Hi Andrade the computational complexity of the topology fixing is square in the number of vertices in the largest defect (actually in the convex hull of it). That means that subjects with small defects are process very rapidly, but large ones can take a long time. Typically if you have a very large defect it means that something went significantly wrong with the processing and you are better off stopping recon-all and finding out what and fixing it. Things like the skull being attached and such. Look at the ?h.orig.nofix and ?h.inflated.nofix surfaces to find the defects. There is a tutorial on our website that covers this.
cheers Bruce On Sat, 24 Nov 2018, Renew Andrade wrote: > External Email - Use Caution > > Dear Community! > I am having trouble processing 2 subjects, twice, it stops in CORRECTING > DEFECTS 1 and 2 respectively. Watching top output shows CPU is still trying > to process the subject. The other day I processed successfully another > subject from the same database. > > Sincerely, > Andrade. > > > > > _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
