I wasn't commenting on the sleep command in terms of benchmarking, but more as a general comment as that was how I feel the original comment on Sleep/Nicewait was being made.
As far as the algorithm - basically I have no real inkling on how to come up with a check so I was looking for an example of how such a check might work.
re: an inc loop count - well that's what I did, and it didn't seem very fast to me (although current I only have one main loop that pastes the GUI images, and then prints the test lines I insert into my display window's array) and while I'm sure its something like 10 - 30 loops per second, its nothing compared to other loop numbers I've seen in the 10's of thousands or 100's of thousands range.
and for a simple loop pasting GUI stuff and printing some text from an array I got the impression that was a little odd.
I do know that the sync rate and/or Vsync plays a part in that though as it will count up much faster with Vsync off and Sync Rate 0
I suppose what would interest me the most, is how I can extrapolate the "increasing too fast to really count" loops per second number I am generating, into a more stable readout of "Average Loops per second" based number, which would be easier for me to read.
I would just like to add, my interest here is purely academic. I don't have any true concerns about performance or optimizations as my program is very early alpha. This is more a case of me trying to satisfy my natural curiosity about how fast it is running, in as concrete (and easy to read) a reference as I can obtain
