When doing an animation with the new MegaPOV Plus, I sometimes get an "Out
of memory" error. It seems to happen at random with something like 10% of
the frames, and these frames are no different from the frames that doesn't
produce errors.

One strange thing is that in the error messages =, !=, and possible other
characters are always replaced with square characters (). Also, the errors
seem to be always have something to do with the = character. Could this be
related to the #set patch (which I did not use in my scene file)?

By the way, I'm using the windows version MegaPov 0.5a Plus mod 0.3.0.

Greetings,

Rune
--
\ Include files, tutorials, 3D images, raytracing jokes,
/ The POV Desktop Theme, and The POV-Ray Logo Contest can
\ all be found at http://rsj.mobilixnet.dk (updated July 23)
/ Also visit http://www.povrayusers.org

Post a reply to this message

In article <39b57a67@news.povray.org>, "Rune" <run### [at] inamecom> 
wrote:

> When doing an animation with the new MegaPOV Plus, I sometimes get an 
> "Out of memory" error. It seems to happen at random with something 
> like 10% of the frames, and these frames are no different from the 
> frames that doesn't produce errors.

Probably a memory leak, if you start over from the frame that caused the 
trouble, things should work fine(until it runs out of memory again). I 
wish I had some software for tracking down memory leaks...


> One strange thing is that in the error messages =, !=, and possible 
> other characters are always replaced with square characters (). 
> Also, the errors seem to be always have something to do with the = 
> character. Could this be related to the #set patch (which I did not 
> use in my scene file)?

This is very strange...I don't see how it could be related to the set 
patch...unless my experiments with getting the "# VarName" syntax to 
work got in there by accident. I will check, this is the most likely 
explanation for what is going wrong.

-- 
Christopher James Huff
Personal: chr### [at] maccom, http://homepage.mac.com/chrishuff/
TAG: chr### [at] tagpovrayorg, http://tag.povray.org/

<><

Post a reply to this message

In article <chrishuff-C86F26.19092605092000@news.povray.org>, Chris 
Huff <chr### [at] maccom> wrote:

> This is very strange...I don't see how it could be related to the set 
> patch...unless my experiments with getting the "# VarName" syntax to 
> work got in there by accident. I will check, this is the most likely 
> explanation for what is going wrong.

Sigh...
I did leave in some code. I can't even figure out why it is isn't 
working, let alone why it is causing those strange effects you mentioned.
I will have version 0.3.1 done soon, with a slightly different syntax 
for the glow effect and this code fixed.
Is this a severe problem? I mean, does it actually cause errors, rather 
than just messing up the reporting of errors? If so, I will go ahead and 
post the affected file(tokenize.c).

-- 
Christopher James Huff
Personal: chr### [at] maccom, http://homepage.mac.com/chrishuff/
TAG: chr### [at] tagpovrayorg, http://tag.povray.org/

<><

Post a reply to this message

Chris Huff <chr### [at] maccom> wrote:
: Probably a memory leak

  I think that someone has said that there's a memory leak in at least 99%
of the programs larger than 10000 lines of code.

  That's one of the reasons why you have to be booting windows from time
to time (and in this case the guilty is not necessarily windows itself).

-- 
main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
):_;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/

Post a reply to this message

In article <39b603a7@news.povray.org>, Warp <war### [at] tagpovrayorg> 
wrote:

>   I think that someone has said that there's a memory leak in at least 99%
> of the programs larger than 10000 lines of code.

Quite true, at least for programs written in the most common languages, 
like C or C++. I suspect there are a couple leaks in the expression 
parsing functions, and maybe two which I added by mistake with the 
turbulence scale patch, but I seem to have been getting leaks since my 
first compile of MegaPOV 0.5a, so it isn't only my code at fault. There 
are probably dozens of small leaks scattered all over the source code.

-- 
Christopher James Huff
Personal: chr### [at] maccom, http://homepage.mac.com/chrishuff/
TAG: chr### [at] tagpovrayorg, http://tag.povray.org/

<><

Post a reply to this message

Chris Huff <chr### [at] maccom> wrote:
: Quite true, at least for programs written in the most common languages, 
: like C or C++.

  In C++ the problem is not so usual because the need for new and delete
has decreased a lot with the introduction of the STL.
  I usually don't use new and delete at all (only when STL really doesn't
do what I want).

-- 
main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
):_;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/

Post a reply to this message

"Chris Huff" wrote:
> Probably a memory leak,

I don't know exactly what a memory leak is, but I can tell you some more
about how MP+ behaved in my case.

Well, I have this animation that doesn't use any MP+ features at all. When I
render it in regular MP everything is fine and the peak memory used is about
the same for every frame.
But when I render the same animation in MP+ the peak memory used increases
rapidly for every frame, like they are added together. This makes it
practically impossible to render a whole animation.

> if you start over from the frame that caused the trouble, things
> should work fine(until it runs out of memory again).

I don't want to start over 20 times to render a short animation.

> > One strange thing is that in the error messages =, !=, and
> > possible other characters are always replaced with square
> > characters ().
>
> Is this a severe problem? I mean, does it actually cause errors,
> rather than just messing up the reporting of errors?

I don't know if they are directly related. Maybe not.

Rune
--
\ Include files, tutorials, 3D images, raytracing jokes,
/ The POV Desktop Theme, and The POV-Ray Logo Contest can
\ all be found at http://rsj.mobilixnet.dk (updated July 23)
/ Also visit http://www.povrayusers.org

Post a reply to this message

In article <chrishuff-875123.08334206092000@news.povray.org> , Chris Huff 
<chr### [at] maccom>  wrote:

> Quite true, at least for programs written in the most common languages,
> like C or C++. I suspect there are a couple leaks in the expression
> parsing functions, and maybe two which I added by mistake with the
> turbulence scale patch, but I seem to have been getting leaks since my
> first compile of MegaPOV 0.5a, so it isn't only my code at fault. There
> are probably dozens of small leaks scattered all over the source code.

POV-Ray has a memory debugging functionality as a compile-time option.  Look
for it in mem.c.  Enabling it should create a log and further it should be
able to tell you which blocks are not freed after rendering.


       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

In article <39b6503f@news.povray.org>, Warp <war### [at] tagpovrayorg> 
wrote:

>   In C++ the problem is not so usual because the need for new and delete
> has decreased a lot with the introduction of the STL.
>   I usually don't use new and delete at all (only when STL really doesn't
> do what I want).

Hmm, I never use the STL any more...it only caused me trouble.

-- 
Christopher James Huff
Personal: chr### [at] maccom, http://homepage.mac.com/chrishuff/
TAG: chr### [at] tagpovrayorg, http://tag.povray.org/

<><

Post a reply to this message

Chris Huff <chr### [at] maccom> wrote:
: Hmm, I never use the STL any more...it only caused me trouble.

  Strange. I don't use new anymore (except only in very few cases) since
the STL is so handy.

  Why would I code dynamic vectors, trees and so on myself when someone
much more expert than me has done it already? And besides, STL is a lot
easier to use :)

-- 
main(i,_){for(_?--i,main(i+2,"FhhQHFIJD|FQTITFN]zRFHhhTBFHhhTBFysdB"[i]
):_;i&&_>1;printf("%s",_-70?_&1?"[]":" ":(_=0,"\n")),_/=2);} /*- Warp -*/

Post a reply to this message