UpTime on multiple processors

1
Peter Lennie wrote,
>There's an interesting wrinkle with
>UpTime(). On a multi-processor machine, calls made on tasks running on the
>different processors return different times (presumably reflecting the
>slightly different times at which the processors were started), so it's
>not ideal as a timing mechanism in an MP environment.
>
>P.

interesting!

i suppose one could deal with that by keeping a table of the temporal
offsets of the multiple processors, and correcting for it.

best

denis

ps
I'm cc-ing this to the psychtoolbox forum
http://color.psych.ucsb.edu/psychtoolbox/forum.html