ftp.xmission.com/2014.06.ftp.xmission.com.tar/ftp.xmission.com/pub/lists/fractdev/archive/v01.n034

Internet Message Format | 2000-08-26 | 42KB

From: [e-mail] (fractdev-digest)
To: [e-mail]
Subject: fractdev-digest V1 #34
Reply-To: fractdev-digest
Sender: [e-mail]
Errors-To: [e-mail]
Precedence: bulk


fractdev-digest        Sunday, August 27 2000        Volume 01 : Number 034




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

Date: Mon, 31 Jul 2000 17:47:15 +1000
From: "Paul de Leeuw" <[e-mail]>
Subject: Re: Ideas for 20.1 based on wishlist

Hi Tim,

I tried to compile the Xwindows float only version but it appears to be full
of 16 bit code (far, seg, offset rtc) which blows the microsnot compiler out
of the water. I would dearly love to merge Manpwin and Fractint to produce a
true colour, 32 bit windows application.

Any suggestions?

Thanks,

Paul.
- ----- Original Message -----
From: Tim Wegner <[e-mail]>
To: <[e-mail]>
Sent: Monday, July 31, 2000 2:33 PM
Subject: Re: Ideas for 20.1 based on wishlist


> Jonathan wrote:
>
> > > All developers,
> >
> > Unfortunately, this still appears to be a fairly small list.
>
> Unfortunately true. I am finding I need to spend my time these days
> learning. This is from my point of view not a bad thing, but it
> pushes any direct fractint work down the priority list.
>
> I checked out Wxwindows again recently, and this still looks like a
> good way to get a portable GUI.
>
> I got ssh (secure shell) installed on my Linux firewall. I just tried
> copying a file from julia.fractalus.org using scp and it works great.
> There is a windows command line utility called pscp available at
> http://www.chiark.greenend.org.uk/~sgtatham/putty/
> that has the same functionality. This makes FTP unnecessary,
> though FTP with a good Windows client is certainly easier.
>
> 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"


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

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

Date: Mon, 31 Jul 2000 07:43:58 -0500
From: "Jonathan Osuch" <[e-mail]>
Subject: Re: Ideas for 20.1 based on wishlist

Paul,

> I tried to compile the Xwindows float only version but it appears to be
full
> of 16 bit code (far, seg, offset rtc) which blows the microsnot compiler
out
> of the water. I would dearly love to merge Manpwin and Fractint to produce
a
> true colour, 32 bit windows application.

If you define XFRACT, all those will go away.  Someday we will need to take
them out, but for now this works.

Since the Allegro package has a Windows port, it may be advantageous to look
at it.  Also, you should probably work from the source I am using.  This has
the advantage of using a driver system for environment specific functions
(graphics & sound).  It still needs a lot of work done so that configure and
make can be used.

Jonathan



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

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

Date: Mon, 31 Jul 2000 07:56:32 -0600
From: Tim Wegner <[e-mail]>
Subject: Re: Ideas for 20.1 based on wishlist

Paiul asked:

> I tried to compile the Xwindows float only version but it appears to be full
> of 16 bit code (far, seg, offset rtc) which blows the microsnot compiler out
> of the water. I would dearly love to merge Manpwin and Fractint to produce a
> true colour, 32 bit windows application.
>
> Any suggestions?

Xfractint has some defines that eliminate all the far keywords etc.
Probably you are just missing something turned on with the
XFRACT define. I'm off to work so I can't find it for you you, but if
you search for XFRACT in the dot h files it should be easy to find.

Tim


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

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

Date: Wed, 02 Aug 2000 15:18:40 GMT
From: "Rupert Millard" <[e-mail]>
Subject: (fractint) Ideas for version 20.1

Johnathon,


About the alternating iterations thing, how about making two new types
manzpoweralt and julzpoweralt which are a bit more advanced than what he
suggested. These could request the exponent pattern as an expression
consisting of numbers and commas with square brackets. The exponents are
separated by commas, for a repeating sequence of length n 'n[a,b,c...]' can
be used. It can be nested too for example:

2,3[4,5],7,3[6,7[2,3]]],[2,3] nb. the last doesn't have a multiplier 'coz it
goes on until it bails-out.


I really think you ought to try implementing the standard escape-time
bailout tests in the parser, they'd be really helpful - and although I don't
know anything about C (do you know where I can get a tutorial in it?) I
can't see why it would be so difficult.


I've found a bug in simplgif which only occurs sometimes and when I try to
simplgif a fractint divide and conquer image. makemig.gif consists of 25
1024x768 'panes' and when I do 'simplgif makemig.gif simplgif.gif' with one
set the top row of pixels changes to the middle colours (cyan in blues.map).
I it only happens on this parameter set which it happens on is:

Test               { ; Version 2000 Patchlevel 12
  reset=2000 type=mandel
  center-mag=-1.41926473620118600/+0.04745164936908675/4194785/1/109.999
  params=0/0 float=y maxiter=5000 inside=0 logmap=223 periodicity=0
  cyclerange=0/255 sound=off
  colors=000000<25>00d00f00h00i00k00m<26>0ew0fx0hx0iy0ky0mz<25>nwzpwzrxz<3\
  >zzz<42>7zz6zz4zz3zz2zz0zz<42>07z06z04z03z02z00z<60>002000000
  }

Makemig this as 5x5 1024x768 images but before you run it uncomment out the
call to simplgif.

I also have another problem with makemig, when fractint looks for
frmig_??.gif it looks in the directory which it is in, shouldn't it look
where it saved them?


From,

Rupert


P.S. Did you know that symmetry very often slows things down? For the
unzoomed M-Set: 2.85 secs without symmetry at 1024x768x256, 10.71 with
symmetry (I'm on a PIII 500 MHz)?
________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com


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

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

Date: Wed, 2 Aug 2000 12:31:54 -0500
From: "Jonathan Osuch" <[e-mail]>
Subject: Re: (fractint) Ideas for version 20.1

Rupert,

> About the alternating iterations thing, how about making two new types
> manzpoweralt and julzpoweralt which are a bit more advanced than what he
> suggested.

I can certainly assist anyone willing to attempt implementing this.  Adding
new fractal types is not on my radar screen at this point.  For various
reasons.

> I really think you ought to try implementing the standard escape-time
> bailout tests in the parser, they'd be really helpful - and although I
don't
> know anything about C (do you know where I can get a tutorial in it?) I
> can't see why it would be so difficult.

The difficulty arises because the fast parser is implemented in assembly
language.  The 'slow' parser is implemented in C, but has none of the
optimizations of the 'fast' parser.  I know I've looked at it, and I know
that I had a good reason for not implementing it.  I just can't remember
what it was.

> I've found a bug in simplgif which only occurs sometimes and when I try to
> simplgif a fractint divide and conquer image.

I believe we know that.  The ability to recombine the individual mig images
was subsequently built into Fractint and that recombination works correctly.
When I just tried it, Fractint generated the file "fractmig.gif"
automatically and it was not necessary to run simplgif.  I did not see the
individual files.  This feature is not very well documented.  We should
remove the line containing the simplgif call from the batch file and not
include simplgif with the distribution.  Also, you can use disk video mode
for images up to 32Kx32K.

> I also have another problem with makemig, when fractint looks for
> frmig_??.gif it looks in the directory which it is in, shouldn't it look
> where it saved them?

It looks in the directory set by the startup parameter workdir.  If you are
setting a different path with the filename= command, this will be a problem.

> Did you know that symmetry very often slows things down? For the
> unzoomed M-Set: 2.85 secs without symmetry at 1024x768x256, 10.71 with
> symmetry (I'm on a PIII 500 MHz)?

That doesn't happen on my machine, but then I only have a slow 233 MHz
pentium.  I might ask, however, why does it take so long (either case) on
your machine?  I get 1.21 seconds with symmetry and 1.32 seconds without at
that resolution.  You *do* have floating point math turned on, don't you?
Or, are you running Xfractint, which changes some of the above discussion?

Jonathan



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

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

Date: Sat, 5 Aug 2000 23:19 0000
From: [e-mail]
Subject: Forky patch question

Hi Jon + Tim + Rich + Folks,

     Just a quick question... I've knocked together a routine for writing
lsystem output as dxf vector file format which works fine BUT
in order to be at all legible it uses a lot of string literals in
printf() statements.

This completely uses up the near ram on the standard dos build on
some machine configurations ( I could get it going by trimming
environment variables right down on my machine forinstance..
*very* tight fit!)

It leaves a few hundred bytes free in a float only build however.

I know I could do the usual substitution of overlaid arrays for
the literals but, as it is, the dxf output routine is a readable reference
on the file format... obfuscating it with near ram saving tricks seems
a shame really and quite against the open source ethic.

Now this build could well be useful to a number of people who want
to
import lsystems in cad packages or whatever so I'd like to release
a patch... the question is what should this patch be releaed as?

For xfractint it should go in fine on all systems (no near mem
problems)
but it shouldn't really  get into the standard dos version as it'll fail
on quite a few machines. The float only build will most likely
work everywhere though.....
so basically I suppose I'm stirring up some
discussion of the issue of source forking and what we do about
it.

Cheers,
          Robin.

P.S. maybe 3d Lsystems might come out of this too... no promises ;-)


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

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

Date: Sat, 5 Aug 2000 19:24:14 -0500
From: "Jonathan Osuch" <[e-mail]>
Subject: Patch 13

Patch 13 is on the ftp site.  The executable is there as fradev20.0.13.zip.

This patch adds parameters p4 and p5 to the formula parser and now checks
all the parameters when the symmetry options XAXIS_NOREAL and XAXIS_NOIMAG
are used.

I purposely did not add the two new parameters to the evolver.  Maybe
someday.

Jonathan




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

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

Date: Sat, 5 Aug 2000 18:57:04 -0500
From: "Jonathan Osuch" <[e-mail]>
Subject: Re: Forky patch question

Robin,

> so basically I suppose I'm stirring up some
> discussion of the issue of source forking and what we do about
> it.

This is already a problem.  We have two versions of Fractint and two
versions of Xfractint.  One with integer math and the other without.  Then
there is the source I am using for the Allegro graphics package, based on
the float only Xfractint.

Once we get the Allegro graphics package working and then port the source
over to djgpp, the DOS code can go away.  But that's a long ways off.

Jonathan



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

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

Date: Sun, 6 Aug 2000 15:47:25 -0400
From: Sylvie Gallet <[e-mail]>
Subject: Patch 13: bug

Hi Jonathan,

  In patch 13, generate any type=3Dformula fractal, go to the <tab> scree=
n,
and return to the image: the top of the image gets corrupted with a
horizontal dotted line.

  Cheers,

        - Sylvie

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

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

Date: Sun, 6 Aug 2000 16:14:10 -0500
From: "Jonathan Osuch" <[e-mail]>
Subject: Re: Patch 13: bug

Sylvie,

> In patch 13, generate any type=formula fractal, go to the <tab> screen,
> and return to the image: the top of the image gets corrupted with a
> horizontal dotted line.

Yes, I can make it do that.  Only if the fractal has not finished
generating.  This is not unique to patch 13, however.  Version 20.0 does
this also.  The good news is that if you use textsafe=save, it doesn't
happen.  Which is why I didn't see it earlier.  I have textsafe=save set in
my SSTOOLS.INI file.

This problem is not unique to the formula type, either.  The lorenz3d types,
rossler3d, sierpinski, and kamtorus3d all show extraneous dots when <tab> is
used before image completion.

There was a while there when I had made textsafe=save the default, but I
changed it back to how it was before because of problems with the memory
routines.  I think those problems have all been ironed out.

OTOH, once we are ported to a windowed environment, the textsafe variable
can go away.  For example, the Allegro graphics package port I'm working on
maintains separate image and text bit maps.  The screens are swapped by
blitting the relevant bit map to the screen bitmap.

Jonathan



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

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

Date: Sun, 6 Aug 2000 19:38:01 -0400
From: Lee Skinner <[e-mail]>
Subject: Patch 13: bug

Jonathan,

  Here's another bug (parser) - its been in Fractint for some time:

sc_not_test {
;
; try real(p1) =3D -1, -2, 1, or 2
;
; p1 =3D -1 should produce the same image as p1 =3D -2
;

if     (p1 =3D=3D  1) ,  par1 =3D 1e+2
elseif (p1 =3D=3D  2) ,  par1 =3D 10^(2)
elseif (p1 =3D=3D -1) ,  par1 =3D 1e-2
elseif (p1 =3D=3D -2) ,  par1 =3D 10^(-2)
endif

z =3D c =3D pixel
:
z =3D par1 * z^100 + c
|z| <=3D 4
}

When p1 =3D -2, the image is different (bigger).  I've noticed this in ot=
her
formulas where replacing 1e-n with 10^(-n) generates different images.

Lee

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

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

Date: Sun, 6 Aug 2000 20:03:21 -0500
From: "Jonathan Osuch" <[e-mail]>
Subject: Re: Patch 13: bug

Lee,

> Here's another bug (parser) - its been in Fractint for some time:

No, this is a misunderstanding that has been around for some time.  The
parser has never understood scientific notation.  Therefore, 1e-2 is
interpreted as: the variable 1e (which is initialized to 0) minus 2.  I
modified your formula so I could better see what was happening.

test_01 {

;  p1=1 produces the same image as p1=5
;  p1=3 is similar due to the relative size of  -2 vs 0.01
;  Note that e by itself is a predefined variable, hence p1=4 is different

if     (p1 ==  1) ,  par1 = 1e-2
elseif (p1 ==  2) ,  par1 = 10^(-2)
elseif (p1 ==  3) ,  par1 = (1e-2) - (10^(-2))
elseif (p1 ==  4) ,  par1 = (e - 2)
elseif (p1 ==  5) ,  par1 = -2
elseif (p1 ==  6) ,  par1 = -1
endif

z = c = pixel
:
z = z * z + par1 + c
|z| <= 4
}

Granted, I would still call the fact that 1e is recognized as a variable a
bug.  OTOH, if we fix it and a lot of formulas use this notation, we break a
lot of images.  Not that it would be the first time <G>.

Jonathan



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

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

Date: Sun, 6 Aug 2000 22:42:27 -0400
From: Lee Skinner <[e-mail]>
Subject: Patch 13: bug

Jonathan,

>>  OTOH, if we fix it and a lot of formulas use this notation, we break =
a
lot of images.  Not that it would be the first time <G>.<<

Thanks for the explanation.  I recommend leaving the bug in for the reaso=
n
you mentioned.

Lee

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

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

Date: Sun, 6 Aug 2000 23:48:47 -0400
From: George Martin <[e-mail]>
Subject: Re: Patch 13: bug

Friends,

The following awoke me from the dead.

>
No, this is a misunderstanding that has been around for some time.  The
parser has never understood scientific notation.  Therefore, 1e-2 is
interpreted as: the variable 1e (which is initialized to 0) minus 2.  I
modified your formula so I could better see what was happening.
<

The parser has all the background code to recognize scientific notation
correctly, but needs a bit of work to finish the implementation. This was=

an unfinished project of mine when v. 20 went out. =


In prior versions, notation such as  1e or 2x was treated as a named
variable, and, as an unintended consequence of the old code, initialized =
to
the lead digit (in the examples, "1" or "2" ), not the usual 0. In versio=
n
20, the error checking is done with new code, and a notation such as 2x
will yield an error (presumably the author meant 2*x). I went through the=

entire orgform compilation before v. 20 went out, and corrected all the
formulas where the more stringent error checking in version 20 was
rendering the formula unworkable.

However, because the new error checking code recognizes scientific
notation, 1e is not recognized as an error and slips through the cracks. =
It
then is evaluated in the old way, not as scientific notation. <sigh>
 =

George Martin

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

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

Date: Mon, 7 Aug 2000 07:53:21 -0500
From: "Jonathan Osuch" <[e-mail]>
Subject: Fractint buglist

All,

Is anybody familiar enough with Spanish to help with the latest entry on the
Fractint buglist?  Here is the message:

************
please need edit zoom correctly in archives with extension gif
why not apply zoom correctly in gif images not purifity this program tilt?

please need help for using this program in medical images

thank you

please response in spanish
************

Jonathan



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

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

Date: Tue, 8 Aug 2000 06:20:32 -0400
From: Sylvie Gallet <[e-mail]>
Subject: Re: Patch 13: bug

Hi Jonathan,

>> Yes, I can make it do that.  Only if the fractal has not finished
>> generating.  This is not unique to patch 13, however.  Version 20.0 do=
es
>> this also.  The good news is that if you use textsafe=3Dsave, it doesn=
't
>> happen.  Which is why I didn't see it earlier.  I have textsafe=3Dsave=
 set
>> in my SSTOOLS.INI file.

  I had never seen it before, this is why I thought it was a new bug! =

You're right, with textsafe=3Dsave, it doesn't happen.

  Cheers,

        - Sylvie

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

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

Date: Fri, 18 Aug 2000 11:50:29 -0500
From: "Jonathan Osuch" <[e-mail]>
Subject: Patch 14

Tim,

The patch 14 diff and executable file are available at
http://www.fractint.org/ftp/

This fixes both the regeneration of completed images and the incorrect
determination of random numbers, which caused the regeneration to have
different subimages.

Jonathan



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

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

Date: Sat, 19 Aug 2000 15:38:50 GMT
From: "Rupert Millard" <[e-mail]>
Subject: Standard bail-out tests in the parser.

Hi Jonathon,

I know its been a long time since you e-mailed me but I haven't had the time
to send you another nag... until now!

Although I'm not a programmer I cannot see why you cannot implement the
bailout tests in the parser. At worst surely you could do what I do all the
time in QBasic, that is copy and paste the relevant code.

Also, I'd like to know what form a true-colour palette editor (or is that a
contradiction in terms?) would take. Would it still be capable of 256 colour
stripey effects or restrictred to smooth gradients like Ultra-fractal and
Fractal-eXtreme?

Finally, what *is* allegro?

Thanks,

From,

Rupert
________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com


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

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

Date: Sat, 19 Aug 2000 13:12:23 -0500
From: "Jonathan Osuch" <[e-mail]>
Subject: Re: Standard bail-out tests in the parser.

Rupert,

> Although I'm not a programmer I cannot see why you cannot implement the
> bailout tests in the parser. At worst surely you could do what I do all
the
> time in QBasic, that is copy and paste the relevant code.

Remember that the parser is actually compiling the formula into machine
code.  What you are asking for is a way to switch the bailout test from the
<z> screen.  Depending on where the compilation is taking place, before or
after the <z> screen, this may be workable.  We would need to change the
designation for the bailout, however.  This would be a good thing for two
reasons, we could accommodate orbit type fractals, and we could then have a
bailout variable instead of having to use part of p1-p5.

Now all we need is someone to do the work.

> Also, I'd like to know what form a true-colour palette editor (or is that
a
> contradiction in terms?) would take. Would it still be capable of 256
colour
> stripey effects or restrictred to smooth gradients like Ultra-fractal and
> Fractal-eXtreme?

I have no idea.  The smooth gradients would seem reasonable, however.  Isn't
that why everybody wants true-color?

> Finally, what *is* allegro?

Have a look at: http://www.talula.demon.co.uk/allegro/
It's a graphics interface that has been ported to several different
environments, including Linux, djgpp, and Windows.

Jonathan



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

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

Date: Sun, 20 Aug 2000 13:29:03 -0600
From: Tim Wegner <[e-mail]>
Subject: News and a version question for Jonathan

I have finally gotten my MSC 7 compiler environment installed on
my new PC. Not too quick considering I got the PC in March. :-)

My computing environment now consists of:

1. An HP 9680C with dual booting with Win98 and Mandrake 7.1.
This is the machine that I just installed MS C/C++ 7.0 on.

2. My old PC with a 150 mhz motherboard bristling with various
upgrades that still has my old computing environment under
Win95, which also has Mandrake 7.0 and Slakware, which I'll
shortly get rid of. This is now my wife Susan's machine.

3. A 166 mhz Dell Optiplex that I bought on Ebay for $75 that at
the moment has Win98 and Mandrake 7.1. I am planning on using
this as a server and for network experimentation. It has 32mb/2gb.

4. My old 66 mhz 486, which is running the one-floppy EigerStein
LRP (Linux Router Project) and the Seawall firewall. This machine
acts as a firewall and router for my home network. The other three
machines connect to my DSL line through the firewall/router. The
firewall machine also hosts a single web page you can access at
twegner.dynodns.net. There isn't much content on this page yet,
it's mostly an experimental.

My college-age son visited this weekend and there were actually
some times that all three of us were accessing the internet
simultaneously through the firewall/router (the family that computes
together stays together?) :-)

Now for the versions question. I am now up to date with patch 14. I
noticed that the patches apply to the "regular" (integer) version of
Fractint. Should I update the float only version and the Xfractint
version to patch 14? Is the float-only strain of Xfractint now the
"official" one?

Tim


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

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

Date: Sun, 20 Aug 2000 16:51:52 -0500
From: "Jonathan Osuch" <[e-mail]>
Subject: Re: News and a version question for Jonathan

Tim,

> Now for the versions question. I am now up to date with patch 14. I
> noticed that the patches apply to the "regular" (integer) version of
> Fractint. Should I update the float only version and the Xfractint
> version to patch 14? Is the float-only strain of Xfractint now the
> "official" one?

I updated the Allegro graphics port source to patch 13.  This source is
based on the float-only strain of Xfractint and required some changes to be
updated by hand.  Since I've already done this once, it would be easier for
me to do it.

I am also behind in uploading the Xfractint source for patches 12-14.  I
still have a source at version 12, so this isn't such a big problem.  Do we
need to include all the MAP files (and others) in the developer's source
distribution?

The problem with making the float-only strain of Xfractint the "official"
one is that there are features that are broken in it that work in the
integer version.  The feature that comes to mind is the 3D image generation
when the <3> key is pressed.  I'm fairly certain it is only due to parts of
a patch getting misplaced.  I just haven't looked into it yet.

Jonathan



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

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

Date: Thu, 24 Aug 2000 10:36:33 -0500
From: "Jonathan Osuch" <[e-mail]>
Subject: Float only source & executable

Tim,

I updated and uploaded the DOS float only source and executables through
patch 14.

Jonathan



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

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

Date: Sat, 26 Aug 2000 12:07:18 GMT
From: "Rupert Millard" <[e-mail]>
Subject: Re: Standard bail-out tests in the parser.

Jonathon,

Thankyou very much for explaining that to me. I *think* I understand now.

From,

Rupert
________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com


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

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

Date: Sat, 26 Aug 2000 10:30:40 -0600
From: Tim Wegner <[e-mail]>
Subject: Re: Float only source & executable

Jonathan wrote:

> I updated and uploaded the DOS float only source and executables
> through patch 14.

I made small changes to the web site - removed the link to the FTP
files via FTP (since it doesn't work) and added Spanky's mirror link.
I also updated the links in the files directory to point to your latest
versions. I am thinking of adding "What's New" from the docs to the
web site.

Perhaps we should move old exe packages to an archive directory
to avoid confusion.

While this is probably less urgent now that development has
slowed down, I was wondering how much interest there would be in
my cutting some CDs that had the full source every patch version
(source and executable). I have found this invaluable for tracking
bugs. I do a binary search running test cases using different
developer versions, and can usually quicky pinpoint exactly which
patch caused a bug. The CD could also contain all the historical
releases that I have.

If we get into a "massive changes" mode, the historical patches will
be of much less interest. However I would be happy to cut such a
CD for any developer or user interested in tracking down bugs.

Tim


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

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

Date: Sat, 26 Aug 2000 13:37:40 -0500
From: "Jonathan Osuch" <[e-mail]>
Subject: Re: Float only source & executable

Tim,

> Perhaps we should move old exe packages to an archive directory
> to avoid confusion.

Yes, that's a good idea.

> While this is probably less urgent now that development has
> slowed down, I was wondering how much interest there would be in
> my cutting some CDs that had the full source every patch version
> (source and executable).

I'm interested.  I had started doing something similar with the version 19.6
patches.  I was putting the patch and the executable on a Zip disk.  Since I
was playing catch up, I lost interest.

> The CD could also contain all the historical
> releases that I have.

That would be useful to me.

BTW, I was unable to compile HC.EXE from the float only source.  Something
about a LIB missing.  Since I had to reinstall MSC, I have no doubt I missed
something.  You might want to try it just to make sure.

Jonathan



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

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

Date: Sat, 26 Aug 2000 20:01:19 -0500
From: "Jonathan Osuch" <[e-mail]>
Subject: Re: Float only source & executable

Tim,

> I am thinking of adding "What's New" from the docs to the
> web site.

I see that you have done so.  It could use a little editing in the place(s)
where a documentation link exists.

Jonathan



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

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

Date: Sat, 26 Aug 2000 20:00:54 -0600
From: Tim Wegner <[e-mail]>
Subject: Web page tweaks

I added a whatsnew.txt file to the ftp directory that we should
maintain when we update patches. There's a link to the file on the
web page. I have not yet dealt with the old versions or created an
archive directory.

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

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

Date: Sat, 26 Aug 2000 20:38:37 -0600
From: Tim Wegner <[e-mail]>
Subject: Web site administrivia

> I see that you have done so.  It could use a little editing in the place(s)
> where a documentation link exists.

Done - I only saw one place. You should have write permission on
the file should you need to edit them. Incidently, I can't change the
files you own (not that I need to). You might do a chmod 664 on
your files. I probably can't move them to an archive myself.

Tim


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

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

Date: Sat, 26 Aug 2000 20:51:55 -0500
From: "Jonathan Osuch" <[e-mail]>
Subject: Re: Web site administrivia

> You should have write permission on
> the file should you need to edit them. Incidently, I can't change the
> files you own (not that I need to). You might do a chmod 664 on
> your files. I probably can't move them to an archive myself.

I haven't been able to change the permissions lately using FTP Voyager.

Jonathan



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

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

Date: Sat, 26 Aug 2000 23:55:34 -0500
From: "Scott D. Boyd" <[e-mail]>
Subject: Re: Web site administrivia

Jonathan,

Try to ftp the site from the command line prompt. Then issue the
"chmod 664 filename" command, and presto, it's done! It should work
regardless of whether you're using Win9x or GNU/Linux.

Scott Boyd

 On Sat, 26 Aug 2000, you wrote:
>
> I haven't been able to change the permissions lately using FTP Voyager.
>
- --
[e-mail]
- -------------------
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"

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

Date: Sun, 27 Aug 2000 11:15:10 -0600
From: Tim Wegner <[e-mail]>
Subject: Re: Web site administrivia

Scott wrote:
Scott wrote:

> Try to ftp the site from the command line prompt. Then issue the
> "chmod 664 filename" command, and presto, it's done! It should work
> regardless of whether you're using Win9x or GNU/Linux.

This works using the built-in Windows FTP command line utility as
long as one knows the FTP commands. I can also  do the chmod
using CuteFTP's custom commands. However, what I like best is
to log in via secure shell. Then I can edit files and do any UNIX
commands at the command line. I can even copy files back and
forth with the secure copy client. I confess I only did this once to
test it, but it seemed to work. I usually transfer files via FTP
because it is very easy with a good client. Voyager probably has
the ability to do commands buried in it somewhere.

The Windows ssh program putty seems much easier to use than
Tiera Pro. It also comes with a command-line copy utility scp.
IMHO eventually this should replace FTP eventually because it is
safer, though I admit that good drag-and-drop windows FTP
programs are hard to beat.

You can always find putty and scp via a www.google.com search
(it pops up first). This is handy because the location has an
obscure URL. Putty is located at:

http://www.chiark.greenend.org.uk/~sgtatham/putty/

Of course from Linux you just use the ssh client. I haven't tried this
yet, but I have installed a ssh server on my Linux Router Project
firewall. I can log in to my Linux ssh server using Putty or Tiera Pro.

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"

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

End of fractdev-digest V1 #34
*****************************