  I mention this because it might be of help with other problems.
//
//  - color-cycling doesn't work.  plasma displays with the default VGA
//  colormap, and thus looks terrible.  This is obviously not fractint's
//  fault -- right?
//
//  - Sometimes, when zooming in (even in 320x200x256), my zoom box gets
//  filled with black.  This makes it hard to zoom.  Also, when TABbing
//  back and forth on some bifurcation fractals, what was previously drawn
//  gets erased -- it's displayed for a brief moment when returning to the
//  picture, then the whole screen becomes black, then the fractal
//  continues drawing from the right edge of what was previously drawn.
//
//  This happens, for example, during the first zoom in basic.key.
//
//  - Drawing an "icons" fractal after a "mandel" fractal at 320x200x256,
//  the top and bottom of the "mandel" fractal (above and below where the
//  "icons" fractal appears) remain on the screen.  The same is true of
//  running "diffusion" or "lsystem", or running "ant" after "plasma", or
//  even "ant" after "ant".  I suspect it's a general problem erasing the
//  screen.
//
//  (In fact, when zooming in on a "mandel" in solid-guessing mode, the old
//  image is progressively erased as the new image is drawn.  I initially
//  thought this was a clever new feature.)
//
//  - When I try the 1280x1024x256 VESA mode, Fractint says it's not
//  supported with my adapter.
//
//  - lsystem doesn't seem to work well at all in the VESA modes.  It looks
//  like fractint first switches to the VESA video mode, then switches back
//  to text-mode to display the L-System thinking message, then fails the
//  same way coming back from TAB or F1 fails.
//
//  OTOH, it works pretty well with the MCGA mode.  *sigh*
//
//  - When saving an image (in MCGA mode), the text message that normally
//  appears in the upper-left-hand corner of the screen when it finishes is
//  only one pixel high, and the rest of that scan line turns to black.
//
//  - Fractint consistently crashed dosemu during the demo:
//  ERROR: unexpected CPU exception 0x06 errorcode: 0x00000000
//  while in vm86 (DOS)
//
//  Program=sigsegv.c, Line=230
//  EIP: 0000:0000008d ESP: 4363:0000f032  VFLAGS(b): 00000
//  00110000 10000110
//  EAX: 00004201 EBX: 000099fe ECX: 000007cc EDX: 00000000
//  VFLAGS(h): 00003086
//  ESI: 00000012 EDI: 0000f52c EBP: 0000f44a DS: 4363 ES: 4363 FS:
//  0000 GS: 0000
//  FLAGS: PF SF IF RF VM  IOPL: 3
//  STACK: 00 00 f8 bf 00 00 00 00 00 00 -> 00 42 91 0c f1 0c 46 32 c2 15
//  OPS  : 01 39 10 00 01 26 01 d2 0a ca -> 26 8e 37 1c 01 d2 0a 41 10 00
//
//  Just after successfully loading new19001.gif.
//
//  (in new19.key.)
//
//  To reproduce: run demo.bat, select option 2, and wait.
//
//  I think this happens every time you try to browse images (with "L").  I
//  seem to remember that this used to crash the computer in real MS-DOS as
//  well.  Running fractint, hitting F3, and typing "L" makes it crash, too
//  (in dosemu, that is).  Sometimes it crashes with different register
//  values.
//
//
//  Anyway, I'm looking forward to looking at all the fractint list
//  formulas and pars with my nifty new toy.  :)
//
//  Kragen
//
//
//
//
//  --------------------------------------------------------------
//  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: [e-mail] (Kragen)
Subject: RE: (fractdev) Fractint in dosemu
Date: 23 Jul 1998 13:42:37 -0400 (EDT)

On Thu, 23 Jul 1998, Peter Gavin wrote:
> I suspect the video problems have alot to do with the fact that your running
> it in a window.  Correct me if I'm wrong, but I think VESA uses direct
> hardware access to display graphics,

I'm pretty sure that it doesn't -- that the VESA drivers give you a
function-call interface, not a bunch of ports and RAM to write to.  But
I don't really know.  At any rate, it's running in a window, so
whatever direct hardware access it's using is emulated anyway.

> which makes it fast, but windows 95
> doesn't like that, so I figure X/Dosemu doesn't either.

I have a cute little game (Nuclear Insanity) which successfully
color-cycles.  I'm pretty sure Nuke does direct hardware access; I'm
contacting the developers to see if they'll show me the asm source for
the color-cycling stuff.

>  In fact, if I
> switch out of fractint while it's in a VESA graphics mode, I can't go back
> and windows tells me it must close the program.  I _can_ use 320x200 or
> 320x240 but only the former works in a window.  All other graphics modes are
> unusable.

800x600 and 1024x768 all work in a window for me.  The only time I've
had dosemu crash is when I do 'L'.

> I suppose part of the problem is caused by fractint trying to
> write straight to the screen, interfering with X in some way.  And dosemu
> looks like it's still in beta, from its version number <g>.

DOSEMU has been in beta for years and years.  Maybe one day it will escape.

Fractint can try to write straight to the screen, but it won't succeed
in a million years...

(btw: I haven't set up the VGA stuff with dosemu, so I *have* to run it
in a window.)

Kragen


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: (fractdev) Fractint in dosemu
Date: 23 Jul 1998 12:57:33 -0500

Peter,

 - In fact, if I switch out of fractint while it's in a VESA graphics mode,
 - I can't go back and windows tells me it must close the program.

This depends very much on your VESA drivers (usually in the BIOS of your
video card) and even the particular mode used.  My Matrox Millennium lets
me switch away and back in 1600x1200 mode (my preferred mode) or 800x600,
but if I use 1024x768 or 640x480, switching back will corrupt the image or,
sometimes, provide no display at all.

 - I _can_ use 320x200 or 320x240 but only the former works in a window.

That's because 320x200 is a standard BIOS-supported video mode, but 320x240
is a "Mode X" display, in which some very clever reprogramming of the VGA
registers allows a lot more flexibility.  Mode X displays allow page
flipping, hardware panning, and displays of 320x or 360x, x200, x240, x400,
and x480.  When I was looking into game programming on the PC we found
information on this.  It's more awkward to work with the screen this way,
due to some of those clever programming tricks, but it's usually worth the
hassle.

Frankly, I'm surprised (pleasantly so) that DOSemu works so well.

Damien M. Jones   \\
[e-mail]  \\  http://www.icd.com/tsd/ (temporary sanity designs)
                    \\  http://www.fractalus.com/ (fractals are my hobby)

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: (fractdev) Fractint in dosemu
Date: 23 Jul 1998 13:03:15 -0500

Kragen,

 - I'm pretty sure that it doesn't -- that the VESA drivers give you a
 - function-call interface, not a bunch of ports and RAM to write to.

You get function calls to switch memory banks and to set up modes.  You
still write to RAM directly.

 - I'm contacting the developers to see if they'll show me the asm source
 - for the color-cycling stuff.

Palette registers are basically in the same location on all VGA cards; this
isn't a VESA issue.  If you want assembly to set VGA palette registers, I
can post it here.  (This is also useful for fading text.)

Damien M. Jones   \\
[e-mail]  \\  http://www.icd.com/tsd/ (temporary sanity designs)
                    \\  http://www.fractalus.com/ (fractals are my hobby)

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: "Peter Gavin" <[e-mail]>
Subject: RE: (fractdev) Fractint in dosemu
Date: 23 Jul 1998 14:12:12 -0400

//  On Thu, 23 Jul 1998, Peter Gavin wrote:
//  > I suspect the video problems have alot to do with the fact
//  that your running
//  > it in a window.  Correct me if I'm wrong, but I think VESA uses direct
//  > hardware access to display graphics,
//
//  I'm pretty sure that it doesn't -- that the VESA drivers give you a
//  function-call interface, not a bunch of ports and RAM to write to.  But
//  I don't really know.  At any rate, it's running in a window, so
//  whatever direct hardware access it's using is emulated anyway.
//

Well, the vesa drivers could do the direct hardware access themselves, and
you tell them what you want done through a function.  I don't even know if
VESA uses that anyway, though...  Could someone else explain this?


//  > which makes it fast, but windows 95
//  > doesn't like that, so I figure X/Dosemu doesn't either.
//
//  I have a cute little game (Nuclear Insanity) which successfully
//  color-cycles.  I'm pretty sure Nuke does direct hardware access; I'm
//  contacting the developers to see if they'll show me the asm source for
//  the color-cycling stuff.
//

Does that nuke game do color cycling in a window?  Because if X is running
at 16- or 24-bit color, that could explain why color cycling doesn't work.
Cycling is just done by rotating the palette, of course, but the higher
depth color modes don't use palettes, so there would have to be some kind of
code in dosemu to rotate them manually pixel by pixel, which would really be
a pain.


//  >  In fact, if I
//  > switch out of fractint while it's in a VESA graphics mode, I
//  can't go back
//  > and windows tells me it must close the program.  I _can_ use
//  320x200 or
//  > 320x240 but only the former works in a window.  All other
//  graphics modes are
//  > unusable.
//
//  800x600 and 1024x768 all work in a window for me.  The only time I've
//  had dosemu crash is when I do 'L'.
//

I don't think I was really clear about this, and I suppose you understood
anyways, but... I should have said that in full screen mode, I can use
anything up to 1024x768 (my monitors max res) but if I switch out which in
graphics mode, Fractint dies.  If I'm in 320x200 or 320x240, it doesn't die,
and fractint will _run_ while windowed at 320x200, because windows likes
that mode, since it's extremely simple.  320x240 will show up ok in a
window, but the program halts until I go back to full screen.


//  DOSEMU has been in beta for years and years.  Maybe one day it
//  will escape.
//
//  Fractint can try to write straight to the screen, but it won't succeed
//  in a million years...
//
//  (btw: I haven't set up the VGA stuff with dosemu, so I *have* to run it
//  in a window.)

That's okay, alot of software doesn't ever get out of beta...  Like ICQ!
Kinda wierd, when you think about it.  7 million people are running a beta
software and probably dont even realize it! :)  If it's stable enough for 7
million people it should be out of beta by now, don't you think?



Pete
<[e-mail]>


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] (Kragen)
Subject: RE: (fractdev) Fractint in dosemuZ
Date: 23 Jul 1998 14:18:57 -0400 (EDT)

On Thu, 23 Jul 1998, Damien M. Jones wrote:
> Frankly, I'm surprised (pleasantly so) that DOSemu works so well.

Me too.  My wife and I are going to analyze some data, and the software
she wants to use runs on DOS, so I thought I'd try it.

Kragen


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] (Kragen)
Subject: RE: (fractdev) Fractint in dosemu
Date: 23 Jul 1998 14:19:54 -0400 (EDT)

On Thu, 23 Jul 1998, Damien M. Jones wrote:
>  - I'm contacting the developers to see if they'll show me the asm source
>  - for the color-cycling stuff.
>
> Palette registers are basically in the same location on all VGA cards; this
> isn't a VESA issue.  If you want assembly to set VGA palette registers, I
> can post it here.  (This is also useful for fading text.)

That's what I thought, too.  But Nuclear Insanity color-cycles, and
Fractint doesn't.

Kragen


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] (Kragen)
Subject: RE: (fractdev) Fractint in dosemu
Date: 23 Jul 1998 14:34:56 -0400 (EDT)

On Thu, 23 Jul 1998, Peter Gavin wrote:
> //  I have a cute little game (Nuclear Insanity) which successfully
> //  color-cycles.  I'm pretty sure Nuke does direct hardware access; I'm
> //  contacting the developers to see if they'll show me the asm source for
> //  the color-cycling stuff.
> //
>
> Does that nuke game do color cycling in a window?

Yes.  I'm running X at 8bpp.

> //  DOSEMU has been in beta for years and years.  Maybe one day it
> //  will escape.
> //
> //  Fractint can try to write straight to the screen, but it won't succeed
> //  in a million years...
> //
> //  (btw: I haven't set up the VGA stuff with dosemu, so I *have* to run it
> //  in a window.)
>
> That's okay, alot of software doesn't ever get out of beta...  Like ICQ!

There are a couple of Linux ICQ clients out there that promise to
escape beta someday soon.

> Kinda wierd, when you think about it.  7 million people are running a beta
> software and probably dont even realize it! :)  If it's stable enough for 7
> million people it should be out of beta by now, don't you think?

Nah.

Kragen


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: (fractdev) Fractint in dosemu
Date: 23 Jul 1998 13:36:43 -0500

Peter,

 - Kinda wierd, when you think about it.  7 million people are running a
 - beta software and probably dont even realize it! :)

Closer to 14 million, actually.

 - If it's stable enough for 7 million people it should be out of beta
 - by now, don't you think?

You must be one of those people who has never had a problem.  I
wholeheartedly agree with the beta status of ICQ (even though in part it
was a marketing position).  ICQ crashes quite frequently, more often than
most of the software I use (more often than Windows, even).  It also
exhibits odd behavior, such as getting confused about who is online and who
is not.  Part of the problem is that ICQ doesn't seem to have a TCP
fallback protocol for when it can't reliably communicate using UDP.  When a
net brown-out occurs, and half the packets go AWOL, ICQ just quits working,
basically, sending duplicate messages, not getting confirmation that
messages were received, etc.

On the whole, though, ICQ performs pretty well.  The ICQ servers have
actually improved over time; they still have odd quirks but for the most
part they work OK.  I'm happy with the package, although I'm definitely NOT
happy AOL ended up buying it.

Damien M. Jones   \\
[e-mail]  \\  http://www.icd.com/tsd/ (temporary sanity designs)
                    \\  http://www.fractalus.com/ (fractals are my hobby)

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: (fractdev) Fractint in dosemu
Date: 23 Jul 1998 13:38:00 -0500

Kragen,

 - That's what I thought, too.  But Nuclear Insanity color-cycles, and
 - Fractint doesn't.

It's possible that Nuclear Insanity color-cycles by actually altering pixel
data, rather than changing palette registers.  For a 256-color mode this is
pretty trivial, even to set up multiple cycle patterns within a single
palette.

Damien M. Jones   \\
[e-mail]  \\  http://www.icd.com/tsd/ (temporary sanity designs)
                    \\  http://www.fractalus.com/ (fractals are my hobby)

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] (Kragen)
Subject: RE: (fractdev) Fractint in dosemu
Date: 23 Jul 1998 14:48:25 -0400 (EDT)

On Thu, 23 Jul 1998, Damien M. Jones wrote:
> Kragen,
>
>  - That's what I thought, too.  But Nuclear Insanity color-cycles, and
>  - Fractint doesn't.
>
> It's possible that Nuclear Insanity color-cycles by actually altering pixel
> data,

No, because the background of my screen (and everything else) changes
color while my mouse is in the Nuclear Insanity window.

> rather than changing palette registers.  For a 256-color mode this is
> pretty trivial, even to set up multiple cycle patterns within a single
> palette.

It might be feasible, but it wouldn't be trivial.

Kragen


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: "Peter Gavin" <[e-mail]>
Subject: RE: (fractdev) Fractint in dosemu
Date: 23 Jul 1998 15:05:02 -0400

//  On the whole, though, ICQ performs pretty well.  The ICQ servers have
//  actually improved over time; they still have odd quirks but for the most
//  part they work OK.  I'm happy with the package, although I'm
//  definitely NOT
//  happy AOL ended up buying it.

Amen to that!  Really, though, they already have Instant Messanger, why ICQ,
too?  I don't even know how Mirabilis (err... AOL, now) makes any money from
it, they don't sell advertising (much), and ICQ is free.


Pete
<[e-mail]>



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: "Dean-Christian Strik" <[e-mail]>
Subject: Re: (fractdev) Fractint in dosemu
Date: 23 Jul 1998 21:17:12 +0200

:Mode X displays allow page
:flipping, hardware panning, and displays of 320x or 360x, x200, x240, x400,
:and x480.  When I was looking into game programming on the PC we found
:information on this.  It's more awkward to work with the screen this way,
:due to some of those clever programming tricks, but it's usually worth the
:hassle.


It can have funny results. Today I assembled a friend's computer with an AMD
K6-233 and a Stealth-II 4MB PCI card (nice card BTW - overclocking it by 20%
gave us an average framerate of about 100 in Unreal under Win98 at
640x480x2^16). When we ran Quake I under windows 98 to test the speed of the
video card using any of the Mode X display resolutions, the framerate dropped
to about 7 - for about 56 at 1024x768 (during this test I did no
overclocking).

But this maybe an exception (I hope)...

Christian




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: "Dean-Christian Strik" <[e-mail]>
Subject: Re: (fractdev) Fractint in dosemu
Date: 23 Jul 1998 22:03:27 +0200

: - If it's stable enough for 7 million people it should be out of beta
: - by now, don't you think?
:
:You must be one of those people who has never had a problem.  I
:wholeheartedly agree with the beta status of ICQ (even though in part it
:was a marketing position).  ICQ crashes quite frequently, more often than
:most of the software I use (more often than Windows, even).  It also
:exhibits odd behavior, such as getting confused about who is online and who
:is not.  Part of the problem is that ICQ doesn't seem to have a TCP
:fallback protocol for when it can't reliably communicate using UDP.  When a
:net brown-out occurs, and half the packets go AWOL, ICQ just quits working,
:basically, sending duplicate messages, not getting confirmation that
:messages were received, etc.


I experience some errors, too. Especially concerning messages I never
received. I still have four of you ICQing fractinters in the 'Waiting
authorization from' area of my ICQ window. Unless they/you (Gedeon Peteri,
John Wilson, John Misterio and Pete Gavin) haven't used ICQ for about two
weeks, I never got an acception confirmation (or denial). I sorted this out
with a friend who also remained in the Auth. area - he sure has sent me one.
There is no point in again giving permission (or not) (if possible anyway).
No, I agree that ICQ can stay Beta for a while...

:On the whole, though, ICQ performs pretty well.  The ICQ servers have
:actually improved over time; they still have odd quirks but for the most
:part they work OK.  I'm happy with the package, although I'm definitely NOT
:happy AOL ended up buying it.


Ah.. Damien - didn't know you were on ICQ. Can I have your # for my list
(http://come.to/fractinticq - sorry 'bout the popup windows - i'll have these
removed soon). And what about Mr. Sitarek? :-)

Christian





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: "Dean-Christian Strik" <[e-mail]>
Subject: Re: (fractdev) Fractint in dosemu
Date: 23 Jul 1998 22:30:22 +0200

:Ah.. Damien - didn't know you were on ICQ. Can I have your # for my list
:(http://come.to/fractinticq - sorry 'bout the popup windows - i'll have these
:removed soon). And what about Mr. Sitarek? :-)


Typo, sorry - of course mean Mr. Sitaker.


Christian




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] (Kragen)
Subject: Re: (fractdev) Fractint in dosemu
Date: 23 Jul 1998 16:37:48 -0400 (EDT)

On Thu, 23 Jul 1998, Dean-Christian Strik wrote:
> :Ah.. Damien - didn't know you were on ICQ. Can I have your # for my list
> :(http://come.to/fractinticq - sorry 'bout the popup windows - i'll have these
> :removed soon). And what about Mr. Sitarek? :-)
>
> Typo, sorry - of course mean Mr. Sitaker.

I'll install Licq or something one of these days.  Then I'll get an ICQ
number.

Kragen


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: (fractdev) Fractint in dosemu
Date: 23 Jul 1998 15:59:23 -0500

Christian,

 - Ah.. Damien - didn't know you were on ICQ. Can I have your # for my
 - list

Well, actually, I'd prefer the number not be given out.  Despite numerous
opportunities to post my number for all to see, I have not done so.  I use
ICQ at work--and work is a small ISP, so I'm *always* online.  It is easy
enough to explain this to a few friends who understand that work means
work, but to explain it to fifty new people clamoring to add me to their
contact list...?  (Yes, the privacy settings I use are pretty strict.)
Thank you, but I will pass. :-)  Anyone who wishes to find me through ICQ
can search on my e-mail address; it's listed.  But I don't generally
broadcast the fact that I use ICQ precisely because I'd prefer most people
don't know.  So I guess I just blew it, announcing it here, huh? :-)

Damien M. Jones   \\
[e-mail]  \\  http://www.icd.com/tsd/ (temporary sanity designs)
                    \\  http://www.fractalus.com/ (fractals are my hobby)

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: (fractdev) Fractint in dosemu
Date: 23 Jul 1998 18:04:24 -0500

Damien M. Jones wrote:
>
> Anyone who wishes to find me through ICQ can search on
> my e-mail address; it's listed.

  819593   Damien           ??
 1945253   damien           [e-mail]
 3620685   ?                [e-mail]
 5985594   Damo             [e-mail]
11634164   Rooboy           [e-mail]
12910213   Thunder In Eye   [e-mail]
14525812   bones            [e-mail]

But nothing for  <[e-mail]>  or  <[e-mail]>

P.N.L.
     Why do most folks hate cynics so much?
       Because we're almost always right.
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: "Damien M. Jones" <[e-mail]>
Subject: Re: (fractdev) Fractint in dosemu
Date: 23 Jul 1998 19:01:37 -0500

Paul,

 - But nothing for  <[e-mail]>  or  <[e-mail]>

Oops, you're right, I removed my e-mail address from my ICQ info.  Heh, I
must've been a bit paranoid at the time.

My last name is in there, though.

Damien M. Jones   \\
[e-mail]  \\  http://www.icd.com/tsd/ (temporary sanity designs)
                    \\  http://www.fractalus.com/ (fractals are my hobby)

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: (fractdev) Fractint in dosemu
Date: 23 Jul 1998 21:29:59 -0400

Kragen,

>> - When I select 800x600x256 from the DEL screen, it works properly, an=
d
dosemu says: vesa_set_SVGA_mode(): mode=3D0x5e.  But if I TAB to the info=

screen, then hit TAB again to get back to the picture, it doesn't print
that message, and it also doesn't redisplay the picture. <<

Have you tried using the textsafe=3Dsave option?

>> - Sometimes, in VESA mode, solid-guessing doesn't fill in the solids o=
n
some lines.  I'll see a fractal with big black cracks across it, with dot=
s
scattered in the cracks.  The cracks go away on later iterations; I menti=
on
this because it might be of help with other problems. <<

What happens with passes=3D1?  Could this be a periodicity problem?  Try
using the command periodicity=3D0.

>> I think this happens every time you try to browse images (with "L").  =
I
seem to remember that this used to crash the computer in real MS-DOS as
well.  Running fractint, hitting F3, and typing "L" makes it crash, too (=
in
dosemu, that is). <<

This may be fixed in the 19.6 developer's version.  Would you please run
Fractint with the debug=3D10000 command line option, and tell me what it
reports?   Also, it would be helpful to know how much memory is free on
your system (below 640K, expanded, extended).  Also, any other pertinent
information which may be useful. For example, a non-Intel CPU.

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] (Kragen)
Subject: Re: (fractdev) Fractint in dosemu
Date: 23 Jul 1998 23:43:19 -0400 (EDT)

On Thu, 23 Jul 1998, Jonathan Osuch wrote:
> Kragen,
> >> - When I select 800x600x256 from the DEL screen, it works properly, and
> dosemu says: vesa_set_SVGA_mode(): mode=0x5e.  But if I TAB to the info
> screen, then hit TAB again to get back to the picture, it doesn't print
> that message, and it also doesn't redisplay the picture. <<
>
> Have you tried using the textsafe=save option?

You know, I should have tried that right away, but I didn't think it
would help any -- I guess I misunderstood the problem it was supposed
to solve.  (I haven't read the textsafe documentation since probably
1994. :).  It works!  Thanks!

> >> - Sometimes, in VESA mode, solid-guessing doesn't fill in the solids on
> some lines.  I'll see a fractal with big black cracks across it, with dots
> scattered in the cracks.  The cracks go away on later iterations; I mention
> this because it might be of help with other problems. <<
>
> What happens with passes=1?  Could this be a periodicity problem?  Try
> using the command periodicity=0.

I'd think that, if it were a periodicity problem, the cracks would not
go away on later passes.  Indeed, fractint textsafe=save periodicity=0
continues to display the cracks.

> >> I think this happens every time you try to browse images (with "L").  I
> seem to remember that this used to crash the computer in real MS-DOS as
> well.  Running fractint, hitting F3, and typing "L" makes it crash, too (in
> dosemu, that is). <<
>
> This may be fixed in the 19.6 developer's version.

I'll probably grab 19.6 one of these days soon.

> Would you please run
> Fractint with the debug=10000 command line option, and tell me what it
> reports?

CPU type: 386  FPU type: 387  IIT FPU: 0  Video: 5  Video chip: 14 (vesa)

 1954 NEAR bytes free
 120928 FAR bytes free
 8612 STACK bytes free
 12680 used by HISTORY structure
 21600 video table used

> Also, it would be helpful to know how much memory is free on
> your system (below 640K, expanded, extended).  Also, any other pertinent
> information which may be useful. For example, a non-Intel CPU.

Some other pertinent information you might be interested in is at the
top of the message you were replying to.

My dosemu settings for amount of RAM available in unconventional ways follow:

# Memory.

dpmi 8192 # 1/4 of all RAM

# I don't think I'll actually use these:
xms 1024
ems 1024

The "1/4 of all RAM" comment refers to the fact that I actually have 32
megs in my machine, physically.

Thanks very much for your interest!  (Particularly since most of these
problems are probably in dosemu or OpenDOS, not Fractint.)

Kragen



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: (fractdev) Fractint in dosemu
Date: 24 Jul 1998 21:32:56 -0400

Kragen,

>> I'd think that, if it were a periodicity problem, the cracks would not=

go away on later passes.  Indeed, fractint textsafe=3Dsave periodicity=3D=
0
continues to display the cracks. <<

Ah, in that case, I'm clueless <G>.  Does this happen with the default
fractal types, or is it a specific zoom?  And, does this particular image=

generate okay on other machines?

>> I'll probably grab 19.6 one of these days soon. <<

I wouldn't count on 19.6 fixing the problem with the browser locking thin=
gs
up.  Although, using the textsafe=3Dsave option should at least let you s=
ee a
message displayed.  My experience has been that with Fractint using
expanded memory is more reliable than using extended memory.  The browser=

needs large amounts of memory, although you have enough far memory
available that it should work.

The developer's version (based on 19.6) has reworked the memory handling
routines such that calls for allocating memory for data structures should=

not fail.  At least, that's my hope.  The theory is that if one type of
memory is requested, but enough isn't available, the next type of memory
will be allocated (or attempted to be allocated).  The sequence is far,
expanded, extended, and then disk.  If you run out of disk space, you've
got big problems.  But, that isn't really a Fractint problem <G>.

I'd recommend using 4 - 8 MB of expanded memory for running Fractint.  Th=
is
will keep the browser happy.

>> Thanks very much for your interest!  (Particularly since most of these=

problems are probably in dosemu or OpenDOS, not Fractint.) <<

You're welcome.  You can never tell when something will flush out an
obscure bug.  Reports of problems are always welcome, especially with new=

and different environments.

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 Gilman <[e-mail]>
Subject: (fractdev) An incredibly small bug
Date: 27 Jul 1998 16:57:11 -0700

While coding up some Mac dialogs to fake the Basic Options screen, I came
across this:

Sound (no, yes, x, y, z)

But the arrow keys actually perform like this:

<->  yes  <->  no  <->  x  <->  y  <->  z  <->


More hard-hitting bug reports coming soon....   ;-P

-= tim gilman
[e-mail]
[e-mail]

"There is no truth, in which passing through awareness, does not lie.
 Yet we chase after it all the same."

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: (fractdev) An incredibly small bug
Date: 27 Jul 1998 21:19:23 -0400

Tim,

>>
Sound (no, yes, x, y, z)

But the arrow keys actually perform like this:

<->  yes  <->  no  <->  x  <->  y  <->  z  <->
<<

Got it, thanks.  I changed the prompt so it reads:

Sound (yes, no, x, y, z)

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] (Kragen)
Subject: Re: (fractdev) Fractint in dosemu
Date: 29 Jul 1998 10:50:24 -0400 (EDT)

Sorry it took so long for me to answer!

On Fri, 24 Jul 1998, Jonathan Osuch wrote:
> >> I'd think that, if it were a periodicity problem, the cracks would not
> go away on later passes.  Indeed, fractint textsafe=save periodicity=0
> continues to display the cracks. <<
>
> Ah, in that case, I'm clueless <G>.  Does this happen with the default
> fractal types, or is it a specific zoom?  And, does this particular image
> generate okay on other machines?

It seems to happen with all escape-time fractal types; I noticed it
first with mandel, zooming in on the crack between the main cardioid
and the big circle next to it.  It happens with lots of images, but not
necessarily all.

> >> I'll probably grab 19.6 one of these days soon. <<
>
> I wouldn't count on 19.6 fixing the problem with the browser locking things
> up.

You mean, when I try to do "L"?

> Although, using the textsafe=save option should at least let you see a
> message displayed.  My experience has been that with Fractint using
> expanded memory is more reliable than using extended memory.

I could turn off XMS without much trouble.  What about DPMI?  Does
Fractint support it?

> The browser
> needs large amounts of memory, although you have enough far memory
> available that it should work.

Well, it could just be a problem with dosemu's memory management --
although it *did* happen under raw DOS on the 5x86.  (I don't remember
whether it was MS-DOS or OpenDOS.)

> The developer's version (based on 19.6) has reworked the memory handling
> routines such that calls for allocating memory for data structures should
> not fail.  At least, that's my hope.  The theory is that if one type of
> memory is requested, but enough isn't available, the next type of memory
> will be allocated (or attempted to be allocated).  The sequence is far,
> expanded, extended, and then disk.  If you run out of disk space, you've
> got big problems.  But, that isn't really a Fractint problem <G>.

Heh :)

In dosemu, it's quite reasonable for me (on my 32M machine) to make 90M
available through EMS, XMS, or DPMI (provided those interfaces can
support such large amounts of memory -- can they?) and let Linux handle
swapping to disk.

> I'd recommend using 4 - 8 MB of expanded memory for running Fractint.  This
> will keep the browser happy.

OK, I'll try that.

> >> Thanks very much for your interest!  (Particularly since most of these
> problems are probably in dosemu or OpenDOS, not Fractint.) <<
>
> You're welcome.  You can never tell when something will flush out an
> obscure bug.  Reports of problems are always welcome, especially with new
> and different environments.

That's wonderful!

Kragen (who just read the Algorithmic Beauty of Plants, and is
salivating over the thought of rendering IFSes as escape-time 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: Ron Barnett <[e-mail]>
Subject: (fractdev) test
Date: 29 Jul 1998 15:43:02 -0400

I don't seem to be getting fractdev list mail. This is a test.

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: (fractdev) Fractint in dosemu
Date: 29 Jul 1998 21:34:52 -0400

Kragen,

>> It seems to happen with all escape-time fractal types; I noticed it
first with mandel, zooming in on the crack between the main cardioid
and the big circle next to it.  It happens with lots of images, but not
necessarily all. <<

Would it be a problem for you to email me a small GIF illustrating the
problem?  Is this only a problem with solid guessing?  Is the final image=

affected?  Is the crack vertical or horizontal?  What happens if you use
"symmetry=3Dno"?

>> > >> I'll probably grab 19.6 one of these days soon. <<
> =

> I wouldn't count on 19.6 fixing the problem with the browser locking
things
> up.  =


You mean, when I try to do "L"?
<<

Yes, pressing "L" invokes the browser unless you are in color cycling or
the palette editor.

>> What about DPMI?  Does Fractint support it? <<

Not at present, and I think DPMI support is way beyond my skills.

>> In dosemu, it's quite reasonable for me (on my 32M machine) to make 90=
M
available through EMS, XMS, or DPMI (provided those interfaces can
support such large amounts of memory -- can they?) and let Linux handle
swapping to disk. <<

I don't know what the limits are.

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: (fractdev) Fractint in dosemu
Date: 30 Jul 1998 21:21:43 -0400

Kragen,

>> Here's a 13K GIF.  It was generated by xv, not fractint, and it
illustrates some problems erasing the zoom box, the "problem" in which
parts of the old image are not erased until the drawing process reaches
them, and the horizontal-crack problems. <<

That's an interesting image.  What video adapter are you using?  Because
this happens with more than one program, the video adapter might be causi=
ng
the problem.  Depending on your video adapter, what happens when you forc=
e
the detection with the command line adapter=3Dxxxx?  This should prevent =
the
use of the VESA routines.

>> >  Is the final image affected?

Interestingly, no, not at all.
<<

That pixel data has to be going somewhere.  Perhaps what is happening is
that we are crossing a boundary when we are doing the pixel write.  Some =
of
the pixels get written because they are on the correct side of the
boundary.  Do the cracks always appear in the same location on the screen=
?

>> >  What happens if you use "symmetry=3Dno"?

Where do I use that? <<

Either on the command line or by pressing the <g> key.

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: "Peter Gavin" <[e-mail]>
Subject: (fractdev) C complex types
Date: 31 Jul 1998 02:03:24 -0400

This is a multi-part message in MIME format.

------=_NextPart_000_0001_01BDBC27.666E6820
Content-Type: text/plain;
    charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

I know this probably sounds really stupid, but it's the *only* thing in =
C that confuses me...  Can someone explain to me how I can interpret and =
create variable types such as:

const int * const annoying[3];

I want to make an array of constant pointers to constant chars, such as:

typedef const char /* a bunch of confusing stuff */ stupid;

stupid adjectives =3D { "idiot", "moron", "Bill Gates" }

Now, as you can see, the 3 strings are constant, but varying length, so =
I want an array of 3 pointers to characters.  The pointers are constant, =
as are the characters.  How the @#$% do I do this? <g>
(I know this is somewhat unnecessary [I suppose using char ** would =
work] but I want to prevent all modifications, since the strings are to =
be global. I also want to learn how to do this without using typedefs =
since alot of people do and I'll need to figure this out eventually.)

Peter Gavin
<[e-mail]>

// End transmission



------=_NextPart_000_0001_01BDBC27.666E6820
Content-Type: text/html;
    charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD W3 HTML//EN">
<HTML>
<HEAD>

<META content=3D"text/html; charset=3Diso-8859-1" =
http-equiv=3DContent-Type>
<META content=3D'"MSHTML 4.72.3110.7"' name=3DGENERATOR>
</HEAD>
<BODY bgColor=3D#ffffff>
<DIV><SPAN class=3D600004805-31071998><FONT color=3D#000000 =
face=3DVerdana size=3D2>I=20
know this probably sounds really stupid, but it's the *only* thing in C =
that=20
confuses me...  Can someone explain to me how I can interpret and =
create=20
variable types such as:</FONT></SPAN></DIV>
<DIV><SPAN class=3D600004805-31071998><FONT color=3D#000000 =
face=3DVerdana=20
size=3D2></FONT></SPAN> </DIV>
<DIV><SPAN class=3D600004805-31071998><FONT size=3D2><FONT =
face=3D"Courier New"><FONT=20
color=3D#000000>const int * const =
annoying[3];</FONT></FONT></FONT></SPAN><FONT=20
color=3D#000000><FONT face=3D"Courier New"><FONT =
size=3D2></FONT></FONT></FONT></DIV>
<DIV><SPAN class=3D600004805-31071998><FONT color=3D#000000 =
face=3DVerdana=20
size=3D2></FONT></SPAN> </DIV>
<DIV><SPAN class=3D600004805-31071998><FONT color=3D#000000 =
face=3DVerdana size=3D2>I=20
want to make an array of constant pointers to constant chars, such=20
as:</FONT></SPAN></DIV>
<DIV> </DIV>
<DIV><SPAN class=3D600004805-31071998><FONT size=3D2><FONT =
face=3D"Courier New"><FONT=20
color=3D#000000>typedef const char /* a bunch of confusing stuff */=20
stupid;</FONT></FONT></FONT></SPAN></DIV>
<DIV><SPAN class=3D600004805-31071998><FONT size=3D2><FONT =
face=3D"Courier New"><FONT=20
color=3D#000000></FONT></FONT></FONT></SPAN><FONT color=3D#000000><FONT=20
face=3D"Courier New"><FONT size=3D2></FONT></FONT></FONT> </DIV>
<DIV><SPAN class=3D600004805-31071998><FONT color=3D#000000 =
face=3D"Courier New"=20
size=3D2>stupid adjectives =3D { "idiot", "moron", =
"Bill=20
Gates" }</FONT></SPAN></DIV>
<DIV> </DIV>
<DIV><FONT color=3D#000000 face=3D"Courier New" size=3D2></FONT><SPAN=20
class=3D600004805-31071998><FONT color=3D#000000 face=3DVerdana =
size=3D2>Now, as you can=20
see, the 3 strings are constant, but varying length, so I want an array =
of 3=20
pointers to characters.  The pointers are constant, as are the=20
characters.  How the @#$% do I do this? =
<g></FONT></SPAN></DIV>
<DIV><SPAN class=3D600004805-31071998><FONT color=3D#000000 =
face=3DVerdana size=3D2>(I=20
know this is somewhat unnecessary [I suppose using char ** would work] =
but I=20
want to prevent all modifications, since the strings are to be global. I =
also=20
want to learn how to do this without using typedefs since alot of people =
do and=20
I'll need to figure this out eventually.)</FONT></SPAN></DIV>
<P><FONT face=3D"" size=3D2><BR>Peter Gavin<BR><<A=20
href=3D"mailto:[e-mail]">[e-mail]</A>><BR><B=
R><EM><FONT=20
face=3D"" size=3D1>// End transmission</FONT></EM></FONT></P>
<DIV> </DIV></BODY></HTML>

------=_NextPart_000_0001_01BDBC27.666E6820--


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] (Kragen)
Subject: Re: (fractdev) C complex types
Date: 31 Jul 1998 10:37:12 -0400 (EDT)

On Fri, 31 Jul 1998, Peter Gavin wrote:
> I want to make an array of constant pointers to constant chars, such as:
>
> typedef const char /* a bunch of confusing stuff */ stupid;
>
> stupid adjectives = { "idiot", "moron", "Bill Gates" }

IIRC, there's 'char const *x' (which is the same as const char *x) and
then there's 'char * const x', which means that x is const, although
what it points to may not be.

I think what you want is an array of 'char const * const'.  So you can say:
typedef char const * const stupid;
stupid adjective = "stupid";
stupid adjectives[] = { "de Raadt", "Randy Loux", NULL};

and get what you want.

But I don't use const much, so all that could be wrong.

Kragen


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] (Kragen)
Subject: Re: (fractdev) Fractint in dosemu
Date: 31 Jul 1998 12:05:59 -0400 (EDT)

On Thu, 30 Jul 1998, Jonathan Osuch wrote:
> Kragen,
>
> >> Here's a 13K GIF.  It was generated by xv, not fractint, and it
> illustrates some problems erasing the zoom box, the "problem" in which
> parts of the old image are not erased until the drawing process reaches
> them, and the horizontal-crack problems. <<
>
> That's an interesting image.  What video adapter are you using?  Because
> this happens with more than one program, the video adapter might be causing
> the problem.

I'm running dosemu in X-windows on a Cirrus Logic CLGD5446.  While it's
possible it might be the video adapter, I believe it's more likely to
be dosemu.  I haven't seen it with any other programs (in dosemu or
otherwise).

>  Depending on your video adapter, what happens when you force
> the detection with the command line adapter=xxxx?  This should prevent the
> use of the VESA routines.

I'm not sure what xxxx should be.

> >> >  Is the final image affected?
>
> Interestingly, no, not at all.
> <<
>
> That pixel data has to be going somewhere.  Perhaps what is happening is
> that we are crossing a boundary when we are doing the pixel write.  Some of
> the pixels get written because they are on the correct side of the
> boundary.  Do the cracks always appear in the same location on the screen?

Well, I'm not sure.  But they do disappear on later passes.

> >> >  What happens if you use "symmetry=no"?
>
> Where do I use that? <<
>
> Either on the command line or by pressing the <g> key.

I tried both.  It said it didn't know what I was talking about.

Kragen


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] (Kragen)
Subject: Re: (fractdev) Fractint in dosemu
Date: 31 Jul 1998 12:05:59 -0400 (EDT)

On Thu, 30 Jul 1998, Jonathan Osuch wrote:
> Kragen,
>
> >> Here's a 13K GIF.  It was generated by xv, not fractint, and it
> illustrates some problems erasing the zoom box, the "problem" in which
> parts of the old image are not erased until the drawing process reaches
> them, and the horizontal-crack problems. <<
>
> That's an interesting image.  What video adapter are you using?  Because
> this happens with more than one program, the video adapter might be causing
> the problem.

I'm running dosemu in X-windows on a Cirrus Logic CLGD5446.  While it's
possible it might be the video adapter, I believe it's more likely to
be dosemu.  I haven't seen it with any other programs (in dosemu or
otherwise).

>  Depending on your video adapter, what happens when you force
> the detection with the command line adapter=xxxx?  This should prevent the
> use of the VESA routines.

I'm not sure what xxxx should be.

> >> >  Is the final image affected?
>
> Interestingly, no, not at all.
> <<
>
> That pixel data has to be going somewhere.  Perhaps what is happening is
> that we are crossing a boundary when we are doing the pixel write.  Some of
> the pixels get written because they are on the correct side of the
> boundary.  Do the cracks always appear in the same location on the screen?

Well, I'm not sure.  But they do disappear on later passes.

> >> >  What happens if you use "symmetry=no"?
>
> Where do I use that? <<
>
> Either on the command line or by pressing the <g> key.

I tried both.  It said it didn't know what I was talking about.

Kragen


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: (fractdev) C complex types
Date: 31 Jul 1998 11:31:33 -0600


In article <000001bdbc48$ed76e060$ea83cdcf@p60>,
    "Peter Gavin" <[e-mail]>  writes:
> I want to make an array of constant pointers to constant chars, such as:
>
> typedef const char /* a bunch of confusing stuff */ stupid;
>
> stupid adjectives =3D { "idiot", "moron", "Bill Gates" }

gcc accepted this perfectly fine:

    const char *const adjectives[3] = { "one", "two", "three" };

did you try that?  This is a const character string:

    const char *constant_string = "initializer";

That says the characters themselves can't be changed, but the pointer
can be changed, i.e. "constant_string++" is valid.  This is a constant
pointer to a constant character string:

    const char *const constant_ptr_to_constant_string = "initializer";

Since you want an array of these strings, just make it an array and
provide an aggregate initializer as in the first example above.  If
the first example doesn't work, I'd begin suspecting your compiler.  I
just tried it on Borland C++ Builder 1.0 on my PC and that also
accepted the construct.
--
http://www.xmission.com/~legalize/    Legalize Adulthood!
[e-mail]
``Ain't it funny that they all fire the pistol,     <URL: http://
  at the wrong end of the race?''--PDBT     www.eden.com/~thewho>

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: "Dean-Christian Strik" <[e-mail]>
Subject: Re: (fractdev) Fractint in dosemu
Date: 31 Jul 1998 22:35:15 +0200

Kragen,

:> >> >  What happens if you use "symmetry=no"?
:>
:> Where do I use that? <<
:>
:> Either on the command line or by pressing the <g> key.
:
:I tried both.  It said it didn't know what I was talking about.


It should be "symmetry=none".

Christian




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: "Peter Gavin" <[e-mail]>
Subject: RE: (fractdev) C complex types
Date: 31 Jul 1998 20:10:41 -0400

Okay... I have something equivalent to this: (I'm using VC++ 4, BTW)

class stupid {
private:
    static const wchar_t * const adjectives[];
};

const wchar_t * const stupid::adjectives[] = { _T("one"), _T("two"),
_T("three") };

It may be the fact that it's a static class member thats messing it up...  I
get
this as a complaint: "error C2230: 'initializing' : indirection to different
types".

Any ideas?



Pete
<[e-mail]>

//  -----Original Message-----
//  From: [e-mail]
//  [mailto:[e-mail]]On Behalf Of Phil McRevis
//  Sent: Friday, July 31, 1998 1:32 PM
//  To: [e-mail]
//  Subject: Re: (fractdev) C complex types
//
//
//
//  In article <000001bdbc48$ed76e060$ea83cdcf@p60>,
//      "Peter Gavin" <[e-mail]>  writes:
//  > I want to make an array of constant pointers to constant
//  chars, such as:
//  >
//  > typedef const char /* a bunch of confusing stuff */ stupid;
//  >
//  > stupid adjectives =3D { "idiot", "moron", "Bill Gates" }
//
//  gcc accepted this perfectly fine:
//
//      const char *const adjectives[3] = { "one", "two", "three" };
//
//  did you try that?  This is a const character string:
//
//      const char *constant_string = "initializer";
//
//  That says the characters themselves can't be changed, but the pointer
//  can be changed, i.e. "constant_string++" is valid.  This is a constant
//  pointer to a constant character string:
//
//      const char *const constant_ptr_to_constant_string = "initializer";
//
//  Since you want an array of these strings, just make it an array and
//  provide an aggregate initializer as in the first example above.  If
//  the first example doesn't work, I'd begin suspecting your compiler.  I
//  just tried it on Borland C++ Builder 1.0 on my PC and that also
//  accepted the construct.
//  --
//  http://www.xmission.com/~legalize/    Legalize Adulthood!
//  [e-mail]
//  ``Ain't it funny that they all fire the pistol,     <URL: http://
//    at the wrong end of the race?''--PDBT     www.eden.com/~thewho>
//
//  --------------------------------------------------------------
//  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: "Peter Gavin" <[e-mail]>
Subject: RE: (fractdev) C complex types
Date: 31 Jul 1998 20:13:46 -0400

Geez, I'm an idiot...  it was the stupid _T() macro...  forgive me, please?
:)



Pete
<[e-mail]>

//  -----Original Message-----
//  From: [e-mail]
//  [mailto:[e-mail]]On Behalf Of Phil McRevis
//  Sent: Friday, July 31, 1998 1:32 PM
//  To: [e-mail]
//  Subject: Re: (fractdev) C complex types
//
//
//
//  In article <000001bdbc48$ed76e060$ea83cdcf@p60>,
//      "Peter Gavin" <[e-mail]>  writes:
//  > I want to make an array of constant pointers to constant
//  chars, such as:
//  >
//  > typedef const char /* a bunch of confusing stuff */ stupid;
//  >
//  > stupid adjectives =3D { "idiot", "moron", "Bill Gates" }
//
//  gcc accepted this perfectly fine:
//
//      const char *const adjectives[3] = { "one", "two", "three" };
//
//  did you try that?  This is a const character string:
//
//      const char *constant_string = "initializer";
//
//  That says the characters themselves can't be changed, but the pointer
//  can be changed, i.e. "constant_string++" is valid.  This is a constant
//  pointer to a constant character string:
//
//      const char *const constant_ptr_to_constant_string = "initializer";
//
//  Since you want an array of these strings, just make it an array and
//  provide an aggregate initializer as in the first example above.  If
//  the first example doesn't work, I'd begin suspecting your compiler.  I
//  just tried it on Borland C++ Builder 1.0 on my PC and that also
//  accepted the construct.
//  --
//  http://www.xmission.com/~legalize/    Legalize Adulthood!
//  [e-mail]
//  ``Ain't it funny that they all fire the pistol,     <URL: http://
//    at the wrong end of the race?''--PDBT     www.eden.com/~thewho>
//
//  --------------------------------------------------------------
//  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: Jonathan Osuch <[e-mail]>
Subject: Re: (fractdev) Fractint in dosemu
Date: 31 Jul 1998 21:40:08 -0400

Kragen,

>> I'm running dosemu in X-windows on a Cirrus Logic CLGD5446.  While it'=
s
possible it might be the video adapter, I believe it's more likely to be
dosemu.  I haven't seen it with any other programs (in dosemu or
otherwise).

>  Depending on your video adapter, what happens when you force
> the detection with the command line adapter=3Dxxxx?  This should preven=
t
the
> use of the VESA routines.

I'm not sure what xxxx should be.
<<

Yes, I see the problem.  First, try: adapter=3Dvga.  This will limit Frac=
tint
to only vga modes, but if it works okay, it's a clue to where the problem=

may lie.

Then, try: vesadetect=3Dno

This second should autodetect your video card.  For some reason we didn't=

implement adapter=3Dcirrus.

>> > >> >  What happens if you use "symmetry=3Dno"?
> =

> Where do I use that? <<
> =

> Either on the command line or by pressing the <g> key.

I tried both.  It said it didn't know what I was talking about.
<<

Sorry, my fault.  As Christian pointed out, it should be symmetry=3Dnone.=


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] (Kragen)
Subject: RE: (fractdev) C complex types
Date: 31 Jul 1998 22:34:06 -0400 (EDT)

On Fri, 31 Jul 1998, Peter Gavin wrote:
> Geez, I'm an idiot...  it was the stupid _T() macro...  forgive me, please?
> :)

Lemme guess... you didn't define UNICODE, so _T("") was a const char *,
not a const wchar_t *?

Kragen


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"