Greetings,
I've been asked to provide some extra information about GPU support in
Freesurfer (being the one guilty of mri_em_register_cuda...).
Firstly, there are no immediate plans for OpenCL support. It would be
very nice to have - with ATI, NVIDIA _and_ x86 multicore backends.
However, it's fa
On Tue, 2010-08-24 at 10:55 -0300, Pedro Paulo de Magalhães Oliveira
Junior wrote:
> Just to clarify one point: when you mentioned CUDA architecture 1.3 I
> think you did mean Cuda Computing Capability 1.3
Indeed - it's the -arch flag to the compiler, hence I got mixed up.
We haven't tested this
On Tue, 2010-08-24 at 09:37 -0400, Richard G. Edgar wrote:
> One final thing: Nick and I found last week that the accelerated
> mri_em_register_cuda doesn't seem to work prior to skull stripping. I'm
> going to work on this this week, but if you want to continue using the
>
On Thu, 2010-08-26 at 23:23 +0200, Georg Homola wrote:
> allow me one additional remark that may be crucial for those considering to
> invest in new cards. Although the Fermi class cards make use of the same
> architecture (Geforce GTX 480 and Tesla C2050 for example), for consumer
> products (GTX
On Fri, 2010-08-27 at 17:37 -0500, Freesurfer Local Archive wrote:
> I am looking at choice of the lower end fermi cards and I was
> wondering if you could comment of Freesurfer's ability to take
> advantage of some seldom used functions included in nVidia GTX 400
> (including tesselation) that w
Greetings,
I've been running more extensive tests on mri_em_register_cuda, trying
to determine why it sometimes gives substantially different results to
mri_em_register. The problem appears to stem from the construction of
the transform matrix on the CPU and GPU. This occurs slightly
differently o
On Wed, 2010-10-20 at 16:18 +0200, Jordi Delgado wrote:
> *** glibc detected *** mri_em_register: malloc(): memory corruption:
> 0x315dfdc0 ***
> === Backtrace: =
> /lib64/libc.so.6[0x3304072fae]
> /lib64/libc.so.6(__libc_calloc+0xcd)[0x330407495d]
> mri_em_register[0x42efad]
On Tue, 2010-11-23 at 15:03 +0100, Marcus N Schmidt wrote:
> I should note that we have a single GeForce GTX 470 card on the
> workstation and have installed the latest CUDA drivers and toolkit
> (3.2).
>
> >
> > Is there a limit to how many processes can use CUDA at the same
> > time? I just set
On Mon, 2010-11-22 at 15:27 -0200, Pedro Paulo de Magalhães Oliveira
Junior wrote:
> recon-all -s bert -use-gpu -all finished in 7hours and 15 minutes.
>
> It is a useful solution when you are in a hurry to finish some cases. It's a
> bit pricey: US$ 2.10 / hour
Right now, on the volume side of
On Tue, 2010-11-23 at 12:31 -0200, Pedro Paulo de Magalhães Oliveira
Junior wrote:
> In FreeSurfer 5.1 can we expect to have CUDA support on OSX?
I haven't tested any of the CUDA code under OSX yet, so I don't have a
timeline.
Regards,
Richard
___
Fr
On Tue, 2011-03-08 at 13:58 +0100, Marcus N Schmidt wrote:
> Just a quick question: Generally speaking, assuming everything else
> remains the same, are subjects run through a CUDA-enabled recon-all
> comparable to subjects run through recon-all without CUDA?
There are differences, which are ine
On Thu, 2011-03-10 at 08:27 -0500, Bruce Fischl wrote:
> If you have enough memory! Maybe Richard can post current CPU recon
> times with the latest hardware and optimizations. I think they are
> much shorter than they used to be
My workstation (3.2 GHz Nehalem) can do a recon-all ernie in about
On Thu, 2011-03-10 at 11:29 -0300, Pedro Paulo de Magalhães Oliveira
Junior wrote:
> That's without GPU?
Yes. CPU only stream.
Richard
___
Freesurfer mailing list
Freesurfer@nmr.mgh.harvard.edu
https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesur
On Fri, 2011-03-11 at 09:11 -0700, Rojas, Don wrote:
> I'm having a problem similar to those that some others have posted
> previously on the list, but I did not see a clear resolution posted
> and I have tried the suggestions as far as I could follow them. The
> error message is:
>
> Acquiring
On Mon, 2011-04-04 at 13:13 +0200, Knut J Bjuland wrote:
> I am in the process of acquiring a new computer to run freesurfer on.
> I am currently think about buying a PC with a geforce 4X0 and a Tesla
> card along with a two screen setup running either Ubuntu or Redhat
> Linux enterprise 6.0. I t
On Tue, 2011-04-05 at 11:23 +0200, Knut J Bjuland wrote:
> Will a Tesla C2050 or another good CPU be able to reduce the running
> time from 20-24 hr to less time like for instance 8hr or below that
> time.
On the standard test case we use here, a full recon-all run takes 8
hours on a 3.2 GHz Neha
On Tue, 2011-04-05 at 16:30 +0100, Ian Malone wrote:
> Richard G. Edgar wrote:
> > On Tue, 2011-04-05 at 11:23 +0200, Knut J Bjuland wrote:
> >
> > > Will a Tesla C2050 or another good CPU be able to reduce the running
> > > time from 20-24 hr to less time li
On Thu, 2011-04-21 at 09:48 -0700, Rongxiang Tang wrote:
>
> I was wondering if I can run several freesurfers in my virtualbox
> simultaneously by setting up several virtual machines...my computer
> has 16G RAM and it's Windows 7 64-bit.
> Would working multiple freesurfers at the same time slow
On Tue, 2011-04-26 at 18:30 +0200, Knut J Bjuland wrote:
> It appears that fressurfer 5.0 is compiled against cuda 3.2.x rather
> than cuda 4. Is there any workaround to force freesurfer 5.0 apart
> from recompilation for it use cuda 4.0. AS the latest version of the
> nvidia driver has change cud
On Tue, 2011-05-03 at 18:05 +0200, Knut J Bjuland wrote:
> I have also installed cuda 3.2 with the same results. There might be an
> error in CUDA Error in file 'devicemanagement.cu' on line 46 : invalid
> argument. which I got when I used the latest nvidia driver with cuda 3.2
> Could anybody
On Tue, 2011-05-03 at 18:08 +0200, Knut J Bjuland wrote:
> is ther any plans to produce an version of freesurfer for openmpi and/or
> cuda 4.0? Do Anders Dale produce an version of freesurfer for openmi
> interface that enable use of freesurfer on supercomputer?
I would not expect a migration
21 matches
Mail list logo
