|
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Kyle wrote:
>
>>Note you have to be more elaborate on what actually happens - 'crash' is
>>an extremely vague description.
>
> What further info can we provide at this time that will help diagnose
> the problem?
'crash' is not much more information than 'it does not work'. Your
description does not even tell us if this is a normal program behaviour
or an error.
> BTW, once the source is available,
The source is available - just not yet packaged for Windows. You can
download the Unix source package and merge it with the 1.1 Windows
source if you want to work on it.
Christoph
--
POV-Ray tutorials, include files, Landscape of the week:
http://www.tu-bs.de/~y0013390/ (Last updated 01 Jul. 2005)
MegaPOV with mechanics simulation: http://megapov.inetart.net/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
George Leroy Tirebiter wrote:
>
> OK, here is exactly what happens to me. I'll describe what happens when
> rendering the mechsim file
As said i can reproduce this on Windows so no need for a more detailed
description here.
Christoph
--
POV-Ray tutorials, include files, Landscape of the week:
http://www.tu-bs.de/~y0013390/ (Last updated 01 Jul. 2005)
MegaPOV with mechanics simulation: http://megapov.inetart.net/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> I can confirm this - it happens in the release version as well as with a
> MinGW build, it however does not happen on Linux.
Such a different behavior of Linux vs other platforms may well be the
sign of uninitialized variables (in particular pointers). Linux seems to be
sometimes quite permissive in this respect, i.e. it does not necessarily make
the program segfault when dereferencing an invalid pointer.
Of course that's not much of an help either...
- NC
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Nicolas Calimet wrote:
>
> Such a different behavior of Linux vs other platforms may well be the
> sign of uninitialized variables (in particular pointers). Linux seems
> to be
> sometimes quite permissive in this respect, i.e. it does not necessarily
> make
> the program segfault when dereferencing an invalid pointer.
Actually we have most likely found the problem and it *should* crash as
it is right now - the good old using-a-pointer-after-it-was-freed bug.
It only causes problems in scenes using Mechsim.
Christoph
--
POV-Ray tutorials, include files, Landscape of the week:
http://www.tu-bs.de/~y0013390/ (Last updated 01 Jul. 2005)
MegaPOV with mechanics simulation: http://megapov.inetart.net/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
dato3q$cpo$1@chho.imagico.de...
>
> It only causes problems in scenes using Mechsim.
FWIW, I just had a bad series of crashes (Megapov 1.2 Windows version, XP
SP2, "POV-Ray has a problem and must be closed" message) on scenes not
involving Mechsim (making HDR images using LightSys).
Some of them are reproducible (one at least was linked to a #debug stream)
while others aren't (the renders works fine for a while for 3-4 runs and
then crash POV). I haven't created a test scene but a slightly modified
version of hdr_environment.pov (with view=3, radiosity commented off
and -w300 -h150 +fh -ohdr_env.hdr in the command line) crashes either at the
beginning of the render or at the end.
G.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Gilles Tran wrote:
>
> FWIW, I just had a bad series of crashes (Megapov 1.2 Windows version, XP
> SP2, "POV-Ray has a problem and must be closed" message) on scenes not
> involving Mechsim (making HDR images using LightSys).
> Some of them are reproducible (one at least was linked to a #debug stream)
> while others aren't (the renders works fine for a while for 3-4 runs and
> then crash POV). I haven't created a test scene but a slightly modified
> version of hdr_environment.pov (with view=3, radiosity commented off
> and -w300 -h150 +fh -ohdr_env.hdr in the command line) crashes either at the
> beginning of the render or at the end.
Could you check if this is related to HDR output? I made some
modifications in the HDR code for version 1.2 so there might be new
problems related to this.
Note that any problem that only occurs in a second or later render will
of course not be observable in the Unix version.
Christoph
--
POV-Ray tutorials, include files, Landscape of the week:
http://www.tu-bs.de/~y0013390/ (Last updated 01 Jul. 2005)
MegaPOV with mechanics simulation: http://megapov.inetart.net/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
dau7vr$ppk$1@chho.imagico.de...
> Could you check if this is related to HDR output? I made some
> modifications in the HDR code for version 1.2 so there might be new
> problems related to this.
It's hard to say. I removed the HDR output but then it has become so erratic
that it now crashes almost every time even on a 1-line scene.
G.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Gilles Tran wrote:
>
> It's hard to say. I removed the HDR output but then it has become so erratic
> that it now crashes almost every time even on a 1-line scene.
Even on a freshly started instance of MegaPOV on scenes not using any
patches?
Note if there is a problem with one of the patches it is well possible
that any render started after once rendering a scene using that patch
shows unusual results. To check always try with a newly started MegaPOV
session.
Christoph
--
POV-Ray tutorials, include files, Landscape of the week:
http://www.tu-bs.de/~y0013390/ (Last updated 01 Jul. 2005)
MegaPOV with mechanics simulation: http://megapov.inetart.net/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Sun, 10 Jul 2005 12:28:08 +0200, Christoph Hormann
<chr### [at] gmxde> wrote:
>The source is available - just not yet packaged for Windows. You can
>download the Unix source package and merge it with the 1.1 Windows
>source if you want to work on it.
>
>Christoph
Ok, I was able to get a debug version compiled in VS 2003 using your
suggestions and a modified version of the solution file from POV-Ray
3.6. Now, of course, I'm having difficulty getting any render to
fail!
I tried the suggestion from Gilles Tran further down in the thread,
using hdr_environment.pov with custom settings, but it did not crash
for me, even in the release version. Maybe the environment was
already trashed, from prior crashes, as you suggested. Anyway, if I
make any headway, I'll post back.
FYI, the scenes I have had issues with are rather simple scenes that
use HDRI lighting and are output to a bitmap, in case this helps.
Kyle
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
daub21$qff$1@chho.imagico.de...
> Even on a freshly started instance of MegaPOV on scenes not using any
> patches?
Well, since it crashes each time, isn't launching it again a freshly started
instance (I couldn't see any suspicious thing in the task lists)? Anyway, I
rebooted the computer, rendered my 1-line file (light_source {0 color rgb
1}) with no problem, then loaded a modified hdr_environment.pov with +fh and
now it crashes again every time. More tests are needed, because I don't know
whether it was +fh or the scene file itself that started the problem.
G.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |