Dear Freesurfer users and developers,
I apologize if this question sounds rhetorically or maybe it has been asked previously (I didn't find the answers in the archives). But here it is... I have a big dataset of images and several workstations. I started the analysis on the first one and afterwards I installed Freesurfer on the others. But just right now I have noticed that their versions are different (I have v5.0 on the first computer, whereas other ones have 5.1). Could you please tell me if I should redo all the analyses from the first workstation or if it is ok and I can upgrade my version and keep it going? Thank you very much beforehand! --- Best Regards, Alex
Hi Alex,
sorry but you should never mix versions.
cheers Bruce On Mon, 5 Sep 2011, Alexander Lebedev wrote:
Dear Freesurfer users and developers,
I apologize if this question sounds rhetorically or maybe it has been asked previously (I didn't find the answers in the archives). But here it is... I have a big dataset of images and several workstations. I started the analysis on the first one and afterwards I installed Freesurfer on the others. But just right now I have noticed that their versions are different (I have v5.0 on the first computer, whereas other ones have 5.1). Could you please tell me if I should redo all the analyses from the first workstation or if it is ok and I can upgrade my version and keep it going? Thank you very much beforehand!
Best Regards, Alex _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Thank you very much for your response!
May I also ask how about running CUDA and non-CUDA analyses for the same dataset? I am asking because I don't have CUDA support on one of my machines as well (It has ATI videocard). I have found a thread on that topic (http://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg16832.html) and even asked the author - and the question still hangs... What do you think? --- Best Regards, Alex
I wouldn't do that either. We have actively worked to minimize the difference between CUDA and non-CUDA versions, but it's not possible to make them identical due to differences in architecture
sorry again Bruce
On Tue, 6 Sep 2011, Alexander Lebedev wrote:
Thank you very much for your response!
May I also ask how about running CUDA and non-CUDA analyses for the same dataset? I am asking because I don't have CUDA support on one of my machines as well (It has ATI videocard). I have found a thread on that topic (http://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg16832.html) and even asked the author - and the question still hangs... What do you think?
Best Regards, Alex
On Tue, Sep 6, 2011 at 1:25 PM, Alexander Lebedev awlman@rambler.ru wrote:
Thank you very much for your response!
May I also ask how about running CUDA and non-CUDA analyses for the same dataset? I am asking because I don't have CUDA support on one of my machines as well (It has ATI videocard). I have found a thread on that topic (http://www.mail-archive.com/freesurfer@nmr.mgh.harvard.edu/msg16832.html) and even asked the author - and the question still hangs... What do you think?
If you require bit-for-bit agreement, then you basically can't change _anything_ about the computer configuration. It is the nature of floating point arithmetic that a+(b+c) != (a+b)+c There's nothing we can do about this - it's how computers work.
Now, the results still ought to agree in some statistical sense, but I never got that far with verifying it across a variety of brains and platforms.
Richard
freesurfer@nmr.mgh.harvard.edu
-
Alexander Lebedev -
Bruce Fischl -
R Edgar