Bruce,

http://surfer.nmr.mgh.harvard.edu/fswiki/CUDADevelopersGuide#Benchmarks Found this link..clearly impressive, are these benchmarks for GPU enabled cards with CUDA or also with ATI Radeon GPU enabled cards?

Best, Cartik

On Oct 24, 2011, at 10:30 AM, Bruce Fischl wrote:

not yet, although it is something we are working on.

cheers Bruce On Mon, 24 Oct 2011, Cartik Sharma wrote:

...

Hi Knut, Good to know we can speed things up with CUDA. I have a multicore CPU and was wondering if there is a parallelized version of Freesurfer on the CPU. Thanks for the speedup! Best regards, Cartik On Oct 24, 2011, at 4:41 AM, Knut J Bjuland wrote:

 Hi Cartik

 You use -use-gpu to offload rendering to the GPU using cuda. 

 Knut J

 > From: Cartik.Sharma@childmind.org
 > To: knutjbj@hotmail.com
 > Subject: Re: [Freesurfer] implementation on Neurodebian
 > Date: Thu, 20 Oct 2011 15:48:10 +0000
 > 
 > Hi Knut:
 > 
 > Cool..is this speeding up things by rendering on the GPU.
 > Is there a way to offload computation to the GPU.
 > 
 > Best regards,
 > Cartik
 > 
 > On Oct 20, 2011, at 10:57 AM, Knut J Bjuland wrote:
 > 
 > > Hi Cartik
 > > 
 > > If you enable cuda with -use-gpu the recon-all will run faster.
 > > 
 > > Knut J
 > > 
 > > Den 17.10.2011 16:41, skrev Cartik Sharma:
 > >> Dear Freesurfer developer(s),
 > >> 
 > >> Is there a way to get Freesurfer run faster on Neurodebian. While the package produces
 > >> some truly out of the world segmentations and labels for the cortical surface, it seems
 > >> to take time.
 > >> 
 > >> For eg:
 > >> Defect correction and retesellation takes an hour, other steps take longer..I'm not sure if
 > >> it's the original dataset, the OS platform (i.e Virtual Machine, Neurodebian) or original
 > >> computational engine itself.
 > >> 
 > >> Would be happy in efforts and hints to make this run faster.
 > >> 
 > >> Best regards,
 > >> Cartik Sharma
 > >> 
 > >> 
 > >> _______________________________________________
 > >> Freesurfer mailing list
 > >> Freesurfer@nmr.mgh.harvard.edu
 > >> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
 > >> 
 > >> 
 > >> The information in this e-mail is intended only for the person to whom it is
 > >> addressed. If you believe this e-mail was sent to you in error and the e-mail
 > >> contains patient information, please contact the Partners Compliance HelpLine at
 > >> http://www.partners.org/complianceline . If the e-mail was sent to you in error
 > >> but does not contain patient information, please contact the sender and properly
 > >> dispose of the e-mail.
 > >> 
 > >> 
 > >> 
 > > 
 > 

just CUDA for now.

Bruce On Mon, 24 Oct 2011, Cartik Sharma wrote:

Bruce,

http://surfer.nmr.mgh.harvard.edu/fswiki/CUDADevelopersGuide#Benchmarks Found this link..clearly impressive, are these benchmarks for GPU enabled cards with CUDA or also with ATI Radeon GPU enabled cards?

Best, Cartik

On Oct 24, 2011, at 10:30 AM, Bruce Fischl wrote:

...

not yet, although it is something we are working on.

cheers Bruce On Mon, 24 Oct 2011, Cartik Sharma wrote:

...

Hi Knut, Good to know we can speed things up with CUDA. I have a multicore CPU and was wondering if there is a parallelized version of Freesurfer on the CPU. Thanks for the speedup! Best regards, Cartik On Oct 24, 2011, at 4:41 AM, Knut J Bjuland wrote:

 Hi Cartik

 You use -use-gpu to offload rendering to the GPU using cuda.

 Knut J

> From: Cartik.Sharma@childmind.org
> To: knutjbj@hotmail.com
> Subject: Re: [Freesurfer] implementation on Neurodebian
> Date: Thu, 20 Oct 2011 15:48:10 +0000
>
> Hi Knut:
>
> Cool..is this speeding up things by rendering on the GPU.
> Is there a way to offload computation to the GPU.
>
> Best regards,
> Cartik
>
> On Oct 20, 2011, at 10:57 AM, Knut J Bjuland wrote:
>
>> Hi Cartik
>>
>> If you enable cuda with -use-gpu the recon-all will run faster.
>>
>> Knut J
>>
>> Den 17.10.2011 16:41, skrev Cartik Sharma:
>>> Dear Freesurfer developer(s),
>>>
>>> Is there a way to get Freesurfer run faster on Neurodebian. While the package produces
>>> some truly out of the world segmentations and labels for the cortical surface, it seems
>>> to take time.
>>>
>>> For eg:
>>> Defect correction and retesellation takes an hour, other steps take longer..I'm not sure if
>>> it's the original dataset, the OS platform (i.e Virtual Machine, Neurodebian) or original
>>> computational engine itself.
>>>
>>> Would be happy in efforts and hints to make this run faster.
>>>
>>> Best regards,
>>> Cartik Sharma
>>>
>>>
>>> _______________________________________________
>>> Freesurfer mailing list
>>> Freesurfer@nmr.mgh.harvard.edu
>>> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
>>>
>>>
>>> The information in this e-mail is intended only for the person to whom it is
>>> addressed. If you believe this e-mail was sent to you in error and the e-mail
>>> contains patient information, please contact the Partners Compliance HelpLine at
>>> http://www.partners.org/complianceline . If the e-mail was sent to you in error
>>> but does not contain patient information, please contact the sender and properly
>>> dispose of the e-mail.
>>>
>>>
>>>
>>
>