Hi,
I am using FS 4.01. I have processed 400 scans with autorecon-all. I am trying to analyse ~350 of them in qdec. I have done the preprocessing (recon-all -qcache -measure thickness) and prepared the required files (attached).
I can start qdec and conduct an analysis. I can also visualise some results but invariably after moving between the contrasts a seg fault occurs (logs attached, I was going to attach 2 but one of them was 12MB in size, does this sound like a normal output length?).
I have a dual quad core Xeon 5355 with 8 Gig of RAM and plenty of disk space.
I don't think it makes any difference but just in case... I computed my own average. Make_average_surface went ok but make_average_volume was killed because it required too much resources. I figured I did not require the volume for the qdec analysis so I used my average (older people) for -qcache smoothing. However, when I first tried to visualise the results qdec did not like the fact that there was no curvature information (I was going to attach a log file but it was 6MB in size) in my average subject so I linked to fsaverage.
Any help is welcome.
Cheers
Nic
Nic,
Are you using the 64bit freesurfer? Try setting this:
setenv QDEC_DONT_USE_HISTOGRAM_EDITOR 1
I have encountered a segfault problem myself on 64bit builds of freesurfer, where some VTK/KWWidgets code is the problem (but I have never been able to find the source of the problem).
Let me know if this 'fixes' the problem, as i may just disable the histogram display permanently until the problem is fixed.
Nick
On Wed, 2008-03-05 at 14:59 +1100, Nicolas Cherbuin wrote:
Hi,
I am using FS 4.01. I have processed 400 scans with autorecon-all. I am trying to analyse ~350 of them in qdec. I have done the preprocessing (recon-all -qcache -measure thickness) and prepared the required files (attached).
I can start qdec and conduct an analysis. I can also visualise some results but invariably after moving between the contrasts a seg fault occurs (logs attached, I was going to attach 2 but one of them was 12MB in size, does this sound like a normal output length?).
I have a dual quad core Xeon 5355 with 8 Gig of RAM and plenty of disk space.
I don't think it makes any difference but just in case... I computed my own average. Make_average_surface went ok but make_average_volume was killed because it required too much resources. I figured I did not require the volume for the qdec analysis so I used my average (older people) for -qcache smoothing. However, when I first tried to visualise the results qdec did not like the fact that there was no curvature information (I was going to attach a log file but it was 6MB in size) in my average subject so I linked to fsaverage.
Any help is welcome.
Cheers
Nic
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Hi Nick,
Yes, I am using the 64 bit fs. I will try your fix and keep you updated. Thanks very much for your help.
Regards,
Nic
-----Original Message----- From: Nick Schmansky [mailto:nicks@nmr.mgh.harvard.edu] Sent: Thu 3/6/2008 3:17 AM To: Nicolas Cherbuin Cc: Freesurfer Mailing List Subject: Re: [Freesurfer] FW: qdec segmentation faults
Nic,
Are you using the 64bit freesurfer? Try setting this:
setenv QDEC_DONT_USE_HISTOGRAM_EDITOR 1
I have encountered a segfault problem myself on 64bit builds of freesurfer, where some VTK/KWWidgets code is the problem (but I have never been able to find the source of the problem).
Let me know if this 'fixes' the problem, as i may just disable the histogram display permanently until the problem is fixed.
Nick
On Wed, 2008-03-05 at 14:59 +1100, Nicolas Cherbuin wrote:
Hi,
I am using FS 4.01. I have processed 400 scans with autorecon-all. I am trying to analyse ~350 of them in qdec. I have done the preprocessing (recon-all -qcache -measure thickness) and prepared the required files (attached).
I can start qdec and conduct an analysis. I can also visualise some results but invariably after moving between the contrasts a seg fault occurs (logs attached, I was going to attach 2 but one of them was 12MB in size, does this sound like a normal output length?).
I have a dual quad core Xeon 5355 with 8 Gig of RAM and plenty of disk space.
I don't think it makes any difference but just in case... I computed my own average. Make_average_surface went ok but make_average_volume was killed because it required too much resources. I figured I did not require the volume for the qdec analysis so I used my average (older people) for -qcache smoothing. However, when I first tried to visualise the results qdec did not like the fact that there was no curvature information (I was going to attach a log file but it was 6MB in size) in my average subject so I linked to fsaverage.
Any help is welcome.
Cheers
Nic
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
freesurfer@nmr.mgh.harvard.edu
-
Nick Schmansky -
Nicolas Cherbuin