yade-dev team mailing list archive

[kan <nicessgg@xxxxxxxxx>]

To sega and Vaclav,

The speed as you stated is good as far as I know.
Could you please let me know what is your hardware. and how many particles
have you used?

Thanks.
Kan

>
> yade-trunk-opt simple-scene.py
> 105485.232068 iterations/sec
>
>
>
What is your hardware? I get to 5e4 iters/sec on my laptop, probably to
double on 2 x dualcore opterons (didn't try). (Sadly, Playstation 3 has

On Wed, Jul 2, 2008 at 5:01 AM, <yade-dev-request@xxxxxxxxxxxxxxxx> wrote:

> Send yade-dev mailing list submissions to
>        yade-dev@xxxxxxxxxxxxxxxx
>
> To subscribe or unsubscribe via the World Wide Web, visit
>        https://lists.berlios.de/mailman/listinfo/yade-dev
> or, via email, send a message with subject or body 'help' to
>        yade-dev-request@xxxxxxxxxxxxxxxx
>
> You can reach the person managing the list at
>        yade-dev-owner@xxxxxxxxxxxxxxxx
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of yade-dev digest..."
>
>
> Today's Topics:
>
>   1. Re: run faster by +16% with r1402 (sega)
>   2. Re: run faster by +16% with r1402 (milauer <eudoxos@xxxxxxxx>)
>   3. Re: [Yade-commits] r1404 - in trunk: extra gui/py gui/qt3
>      pkg/common/Engine/MetaEngine pkg/common/Engine/StandAloneEngine
>      pkg/common/RenderingEngine/GLDrawGeometricalModel
>      pkg/dem/Engine/StandAloneEngine (milauer <eudoxos@xxxxxxxx>)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Tue, 1 Jul 2008 17:59:22 +0400
> From: sega <dorofeenko@xxxxxxxxx>
> Subject: Re: [yade-dev] run faster by +16% with r1402
> To: Yade Development Group <yade-dev@xxxxxxxxxxxxxxxx>
> Message-ID: <200807011759.22775.dorofeenko@xxxxxxxxx>
> Content-Type: text/plain;  charset="iso-8859-1"
>
> Hi!
> I can not understand how to use a parallel engine.
> I have the following results:
>
> export OMP_NUM_THREADS=2; yade-trunk-opt simple-scene-parallel.py
> 36114.1206212 iterations/sec
>
> export OMP_NUM_THREADS=1; yade-trunk-opt simple-scene-parallel.py
> 112233.445567 iterations/sec
>
>
> yade-trunk-opt simple-scene.py
> 105485.232068 iterations/sec
>
>
> --
> Best regards, Sergei D.
>
>
> ------------------------------
>
> Message: 2
> Date: Tue, 01 Jul 2008 20:27:46 +0200
> From: V?clav S(milauer <eudoxos@xxxxxxxx>
> Subject: Re: [yade-dev] run faster by +16% with r1402
> To: Yade Development Group <yade-dev@xxxxxxxxxxxxxxxx>
> Message-ID: <486A7722.8030304@xxxxxxxx>
> Content-Type: text/plain; charset=ISO-8859-1
>
>
>
>
>
> > export OMP_NUM_THREADS=2; yade-trunk-opt simple-scene-parallel.py
> > 36114.1206212 iterations/sec
> >
> > export OMP_NUM_THREADS=1; yade-trunk-opt simple-scene-parallel.py
> > 112233.445567 iterations/sec
> >
> > yade-trunk-opt simple-scene.py
> > 105485.232068 iterations/sec
> >
> >
> >
> What is your hardware? I get to 5e4 iters/sec on my laptop, probably to
> double on 2 x dualcore opterons (didn't try). (Sadly, Playstation 3 has
> max. 256MB of memory, otherwise it would be killer thing to run
> computations on - with 7 cores for massively vectorized floating point...)
>
> openMP defaults to number of threads = number of cores (IIRC).... Second
> and third cases being roughly equal would mean you run on 1 core
> machine? No idea.
>
> There is some overhead of starting the parallel section, which in the
> case of simple-scene probably ouweights the gains of it. I suspect
> however that it might be worth it for larger simulation, where we are at
> few iterations / sec.
>
> Vaclav
>
>
> ------------------------------
>
> Message: 3
> Date: Tue, 01 Jul 2008 20:32:56 +0200
> From: V?clav S(milauer <eudoxos@xxxxxxxx>
> Subject: Re: [yade-dev] [Yade-commits] r1404 - in trunk: extra gui/py
>        gui/qt3 pkg/common/Engine/MetaEngine
>        pkg/common/Engine/StandAloneEngine
>        pkg/common/RenderingEngine/GLDrawGeometricalModel
>        pkg/dem/Engine/StandAloneEngine
> To: Yade Development Group <yade-dev@xxxxxxxxxxxxxxxx>
> Message-ID: <486A7858.6070202@xxxxxxxx>
> Content-Type: text/plain; charset=ISO-8859-1
>
>
> > Fix in svn1409.
> >
> Thanks.
> > sega@linux:~/work/yade/trunk/scripts> yade-trunk-nopermissive
> simple-scene.py
> > ...
> > ERROR gui/qt3/QtGUI.cpp:57 run: Unprocessed non-option argument:
> `simple-scene.py'
> >
> If you don't have boost::python, scons will (silently) build without
> python enabled, hende QtGUI will not be able to process this argument.
> Boost::python uses some code that doesn't compile with the stricter g++
> 4.3, unless you use -fpermissive; that is, I think, why already the
> check for boost::python will fail.
> > Checking for boost::python::scope() in C++ library boost_python-mt... no
> > Checking for boost::python::scope() in C++ library boost_python... no
> >
> You can look in config.log, you will see that the compiler chokes on
> some boost::python header.
> > 1. Viewer size does not change with SimulationPlayer window size.
> >
> The viewer is fixed-size (that is how it was designed and I didn't chage
> it yet).
> > 2. The Primary View of Simulation Controller does not show at first start
> (seems, scene size is not correct).
> >
> This was also fixed in r1409 (did you svn update?), at least for me (the
> secene was realoding spuriously and rootBody didn't have the AABB). Use
> the C (AABB center or average scene center) or Alt-C (median scene
> center) to get the scene right if there is a problem.
>
> V.
>
>
> ------------------------------
>
> _______________________________________________
> yade-dev mailing list
> yade-dev@xxxxxxxxxxxxxxxx
> https://lists.berlios.de/mailman/listinfo/yade-dev
>
>
> End of yade-dev Digest, Vol 20, Issue 2
> ***************************************
>

_______________________________________________
yade-dev mailing list
yade-dev@xxxxxxxxxxxxxxxx
https://lists.berlios.de/mailman/listinfo/yade-dev