I'm working on a gaze-contingent experiment which requires a very precise presentation of the stimulus. It seems like Psychtoolbox has an experimental feature that takes advantage of the FreeSync/Gsync technology for this purpose: the 'kPsychUseFineGrainedOnset' flag for 'specialFlags' in Screen('OpenWindow', ...). However, the manual seems to imply that this is experimental in nature and it may not work properly in practice. Has anyone tried this (or any other features) in an experiment? If so, can you please comment on your experience with that?
Also, the manual says that we should use a proper combination of hardware+software but does not say specifically what it should be. It's great if you can give some suggestion on that as well.
Thanks,
Long