Thanks!

That's all I wanted to hear :). We have tried to dabble with ETAC but found it still to be very conservative. But that was in the early days, perhaps time to give it a new go! =)


Something I noticed as a bonus:
Not sure this was from @SSwarper of afniproc since my script is running them in sequence but one of those processes created temp-files in the directory where I run my main bash-script (which makes me believe it's the @SSwarper since afni-proc cd's into the .results folder):

__eac_tmp_0_cp.nii __eac_tmp_1_ulay_shrp.nii __eac_tmp_2_res.nii __eac_tmp_3_mask.nii __eac_tmp_4_mfilt.nii __eac_tmp_5_edgy.nii __eac_tmp_6_edge_mskd.nii

Running subjects in parallel on the same machine or from another server in the same network mount could (with very unlucky timing since they are short-lived) cause a collision of file names (since they don't seem to have a unique identifier) and perhaps result in fatal error or read the wrong files. Unless your @SSwarper is super smart and checks for files prior to generating them. I just wanted to flag this since with this new recommended approach I guess more people will make scripts like mine (@SSwarper in sequence with afniproc) and I would guess we are not the only ones with more than one analysis machine with access to the same network home directory.



Edited 1 time(s). Last edit at 02/21/2020 09:00AM by Robin.