Tyler, thanks for this. I think I found why probtrackx2 takes up more RAM in this context, but I need some time before I can fix it.
For the time being, you can perhaps use the trick I mentioned a couple of emails ago? Will try to fix this memory issue very soon.
Cheers
Saad
On 4 May 2012, at 17:52, Tyler Blazey wrote:
> Hi Saad,
>
> Thanks for responding to another one of my questions. The high memory requirement occurs whether my targets are volumes or surfaces. In case it helps, the commands I used to run the volume seeds/targets are below:
>
> probtrackx --samples=../dti.bedpostX/merged --mask=../dti.bedpostX/nodif_brain_mask --seed=vol_targets/ctx-rh-insula --dir=ctx-rh-insula_1 --forcedir --opd --os2t --s2tastext --targetmasks=vol_targets.txt --stop=subj_stop.nii.gz --xfm=../atlas/subj_free_to_dti.mat --seedref=../atlas/subj_T1.nii.gz --loopcheck --cthr=0.2 --nsteps=2000 --nsamples=5000
>
> probtrackx2 --samples=../dti.bedpostX/merged --mask=../dti.bedpostX/nodif_brain_mask --seed=vol_targets/ctx-rh-insula --dir=ctx-rh-insula_2 --forcedir --opd --os2t --s2tastext --targetmasks=vol_targets.txt --stop=subj_stop.nii.gz --xfm=../atlas/subj_free_to_dti.mat --seedref=../atlas/subj_T1.nii.gz --loopcheck --cthr=0.2 --nsteps=2000 --nsamples=5000
>
> Is there a mistaken in my probtrackx2 command line that could be causing the high memory usage?
>
> -Tyler
>
--
Saad Jbabdi
University of Oxford, FMRIB Centre
JR Hospital, Headington, OX3 9DU, UK
(+44)1865-222466 (fax 717)
www.fmrib.ox.ac.uk/~saad
|