"Scott McPhillips [MVP]" <org-dot-mvps-at-scottmcp> wrote: "David Webber" <dave@musical-dot-demon-dot-co.uk> wrote in message: The topic "About timers" in the documentation says: "If a high-resolution performance counter exists on the system, you can use the QueryPerformanceFrequency function to express the frequency, in counts per second" The "if" is a bit worrying
"David Webber" wrote: The topic "About timers" in the documentation says: "If a high-resolution performance counter exists on the system, you can use the QueryPerformanceFrequency function to express the frequency, in counts per second" The "if" is a bit worrying: am I likely to encounter systems on which there isn't one? A high-resolution performance counter is an euphemism
"Michael" <michael.at.gmail.dot.com> a écrit dans le message de news: Xns9890BC2304862michaelatgmaildotcom@212.27.60.37... Oui, il faut appeler QueryPerformanceFrequency pour avoir la fréquence du Timer. C'est ce qui vous permettra de relier à un temps, la valeur retournée par QueryPerformanceCounter QueryPerformanceFrequency renvoie usuellement (et historiquement
VERY GOOD! When in doubt, measure. I spent 15 years doing performance measurement, and learned that if you ask a programmer where the performance bottleneck is in the program, you will get the wrong answer. Always. (In fact, there were never any exceptions to this the whole time I was doing performance measurement). But once you've measured, you have facts. Then you can make informed decisions
Joseph M. Newcomer wrote: Has it occurred to you that you can actually *measure* the time yourself? Look at ::QueryPerformanceCounter and ::QueryPerformanceFrequency. Thanks, I think I can test it myself. Why depend on secondary sources when you can get the raw data on your own? Sorry, Joe, I can't understand this sentence above. Do you mean that I should depend on the source
