Strangest sf::sleep bug

[Laurent]

the same effect as opening chrome.

Which is? To solve the problem? (sorry, I don't really remember the begining of this discussion)

[Daddi]

Yes, the moment the chrome browser starts, the bug seems to vanish.
Could it be, that chrome sets windows sleep-time temporarily down to 1 (instead of 15.6 which was named earlier) and back after it gets closed?

[Laurent]

Could it be, that chrome sets windows sleep-time temporarily down to 1 (instead of 15.6 which was named earlier) and back after it gets closed?

I don't know. It could be, yes.

[cire]

It would seem that it does.

Some interesting info here.

[Daddi]

Thats very intersting! But I wonder, why this doesn't happen on every windows machine. At home, everythings fine, at work -> sleep "bug".

[Foaly]

Wow this is a very interesting article! So using timeBeginPeriod(1); seems to be quiet common for using high resolution timer/sleep.

[Daddi]

The earlier posted fix sets timeBeginPeriod every time before a single sleep is engaged. A better solution may be to do this once during the initialization of the application.

[binary1248]

So that when the application terminates in some undefined way, the system gets stuck in high resolution mode? Changing the mode does not incur that much of a performance penalty and since it is used during sf::Sleep, whether that time is spent setting the mode or sleeping doesn't really matter.

[Daddi]

True, I didn't think of this. But theoretically, this could happen with chrome too

[Foaly]

Im sorry for reactivating this old topic. I know it's not a good practice, but since there where quiet a few topics recently about sf::sleep being a bit off (for example through setFramerateLimit() ) I remembered this thread. I dug it up and saw that there was quiet a lot of information gathered and even a solution proposed, but the problem was never actually solved. So shouldn't we investigate further or at least create a ticket on the tracker?

[eXpl0it3r]

So shouldn't we investigate further or at least create a ticket on the tracker?

Hehe, I was thinking the same, but didn't go looking for the thread. Would be nice if we can find a good solution.

[Foaly]

Hehe, I was thinking the same, but didn't go looking for the thread. Would be nice if we can find a good solution.

Good to hear! So I am not the only one I created an issue on the tracker: #439

[Cornstalks]

Hehe, I was thinking the same, but didn't go looking for the thread. Would be nice if we can find a good solution.

Good to hear! So I am not the only one I created an issue on the tracker: #439

I personally think a "high resolution sleep" is a bad idea, as it's very platform dependent, and on some platforms completely impossible.

I think instead something that yield's the current thread's timeslice back to the system would be much more effective. (i.e. SwitchToThread() on Windows (not Sleep(0), which is different)); this would be most useful for programs that don't want to unnecessarily max out the CPU (like mobile devices, where battery is important).