Is sf::Clock accurate enough for fixed Time step game loops?

[sonicd007]

While asking on gamedev about fixed time steps, I was told clocks should never be used as a game timer and to use QueryPerformanceCounter() instead.  That solution however is Windows specific and I read somewhere that on multicore systems it could return different times since each cpu may not be synced.  That being said, does sf::Clock use something like QueryPerformanceCounter or does it use something like std::Clock.  Sorry I haven't checked the source code, I'm at work and this has been bugging me.

[Nexus]

sf::Clock uses QueryPerformanceCounter() on Windows, so you're on the safe side

[FRex]

/src/SFML/System/Win32/ClockImpl.cpp

Time ClockImpl::getCurrentTime()
{
    // Force the following code to run on first core
    // (see http://msdn.microsoft.com/en-us/library/windows/desktop/ms644904(v=vs.85).aspx)
    HANDLE currentThread = GetCurrentThread();
    DWORD_PTR previousMask = SetThreadAffinityMask(currentThread, 1);

    // Get the frequency of the performance counter
    // (it is constant across the program lifetime)
    static LARGE_INTEGER frequency = getFrequency();

    // Get the current time
    LARGE_INTEGER time;
    QueryPerformanceCounter(&time);

    // Restore the thread affinity
    SetThreadAffinityMask(currentThread, previousMask);

    // Return the current time as microseconds
    return sf::microseconds(1000000 * time.QuadPart / frequency.QuadPart);
}

 

Yes it does, apparently. I think Evil Laurent said that clocks always go for highest avaible resolution but I might be remembering wrong.

[eXpl0it3r]

For Windows it uses QueryPerformanceCounter (see here), for Mac OS X it seems to use mach_timebase_info (see here) and for other Unix based system it seems to use clock_gettime (see here).

None of the functions are perfect, but for nearly all games I can think of it's precise enough and you'll have a hard time finding something better.

[sonicd007]

Thank you very much!