Re: Timer accuracy
From: Kurt Barthelmess (kbarthelmess_at_compuserve.com)
Date: 10/02/03
- Next message: CARLO GARCIA: "I've a problem using FindNextPrinterChangeNotification"
- Previous message: Kyle A. Miller: "Re: Windows Refuses to Shutdown With My App Running"
- In reply to: Jens Gruschel: "Timer accuracy"
- Next in thread: Brian Cook: "Re: Timer accuracy"
- Reply: Brian Cook: "Re: Timer accuracy"
- Reply: Ndi: "Re: Timer accuracy"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Thu, 02 Oct 2003 15:46:25 GMT
"Jens Gruschel" <nospam@pegtop.net> wrote:
>My question: Does anyone know about accuracy of Win9x / WinNT?
No version of WIndows will provide an accurate timer "interrupt" under
all conditions. Timers have one of the lowest priorities around and
your results were a function of the idle status of your system.
>What about
>slower machines?
I t is not the speed so much as it is whether Windows is "busy" when
the timer "ticks". If so, that timer has to wait.
>And finally: Can anyone explain why it can be that
>accurate?
Nominally a PC's clock ticks at 55 msec. intervals, or about 1100 a
minute. (Some versions of WIndows tweak that rate up even higher.)
iac, handling one notification for every 100 possible ones gives lots
of room for error.
>I mean, it takes some time to process the WM_TIMER message,
>execute another SetTimer api call etc.
One a timer is set, messages continue to come in unless you disable /
reset the timer. iac, with a 2GHz system, you can easily process an
uncountable (well, several billion<g>) such interrupts a minute.
Good luck.
Kurt
- Next message: CARLO GARCIA: "I've a problem using FindNextPrinterChangeNotification"
- Previous message: Kyle A. Miller: "Re: Windows Refuses to Shutdown With My App Running"
- In reply to: Jens Gruschel: "Timer accuracy"
- Next in thread: Brian Cook: "Re: Timer accuracy"
- Reply: Brian Cook: "Re: Timer accuracy"
- Reply: Ndi: "Re: Timer accuracy"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|