Glenn,
There is not a problem mixing subjects processed with the 32bit version and 64bit version, nor between OS type.
The variance of Freesurfer results resulting from the randomness of some of the algorithms (about 0.01%) is actually greater than the variance of results between different addressing-widths and OS types (about 0.002%).
Nick
On Wed, 2006-12-13 at 12:32 +0100, Glenn Lawyer wrote:
Hi,
Is there any problem mixing subjects processed with the 32 bit version and the 64 bit version, using FS 3.? How about between the RH9 build and the Centos build?
I would be very surprised if this did present a problem, but my supervisor wants me to check.
Yes, I did see Nick's email (28 mar 2006) on the slight differences in the floating point optimizations, which he felt were minor.
+glenn
I'm only in it for the glory.
Glenn Lawyer < +352 661 967 244 < Instituttgruppe for psykiatri < Postboks 1130 Blindern < 0318 Oslo <
< http://folk.uio.no/davidgl > <><><><><><><><><><><><><><><><><:) _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer