 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Would it be possible to obtain the VC6 EXE compiled with debugging
information?
I've tried to reproduce this floating pointer overflow/underflow on my
Pentium II 450 and have been unable too. Which leads me to believe that
either there's some incompatibility with the Athlon processor, or there's
something wrong with this computer, which really bugs me since it never
occured before Beta 13 and I can't find anything in the changes that would
effect the scene I was writing.
Last night I tried it with Beta 14 and I got a blue screen saying something
like VMM32.VXD has caused an exception. It only seems to occur when
rendering a very long scene. It's not just isosurfaces as I originally
thought because I had it freeze on a different scene where none were
present. Tonight I'm going to try to render the scene with the isosurfaces
running it in VC6 to see if it can pick anything up. However, a Debug EXE
would be of great help since it could detect such things as memory leaks,
damaged blocks and assertions.
Thanks in advance for anything anyone can think of....
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Sorry, here's the specs again:
Windows 98, 512meg DDR, Athlon 1900+
"Chris Becker" <cmb### [at] rit edu> wrote in message
news:3c9f5e38@news.povray.org...
> Would it be possible to obtain the VC6 EXE compiled with debugging
> information?
>
> I've tried to reproduce this floating pointer overflow/underflow on my
> Pentium II 450 and have been unable too. Which leads me to believe that
> either there's some incompatibility with the Athlon processor, or there's
> something wrong with this computer, which really bugs me since it never
> occured before Beta 13 and I can't find anything in the changes that would
> effect the scene I was writing.
>
> Last night I tried it with Beta 14 and I got a blue screen saying
something
> like VMM32.VXD has caused an exception. It only seems to occur when
> rendering a very long scene. It's not just isosurfaces as I originally
> thought because I had it freeze on a different scene where none were
> present. Tonight I'm going to try to render the scene with the isosurfaces
> running it in VC6 to see if it can pick anything up. However, a Debug EXE
> would be of great help since it could detect such things as memory leaks,
> damaged blocks and assertions.
>
> Thanks in advance for anything anyone can think of....
>
>
>
>
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <3c9f5e38@news.povray.org> , "Chris Becker" <cmb### [at] ritedu>
wrote:
> I've tried to reproduce this floating pointer overflow/underflow
The problem you had was unique to beta 13 and had been confirmed in the
thread:
Newsgroups: povray.beta-test
Subject: strange recursion error
Date: Fri, 15 Mar 2002 17:46:31 +0100
Message-ID: <at849uk68rnqhu92mhbkng6ho01tl23178@4ax.com>
Xref: news.povray.org povray.beta-test:5110
Thorsten
____________________________________________________
Thorsten Froehlich
e-mail: mac### [at] povrayorg
I am a member of the POV-Ray Team.
Visit POV-Ray on the web: http://mac.povray.org
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Well, I tried it with Beta 14 and that's when I recieved the blue screen.
I'll try it again tonight and see if it happens again. In the second scene I
tried I experience a freeze and I wasn't using functions. I was doing a
simple scene but with a very high quality focul blur, so the scene took
about 2 hours. The first time I tried to render it it got to 50% and froze.
The second time I rebooted first and then it was able to do it. So, to me,
it seems to occur with very large extensive calculations.
If it is a floating point stack overflow/underflow, it may be something
different in the Athlon XP handling of floating point numbers than the
Pentium II. Or it could be some simple memory leak that's damaging the
program somehow. Or it may be something is wrong with my computer. I ran the
long BIOS memory test and it didn't find any errors so I'm still confused on
what this could be. Plus, no one else seems to have the same problem that I
am.
If anyone wants to try this on an Athlon XP I could post the full scene
code. I've never been able to get a full rendering since Beta 12. But note
that it would take all night and probably then some.
Arg, I'm so confused and frustrated...
"Thorsten Froehlich" <tho### [at] trfde> wrote in message
news:3c9f687c@news.povray.org...
> In article <3c9f5e38@news.povray.org> , "Chris Becker" <cmb### [at] ritedu>
> wrote:
>
> > I've tried to reproduce this floating pointer overflow/underflow
>
> The problem you had was unique to beta 13 and had been confirmed in the
> thread:
>
> Newsgroups: povray.beta-test
> Subject: strange recursion error
> Date: Fri, 15 Mar 2002 17:46:31 +0100
> Message-ID: <at849uk68rnqhu92mhbkng6ho01tl23178@4ax.com>
> Xref: news.povray.org povray.beta-test:5110
>
>
> Thorsten
>
>
> ____________________________________________________
> Thorsten Froehlich
> e-mail: mac### [at] povrayorg
>
> I am a member of the POV-Ray Team.
> Visit POV-Ray on the web: http://mac.povray.org
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <3c9f938b$1@news.povray.org> , "Chris Becker" <cmb### [at] ritedu>
wrote:
> If it is a floating point stack overflow/underflow, it may be something
> different in the Athlon XP handling of floating point numbers
You are on the wrong track here: The message is generated by the function VM
and has nothing to do with the program stack.
Thorsten
____________________________________________________
Thorsten Froehlich, Duisburg, Germany
e-mail: tho### [at] trfde
Visit POV-Ray on the web: http://mac.povray.org
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
My mistake. So what exactly is causing these lock ups and blue screens that
I'm experiencing? It doesn't always say floating point error. How do I avoid
or fix it?
"Thorsten Froehlich" <tho### [at] trfde> wrote in message
news:3ca032f2@news.povray.org...
> In article <3c9f938b$1@news.povray.org> , "Chris Becker" <cmb### [at] ritedu>
> wrote:
>
> > If it is a floating point stack overflow/underflow, it may be something
> > different in the Athlon XP handling of floating point numbers
>
> You are on the wrong track here: The message is generated by the function
VM
> and has nothing to do with the program stack.
>
> Thorsten
>
> ____________________________________________________
> Thorsten Froehlich, Duisburg, Germany
> e-mail: tho### [at] trfde
>
> Visit POV-Ray on the web: http://mac.povray.org
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
If I may interject...
This sounds more like bad RAM to me.
In my experience RAM problems on a PC don't show themselves until you get a
hungry
program in there like POV-Ray, actually using it.
Have you tried running any other memory-hungry applications, like a game or
something?
"Chris Becker" <cmb### [at] ritedu> wrote in message
news:3ca08767$1@news.povray.org...
> My mistake. So what exactly is causing these lock ups and blue screens
that
> I'm experiencing? It doesn't always say floating point error. How do I
avoid
> or fix it?
>
> "Thorsten Froehlich" <tho### [at] trfde> wrote in message
> news:3ca032f2@news.povray.org...
> > In article <3c9f938b$1@news.povray.org> , "Chris Becker"
<cmb### [at] ritedu>
> > wrote:
> >
> > > If it is a floating point stack overflow/underflow, it may be
something
> > > different in the Athlon XP handling of floating point numbers
> >
> > You are on the wrong track here: The message is generated by the
function
> VM
> > and has nothing to do with the program stack.
> >
> > Thorsten
> >
> > ____________________________________________________
> > Thorsten Froehlich, Duisburg, Germany
> > e-mail: tho### [at] trfde
> >
> > Visit POV-Ray on the web: http://mac.povray.org
>
>
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Mon, 25 Mar 2002 12:30:30 -0500, "Chris Becker" <cmb### [at] ritedu>
wrote:
>Last night I tried it with Beta 14 and I got a blue screen saying something
>like VMM32.VXD has caused an exception. It only seems to occur when
>rendering a very long scene.
Sounds like a hardware problem to me. Check out the RAM, cooling and
power supply (i.e. how stable the core voltage is, though for your
beast you'll need a 300+ MHz oscilloscope)
Peter Popov ICQ : 15002700
Personal e-mail : pet### [at] vipbg
TAG e-mail : pet### [at] tagpovrayorg
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I ran a memory test, it found 1 bit bad. What a kick in the teeth.
I'll try switching the slot. Let's just hope that "life time" warranty is
still intact since January.
Somehow though, this is more relieving than a bug in POV-Ray that no one
else can seem to find, at least its fixable by me. : ) [/end happy thought]
"Chris Becker" <cmb### [at] ritedu> wrote in message
news:3c9f5e38@news.povray.org...
> Would it be possible to obtain the VC6 EXE compiled with debugging
> information?
>
> I've tried to reproduce this floating pointer overflow/underflow on my
> Pentium II 450 and have been unable too. Which leads me to believe that
> either there's some incompatibility with the Athlon processor, or there's
> something wrong with this computer, which really bugs me since it never
> occured before Beta 13 and I can't find anything in the changes that would
> effect the scene I was writing.
>
> Last night I tried it with Beta 14 and I got a blue screen saying
something
> like VMM32.VXD has caused an exception. It only seems to occur when
> rendering a very long scene. It's not just isosurfaces as I originally
> thought because I had it freeze on a different scene where none were
> present. Tonight I'm going to try to render the scene with the isosurfaces
> running it in VC6 to see if it can pick anything up. However, a Debug EXE
> would be of great help since it could detect such things as memory leaks,
> damaged blocks and assertions.
>
> Thanks in advance for anything anyone can think of....
>
>
>
>
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Tue, 26 Mar 2002 16:13:28 -0500, "Chris Becker" <cmb### [at] ritedu>
wrote:
>I ran a memory test, it found 1 bit bad. What a kick in the teeth.
1 bit bad is 1 bit too many. Replace it if the warranty holds.
>I'll try switching the slot. Let's just hope that "life time" warranty is
>still intact since January.
Is it a Viking Lifetime? If so, be content, the warranty is for real.
Peter Popov ICQ : 15002700
Personal e-mail : pet### [at] vipbg
TAG e-mail : pet### [at] tagpovrayorg
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
