The time of the sleep doesn't seem to matter, if I sleep for 0.01, it runs fine the first time, but slows down the more times I run through it.
On Tue, Jan 20, 2015, 05:43 Andrew Graham <a...@agraham.demon.co.uk> wrote: > You don’t say how long you are trying to sleep for or why you want to > sleep, but every time you call Sleep you abandon the rest of your > timeslice. If you are trying to sleep for a few milliseconds you may be > suprised by how long you actually sleep as your thread has to be > rescheduled after the sleep time expires. I haven’t needed to check this on > Windows 8 but a few years ago we found that on Windows 7 the minimum sleep > time was about 15mS – one system timer tick. In fact I believe this minimum > can be shorter if there are no other threads of the same or higher priority > are waiting to run but we didn’t notice this happening. Sleep is fine for > arbitrary length waits of several hundred milliseconds or more but is > useless for accurate timing and doesn’t work properly for short timeouts. > > Andy Graham > > *From:* Slide <slide.o....@gmail.com> > *Sent:* Monday, January 19, 2015 10:27 PM > *To:* ironpython-users@python.org > *Subject:* [Ironpython-users] time.sleep slowdown > > We're seeing some weirdness in some embedded scripting. Some of our > scripts have calls to time.sleep(), with these calls in place, from run to > run of the script, performance drops significantly, but if we remove these > calls, we don't see any slowdown. The sleep function is implemented with a > call to Thread.Sleep, so I am not sure why this would slow things down so > much. Has anyone seen anything like this, and have a way to overcome it? > > Thanks, > > slide > > ------------------------------ > _______________________________________________ > Ironpython-users mailing list > Ironpython-users@python.org > https://mail.python.org/mailman/listinfo/ironpython-users >
_______________________________________________ Ironpython-users mailing list Ironpython-users@python.org https://mail.python.org/mailman/listinfo/ironpython-users
