ChessUSA.com TalkChess.com
Hosted by Your Move Chess & Games
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

microsecond-accurate timing on Windows
Post new topic    TalkChess.com Forum Index -> Computer Chess Club: Programming and Technical Discussions Flat
View previous topic :: View next topic  
Author Message
Martin Sedlak



Joined: 26 Nov 2010
Posts: 701
PostPost subject: microsecond-accurate timing on Windows    Posted: Mon May 28, 2012 10:20 am Reply to topic Reply with quote
I have been thinking recently about time measurement in Windows.
GetTickCount (has granularity of ~15 msec which is bad)
timeGetTime depends on last period set using with timeBeginPeriod,
the docs say it is system-global, so if an app uses timeBeginPeriod(40)
while your program is running, you immediately get worse granularity
than GetTickCount. Also it is said that this affects thread scheduling so better keep hands off it.
QueryPerformanceCounter/QueryPerformanceFrequency:
ultra high resolution, if i remember these calls used to take longer than GetTickCount. Not always available.
rdtsc instruction: very fast but there may be problems with multiple cores and dynamic changes in CPU freq, nonportable
Are there better alternatives I missed?

Now what if I wanted to measure in microseconds instead of some units (usually close to CPU clock freq) which QPF provides?
Here's my solution using QPC/QPF. Note that it's fairly slow on a 32-bit machine here, getMicrosec() itself takes about 1-2 microseconds here () which is already a lot.
Anyway, here's my solution with implementation in case it's useful to someone: (left out some impl. details but I guess it's self-explanatory)

Code:

static volatile signed char init = 0;
static i64 freq;
static u32 shortFreq;
static i64 lastTick;
static i64 remainder = 0;
static i64 emul = 0;
static Mutex usMutex;
static Mutex gMutex;
static i32 lastTC;

i64 getMicrosec()
{
   if ( init == -1 )
   {
      LARGE_INTEGER tmp;
      QueryPerformanceCounter( &tmp );
      i64 cur = (i64)tmp.QuadPart;
      {
         MutexLock _( usMutex );
         i64 delta = cur - lastTick + remainder;
         i64 us = delta * 1000000 / shortFreq;
         emul += us;
         remainder = delta - us * freq / 1000000;
         lastTick = cur;
         return emul;
      }
   }
   if ( init == 1 )
   {
      LARGE_INTEGER tmp;
      QueryPerformanceCounter( &tmp );
      i64 cur = (i64)tmp.QuadPart;
      {
         MutexLock _( usMutex );
         i64 delta = cur - lastTick + remainder;
         i64 us = delta * 1000000 / freq;
         emul += us;
         remainder = delta - us * freq / 1000000;
         lastTick = cur;
         return emul;
      }
   }
   if ( init == 2 )
   {
      MutexLock _( usMutex );
      i32 cur = (i32)GetTickCount();
      i32 delta = cur - lastTC;
      i64 tmp = (i64)delta * 1000;
      lastTC = cur;
      return emul += tmp;
   }
   
   // initialize

   MutexLock _( gMutex );

   LARGE_INTEGER frq;
   if ( QueryPerformanceFrequency( &frq ) == FALSE )
   {
      // not available => use milisec emulation
      init = 2;
      lastTC = GetTickCount();
   } else {
      freq = (i64)frq.QuadPart;
      LARGE_INTEGER tmp;
      QueryPerformanceCounter( &tmp );
      lastTick = (i64)tmp.QuadPart;

      if ( freq <= 0xffffffffU )
      {
         shortFreq = (u32)freq;
         init = -1;
      }
      else
      {
         init = 1;
      }
   }
   return getMicrosec();
}

i32 getMillisec()
{
       return (i32)(getMicrosec()/1000 & 0xffffffffU);
}


Basically what this does is similar to GetTickCount except that it has a microsecond resolution and uses 64-bit output counter. And getMillisec() is a 1-millisecond accurate version.
Of course if perfcounter is not available the code falls back to emulation using GetTickCount().
Back to top
View user's profile Send private message
Display posts from previous:   
Subject Author Date/Time
microsecond-accurate timing on Windows Martin Sedlak Mon May 28, 2012 10:20 am
      Re: microsecond-accurate timing on Windows Martin Sedlak Mon May 28, 2012 12:56 pm
      Re: microsecond-accurate timing on Windows Vincent Diepeveen Mon May 28, 2012 2:44 pm
            Re: microsecond-accurate timing on Windows Martin Sedlak Mon May 28, 2012 3:22 pm
                  Re: microsecond-accurate timing on Windows Vincent Diepeveen Mon May 28, 2012 3:30 pm
                        Re: microsecond-accurate timing on Windows Martin Sedlak Mon May 28, 2012 5:10 pm
                              Re: microsecond-accurate timing on Windows Vincent Diepeveen Mon May 28, 2012 7:05 pm
            Re: microsecond-accurate timing on Windows Martin Sedlak Mon May 28, 2012 3:36 pm
                  Re: microsecond-accurate timing on Windows Vincent Diepeveen Mon May 28, 2012 3:39 pm
                        Re: microsecond-accurate timing on Windows Martin Sedlak Mon May 28, 2012 5:20 pm
                              Re: microsecond-accurate timing on Windows Vincent Diepeveen Mon May 28, 2012 6:23 pm
                                    Re: microsecond-accurate timing on Windows Martin Sedlak Mon May 28, 2012 7:13 pm
                                          Re: microsecond-accurate timing on Windows Vincent Diepeveen Mon May 28, 2012 7:18 pm
                                                Re: microsecond-accurate timing on Windows Martin Sedlak Mon May 28, 2012 8:33 pm
                                                      Re: microsecond-accurate timing on Windows Aleks Peshkov Tue May 29, 2012 7:50 am
                                                            Re: microsecond-accurate timing on Windows Martin Sedlak Tue May 29, 2012 8:26 am
                                          Re: microsecond-accurate timing on Windows Vincent Diepeveen Mon May 28, 2012 7:37 pm
                                                Re: microsecond-accurate timing on Windows Martin Sedlak Mon May 28, 2012 8:07 pm
                                                      Re: microsecond-accurate timing on Windows Vincent Diepeveen Tue May 29, 2012 12:44 pm
                                                            Re: microsecond-accurate timing on Windows Martin Sedlak Tue May 29, 2012 1:15 pm
                                                                  Re: microsecond-accurate timing on Windows Vincent Diepeveen Tue May 29, 2012 3:07 pm
                                                                        Re: microsecond-accurate timing on Windows Martin Sedlak Tue May 29, 2012 4:38 pm
                                                      Re: microsecond-accurate timing on Windows Vincent Diepeveen Tue May 29, 2012 12:55 pm
                                                      Re: microsecond-accurate timing on Windows Vincent Diepeveen Tue May 29, 2012 1:11 pm
                                                            Re: microsecond-accurate timing on Windows Martin Sedlak Tue May 29, 2012 1:29 pm
                                                                  Re: microsecond-accurate timing on Windows Vincent Diepeveen Tue May 29, 2012 3:18 pm
                                                                        Re: microsecond-accurate timing on Windows Martin Sedlak Tue May 29, 2012 5:00 pm
                                                                              Re: microsecond-accurate timing on Windows Wylie Garvin Tue May 29, 2012 9:38 pm
                                                                                    Re: microsecond-accurate timing on Windows Vincent Diepeveen Tue May 29, 2012 9:48 pm
                                                                                          Re: microsecond-accurate timing on Windows Martin Sedlak Wed May 30, 2012 9:10 am
                                                                                          Re: microsecond-accurate timing on Windows Vincent Diepeveen Fri Jun 01, 2012 9:24 am
                                                                                          Re: microsecond-accurate timing on Windows Martin Sedlak Fri Jun 01, 2012 10:52 am
                                                                                          Re: microsecond-accurate timing on Windows Vincent Diepeveen Sat Jun 02, 2012 3:36 pm
                                                                                          Re: microsecond-accurate timing on Windows Martin Sedlak Sun Jun 03, 2012 7:45 am
                                                                                          Re: microsecond-accurate timing on Windows Vincent Diepeveen Mon Jun 04, 2012 11:25 am
                                                                                          Re: microsecond-accurate timing on Windows Vincent Diepeveen Mon Jun 04, 2012 11:39 am
                                                                                          Re: microsecond-accurate timing on Windows Martin Sedlak Mon Jun 04, 2012 12:31 pm
                                                                                          Re: microsecond-accurate timing on Windows Martin Sedlak Mon Jun 04, 2012 12:12 pm
                                                                                    Re: microsecond-accurate timing on Windows Martin Sedlak Wed May 30, 2012 9:13 am
                                                                                          Re: microsecond-accurate timing on Windows Vincent Diepeveen Fri Jun 01, 2012 9:29 am
                                                                        Re: microsecond-accurate timing on Windows Martin Sedlak Tue May 29, 2012 5:25 pm
                                                                              Re: microsecond-accurate timing on Windows Vincent Diepeveen Tue May 29, 2012 7:19 pm
                                                                                    Re: microsecond-accurate timing on Windows Martin Sedlak Wed May 30, 2012 11:41 am
                                                                                          Re: microsecond-accurate timing on Windows Thomas Petzke Wed May 30, 2012 2:57 pm
                                                                              Re: microsecond-accurate timing on Windows Vincent Diepeveen Tue May 29, 2012 7:27 pm
Post new topic    TalkChess.com Forum Index -> Computer Chess Club: Programming and Technical Discussions

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum




Powered by phpBB © 2001, 2005 phpBB Group
Enhanced with Moby Threads