Higher-resolution time for win32

Marc Lehmann schmorp at schmorp.de
Thu Jul 2 18:49:35 CEST 2009


On Thu, Jul 02, 2009 at 09:21:09AM -0400, Andy Grundman <andy at hybridized.org> wrote:
> value every 1ms.)  That's the main issue I was trying to fix.  It's a 

Yeah, but you can't really fix it. Such exact clocks are simply not
generally available on typical PC style hardware.

Note that if you want to benchmark a piece of code, than using the
performance counters might be a good idea. For measuring time, they are a
bad idea.

>> Also note that libev doesn't guarantee any higher precision then  
>> around
>> 1ms even on systems supporting higher resolutions.
>
> Really?  I thought Linux/Mac would be fine in this respect.

The epoll *API* is not more exact than 1ms, no matter how good the
implementation is (and the implementation usually worse, depends on amny
factors, however).

And I doubt anything is fine on Mac. The fact that instead of fixing
kqueue they now also broke poll is a typical example, not the exception. I
would be surprised if macs had better then 100hz granularity in libev, and
seriously shocked if it was better than 1000hz.

-- 
                The choice of a       Deliantra, the free code+content MORPG
      -----==-     _GNU_              http://www.deliantra.net
      ----==-- _       generation
      ---==---(_)__  __ ____  __      Marc Lehmann
      --==---/ / _ \/ // /\ \/ /      pcg at goof.com
      -=====/_/_//_/\_,_/ /_/\_\



More information about the libev mailing list