Nekonomicon - Blender Irix Development

fzalfa
From avignon ,provence, france
Who joined June 19, 2005, 11:38 a.m.

Wrote the following at Jan. 1, 2007, 11:23 a.m...

thankx for the correction

is it possible to you to build it with more thread number ???
16 if possible or more.....

Laurent

SGI or die !!!
:O2:

HP proliant DL 585 Quad Opteron dual core 2.5Ghz 16Gb

Jarndyce911
From Silicon Valley
Who joined Jan. 15, 2004, 2 p.m.

Wrote the following at Jan. 1, 2007, 12:08 p.m...

fzalfa wrote: thankx for the correction

is it possible to you to build it with more thread number ???
16 if possible or more.....

Laurent

The limitation on the number of threads is built in to the Blender system
so that is not a function of the compilation.

Also as far as the missing scripts files. The .blender file is included in the
main directory, so just copy it to your home directory and all the scripts
you need will be there.

Once you have copied files try the 'q' command again and please let me
know. I can't test it at the moment as I am away from my system, so I
will try it as soon as I can.

Thanks for the continued help!

Charles

Vagabondo
Who joined Sept. 6, 2005, 6:05 p.m.
and authored 337 notes

Wrote the following at Jan. 1, 2007, 3:35 p.m...

I concur...just finished installing the new blender...

Using "test.blend" as instructed on the benchmarking site, but running 8 threads and 8*8 matrix, I achieved a 0:59.77 for a 32p O3800...

Top score on the benchmarking list was 2:01...

This is very promising. I am not aware how difficult it is to increase the number of threads...hope it can be done relatively easilly...would definately like to see what would happen if I could run 64 threads simultaneously!! Not sure where the point of scheduling diminishing returns is, but would definately bring new life to those of us with >8p Iron!!

Kudos for achieving this level of multi-threading. I dont want to appear demanding since this is a great step forward. Thanks to all who have worked to make this possible

Nevermind...looks as though the new years rebound got me: Quad 3GHz (Dual Dual core) with 16GB ram posted 2.XX SECONDS! Oh well...still room for improvement though :wink:

Vagabondo
Who joined Sept. 6, 2005, 6:05 p.m.
and authored 337 notes

Wrote the following at Jan. 1, 2007, 3:38 p.m...

@fzalfa...what file were you using to benchmark your test??

joerg
Who joined Jan. 8, 2004, 6:57 a.m.
and authored 2223 notes

Wrote the following at Jan. 1, 2007, 4:29 p.m...

Vagabondo wrote:

@fzalfa...what file were you using to benchmark your test??

I think the file is from this webside http://www.eofw.org/bench/

regards
Joerg

_________________
http://www.irixworld.net

fzalfa
Who joined June 19, 2005, 10:38 a.m.
and authored 739 notes

Wrote the following at Jan. 2, 2007, 1:49 a.m...

i have post the O2k test on this site......
but i do the Octane2 / o2k comparison with a personal scene.

Laurent

_________________
SGI or die !!!
:O2:

HP proliant DL 585 Quad Opteron dual core 2.5Ghz 16Gb

Tartiflette
Who joined Jan. 2, 2007, 12:32 p.m.
and authored 5 notes

Wrote the following at Jan. 2, 2007, 12:40 p.m...

joerg wrote:

Vagabondo wrote:

@fzalfa...what file were you using to benchmark your test??

I think the file is from this webside http://www.eofw.org/bench/

regards
Joerg

WOW !!

The o2k are very impressive, if you ask me !

If you consider as well that the processors in there are very old ones now, it's even more impressive !
The r10k is about twice as fast as the so called "revolutionnary" CoreDuo on a per Mhz basis !!! :shock:

It's even more depressing now to think where SGI is...
BTW, the fact that Blender still runs on SGI is a good thing !

Regards,
Laurent aka Tartiflette

P.S. : And hello to all the members of the board, it's my first post here !

Vagabondo
Who joined Sept. 6, 2005, 6:05 p.m.
and authored 337 notes

Wrote the following at Jan. 2, 2007, 12:41 p.m...

Well, considering the fact that the O3800 is only running 8 threads when it could easilly handle 4X that at 1/CPU and it came in equivalent to a quad core opteron running at 2.6GHz...not too shabby...plus, I can use it to dry my hair in the morning :wink:

.

Can't wait to see if the 8 threads ability was a proof of principle and can just be extended...and just spawn threads to the OS and have that take care of scheduling and resources.

Welcome Tartiflette!! The more the merrier

Tartiflette
Who joined Jan. 2, 2007, 12:32 p.m.
and authored 5 notes

Wrote the following at Jan. 2, 2007, 1:07 p.m...

Vagabondo wrote:

Well, considering the fact that the O3800 is only running 8 threads when it could easilly handle 4X that at 1/CPU and it came in equivalent to a quad core opteron running at 2.6GHz...not too shabby...plus, I can use it to dry my hair in the morning :wink:

.

Can't wait to see if the 8 threads ability was a proof of principle and can just be extended...and just spawn threads to the OS and have that take care of scheduling and resources.

Welcome Tartiflette!! The more the merrier

From what i've seen, digging in the blender code when making my own build on Mac OS X, you can raise the number of threads in the code as you wish, but i've read that the scaling isn't that good past a certain number of thread...

But i'm not even a programmer so i don't know for sure if that's true or not ? :?:

But it would be cool to see your O3800 running full throttle the bench just to know what to expect from such a beast !

It's good to see that those processor and systems were really so good in the past, and that they can still compete with state of the art computers ! It says a lot about the possibilities back in thos days..

Anyway, i hope that Blender will live for a long time and that it will still be available for the SGI platform.

And thank you for the wlecome ! :wink:

Regards,
Laurent aka Tartiflette

Jarndyce911
From Silicon Valley
Who joined Jan. 15, 2004, 2 p.m.

Wrote the following at Jan. 6, 2007, 10:32 a.m...

Alright, I have answers! First, the reason that 'q' doesn't work anymore
is because it was changed to 'cntrl+q', so give that a try and let me know
if it works for you.

Also, I recompiled and found that I could compile blenderplayer, so you
will notice that the download is now about twice the size.

The third and possibly most exciting part is that someone on the blender
developer list pointed out where to change the number of possible threads
in the code from 8 to whatever I wanted.

Here is the version with a maximum of 1024 threads!

http://www.jarndyce.org/blender-2.43RC1 ... ps.tar.bz2

Please let me know if anything is not working properly.

Thanks

Charles

Jarndyce911
Who joined Jan. 15, 2004, 2 p.m.
and authored 240 notes

Wrote the following at Jan. 6, 2007, 10:34 a.m...

Here is the newly compiled version with a maximum of 1024 threads:

http://www.jarndyce.org/blender-2.43RC1 ... ps.tar.bz2

I also update the other topic, so check it out:

viewtopic.php?t=11942&start=15

Let me know if you have comments or questions!

Thanks

_________________
Charles

joerg
Who joined Jan. 8, 2004, 6:57 a.m.
and authored 2223 notes

Wrote the following at Jan. 6, 2007, 12:28 p.m...

Sorry....but its not working with more than 8 threads. It takes a maximum of 790% cputime.

But i have some questions.

Why it always creates a /tmp/0001.jpg when specify -F PNG?

How to set the tilesets when using the command line only?

Code:

./blender -t 24 -F PNG -g -b test.blend -f 1

On a multi cpu system you can create a cpuset in combination with a simple shell wrapper

Code:

  [o2k]:~/blender-2.43RC1-irix-6.5-mips $ cpuset -q blender -A ./bench.sh 8
  
  guessing './blender' == '/usr/people/beh/blender-2.43RC1-irix-6.5-mips/./blender'
  
  Compiled with Python version 2.4.
  
  Checking for installed Python... got it!
  
  Fra:1 Mem:3.54M Sce: Scene Ve:998 Fa:985 La:1
  
  Fra:1 Mem:23.25M | Part 1-16
  
  Fra:1 Mem:22.82M | Part 5-16
  
  Fra:1 Mem:23.25M | Part 2-16
  
  Fra:1 Mem:23.25M | Part 6-16
  
  Fra:1 Mem:23.19M | Part 4-16
  
  Fra:1 Mem:23.19M | Part 3-16
  
  Fra:1 Mem:23.25M | Part 7-16
  
  Fra:1 Mem:23.25M | Part 9-16
  
  Fra:1 Mem:23.25M | Part 8-16
  
  Fra:1 Mem:21.84M | Part 11-16
  
  Fra:1 Mem:20.42M | Part 10-16
  
  Fra:1 Mem:19.00M | Part 13-16
  
  Fra:1 Mem:17.59M | Part 12-16
  
  Fra:1 Mem:16.11M | Part 14-16
  
  Fra:1 Mem:14.69M | Part 15-16
  
  Fra:1 Mem:13.27M | Part 16-16
  
  Saved: /tmp/0001.jpg Time: 02:12.91

regards
Joerg

_________________
http://www.irixworld.net

Jarndyce911
Who joined Jan. 15, 2004, 2 p.m.
and authored 240 notes

Wrote the following at Jan. 6, 2007, 1:30 p.m...

hmmm, I will have to check into it more tomorrow (computer is shutdown
till tomorrow morning). There must be a setting for it in another place.

As for the command line tile set, I will have to check on that as well.

Thanks for the feedback, I appreciate it.

_________________
Charles

Vagabondo
Who joined Sept. 6, 2005, 6:05 p.m.
and authored 337 notes

Wrote the following at Jan. 6, 2007, 6:55 p.m...

Unfortunately, I can confirm that the max of 8 threads still remains.

Installed the newest compile...It does allow you to SET >8 threads, but does not carry through to the process.

One step in the right direction though.

Thanks Jarndyce for the extra effort :wink:

deBug
Who joined Feb. 27, 2006, 1:44 p.m.
and authored 839 notes

Wrote the following at Jan. 6, 2007, 7:12 p.m...

fzalfa wrote:

hello

i have try the latest Blender provided in the previous topic, support up to 8 cpu is great

i cannot resist to do some benchmark and i observe some strange behaviour in multithread

i use the same test scene between the Octane2 dual R12k and the O2k 8 R10k 250

the first try i do with the O2k was about 1m17s75.....slower than my Octane2.... ??

parametters was, 8 threads, render tile 4x4

at the begining of the rendering all the cpu are loaded, and gradualy as therendering progress, the used cpu number goes lowering. why ??

at the begining the rendering was very fast, and when the most rendering parts are done, the rendering brake seriousely

so i have found a solution, i adjuste the tile render zone as multiple of numbers of cpu, so this is the result with this mods:

8 threads, render tiles 8x8 done in 25s63...... awesome no ?

from to start to the end of the rendering, all the cpu are used....

regards & happy new year

Laurent

I'm I missing something deeper in this?
Because for me it looks obvious what’s going on here.

When there are less than 8 tiles left, some CPUs will be unused as they will not be able to pick any more tiles, right?
So at the end there will always be unused CPUs.
As the tiles completes, more and more CPUs will be unallocated until there is just one CPU doing the last tile.
So the total amount of processing power goes down when there is less than 8 tiles left, right?

So the trick is making enough tiles in the job so that the last 8 tiles is a very small percentage of the total number of tiles, thus affecting the job the least.

In your first example with 4x4 tiles (16) the last 8 tiles is half of the entire job, thus giving big impact on the job.
But in your second example 8x8 tiles (64) the last 8 tiles is just one 1/8 of the entire job, giving less impact on the job.

So your right in that more tiles are better but I don’t believe it has to be a multiple of the number of CPUs, for example, a job with 6x10 tiles are probably as fast as a job with 8x8 tiles.

lhaza
From Milchstrasse, Kakania
Who joined April 20, 2005, 6:35 p.m.

Wrote the following at Jan. 9, 2007, 4:24 p.m...

I still get the same behaviour...

(no filenames, only sizes, and blender cries for icons-dir...)

Code: Select all


   tanger 13% ls
   

   BlenderQuickStart.pdf  blender                copyright.txt
   

   GPL-license.txt        blender.html           plugins
   

   Python-license.txt     blenderplayer          release_242.txt
   

   tanger 14% ./blender
   

   guessing './blender' == '/disc2/nekoware/blender-2.43RC1-irix-6.5-mips/./blender'
   

   Compiled with Python version 2.4.
   

   Checking for installed Python... got it!
   

   /usr/people/m/.blender/icons: No such file or directory
   

   iconv_open Erroriconv_open Erroriconv_open Erroriconv_open Error
   

   Blender quit

ka0s
From The Netherlands
Who joined March 20, 2004, 5:23 a.m.

Wrote the following at Jan. 13, 2007, 2:27 a.m...

Same here.......

I can feel it, my mind is going ....

foetz
Moderator
Who joined April 14, 2003, 4:34 a.m.

Wrote the following at Jan. 13, 2007, 10:39 a.m...

lhaza wrote: I still get the same behaviour... (no filenames, only sizes, and blender cries for icons-dir...)

Code: Select all
tanger 13% ls BlenderQuickStart.pdf blender copyright.txt GPL-license.txt blender.html plugins Python-license.txt blenderplayer release_242.txt tanger 14% ./blender guessing './blender' == '/disc2/nekoware/blender-2.43RC1-irix-6.5-mips/./blender' Compiled with Python version 2.4. Checking for installed Python... got it! /usr/people/m/.blender/icons: No such file or directory iconv_open Erroriconv_open Erroriconv_open Erroriconv_open Error Blender quit

wrong libiconv.
which was used for compiling?

Jarndyce911
From Silicon Valley
Who joined Jan. 15, 2004, 2 p.m.

Wrote the following at Jan. 13, 2007, 4:12 p.m...

I will have to check on the libiconv version. But it should not
matter since everything was compiled in statically except standard
Irix libraries. I will also do an ldd to ensure that is still true, but
if you would also check on your system and let me know that
you don't have any odd dependencies I would appreciate it.

Thanks

Charles

lhaza
From Milchstrasse, Kakania
Who joined April 20, 2005, 6:35 p.m.

Wrote the following at Jan. 15, 2007, 5:32 a.m...

ldd of RC:

Code: Select all


   tanger 62% ldd ./blender
   

   libmalloc.so  =>         /usr/lib32/libmalloc.so
   

   libfastm.so  =>  /usr/lib32/libfastm.so
   

   libm.so  =>      /usr/lib32/libm.so
   

   libGL.so  =>     /usr/lib32/libGL.so
   

   libGLU.so  =>    /usr/lib32/libGLU.so
   

   libmovie.so  =>  /usr/lib32/libmovie.so
   

   libXi.so  =>     /usr/lib32/libXi.so
   

   libXmu.so  =>    /usr/lib32/libXmu.so
   

   libXext.so  =>   /usr/lib32/libXext.so
   

   libX11.so.1  =>  /usr/lib32/libX11.so.1
   

   libc.so.1  =>    /usr/lib32/libc.so.1
   

   libdmedia.so  =>         /usr/lib32/libdmedia.so
   

   libcl.so  =>     /usr/lib32/libcl.so
   

   libaudio.so  =>  /usr/lib32/libaudio.so
   

   libCio.so.1  =>  /usr/lib32/libCio.so.1
   

   libpthread.so  =>        /usr/lib32/libpthread.so
   

   libCsup.so  =>   /usr/lib32/libCsup.so
   

   libC.so.2  =>    /usr/lib32/libC.so.2
   

   libGLcore.so  =>         /usr/lib32/libGLcore.so
   

   libXsgivc.so  =>         /usr/lib32/libXsgivc.so
   

   libmovieplay.so  =>      /usr/lib32/libmovieplay.so
   

   libmoviefile.so  =>      /usr/lib32/libmoviefile.so
   

   libXt.so  =>     /usr/lib32/libXt.so
   

   libgen.so  =>    /usr/lib32/libgen.so
   

   libmutex.so  =>  /usr/lib32/libmutex.so
   

   libawareaudio.so  =>     /usr/lib32/libawareaudio.so

ldd of neko_blender2.42a

Code: Select all


   tanger 65% ldd /usr/nekoware/bin/blender
   

   libmalloc.so  =>         /usr/lib32/libmalloc.so
   

   libfastm.so  =>  /usr/lib32/libfastm.so
   

   libm.so  =>      /usr/lib32/libm.so
   

   libIlmImf.so.3  =>       /usr/nekoware/lib/libIlmImf.so.3
   

   libImath.so.3  =>        /usr/nekoware/lib/libImath.so.3
   

   libHalf.so.3  =>         /usr/nekoware/lib/libHalf.so.3
   

   libIex.so.3  =>  /usr/nekoware/lib/libIex.so.3
   

   libavformat.so  =>       /usr/nekoware/lib/libavformat.so
   

   libavcodec.so  =>        /usr/nekoware/lib/libavcodec.so
   

   libGL.so  =>     /usr/lib32/libGL.so
   

   libGLU.so  =>    /usr/lib32/libGLU.so
   

   libmovie.so  =>  /usr/lib32/libmovie.so
   

   libXmu.so  =>    /usr/lib32/libXmu.so
   

   libXext.so  =>   /usr/lib32/libXext.so
   

   libX11.so.1  =>  /usr/lib32/libX11.so.1
   

   libc.so.1  =>    /usr/lib32/libc.so.1
   

   libdmedia.so  =>         /usr/lib32/libdmedia.so
   

   libcl.so  =>     /usr/lib32/libcl.so
   

   libaudio.so  =>  /usr/lib32/libaudio.so
   

   libCio.so.1  =>  /usr/lib32/libCio.so.1
   

   libpthread.so  =>        /usr/lib32/libpthread.so
   

   libCsup.so  =>   /usr/lib32/libCsup.so
   

   libC.so.2  =>    /usr/lib32/libC.so.2
   

   libavutil.so.49  =>      /usr/nekoware/lib/libavutil.so.49
   

   libz.so  =>      /usr/nekoware/lib/libz.so
   

   libmp3lame.so.1  =>      /usr/nekoware/lib/libmp3lame.so.1
   

   libvorbis.so.1  =>       /usr/nekoware/lib/libvorbis.so.1
   

   libvorbisenc.so.3  =>    /usr/nekoware/lib/libvorbisenc.so.3
   

   libogg.so.1  =>  /usr/nekoware/lib/libogg.so.1
   

   libx264.so.54  =>        /usr/nekoware/lib/libx264.so.54
   

   libfaac.so.1  =>         /usr/nekoware/lib/libfaac.so.1
   

   libfaad.so.1  =>         /usr/nekoware/lib/libfaad.so.1
   

   libGLcore.so  =>         /usr/lib32/libGLcore.so
   

   libXsgivc.so  =>         /usr/lib32/libXsgivc.so
   

   libmovieplay.so  =>      /usr/lib32/libmovieplay.so
   

   libmoviefile.so  =>      /usr/lib32/libmoviefile.so
   

   libXt.so  =>     /usr/lib32/libXt.so
   

   libgen.so  =>    /usr/lib32/libgen.so
   

   libmutex.so  =>  /usr/lib32/libmutex.so
   

   libawareaudio.so  =>     /usr/lib32/libawareaudio.so

I tried with the sequenceeditor and loaded pond.dv (DV-sample from libdv see http://libdv.sourceforge.net and http://prdownloads.sourceforge.net/libd ... v?download for the file(50 mb)).

with 2.42a:
SHIFT-A->Import Movie
coredump
SHIFT-A->Import Movie + Audio (HD)

Code: Select all


   Input #0, dv, from '/disc2/pond.dv':
   

   Duration: 00:00:30.5, start: 0.000000, bitrate: 28771 kb/s
   

   Stream #0.0: Video: dvvideo, yuv411p, 720x480, 28771 kb/s, 29.97 fps(r)
   

   Stream #0.1: Audio: pcm_s16le, 44100 Hz, stereo, 1411 kb/s

The video is imported (no audio track seen, but audiodata mentioned on console).

with RC1, both imports seem to work (both tracks created), but the preview in the upper right corner stays black and <F12> (render) also (with 2.42a and sequence checked in <F10> the current frame is rendered).

SGI: Computer Graphics

Blender Irix Development - Page 2