Normal
I think that scythe42 mentioned something about that (for Win8) a while ago, so it may be a 'feature' of Win8 Do you mean a float (as you had decimal before with the Win32 API) ? MPAR really needs a very accurate measurement of refresh rate (i.e. the one that EVR presenter estimates), otherwise it can create worse dropped/repeated frames (instead of making it better). So it's better to force MPAR bias to 1.0 (no correction) unless we know for sure that the video is running on the primary monitor (and MP was started there etc.). We need to disable EVR presenter vsync correction for the same reason in the 'no primary' situation as well. Of course this hopefully doesn't apply to Win8 I'm happy to make the changes to dshowhelper (for MPAR and vsync correction control), but I think it might be better for MP C# to handle the 'logic' of this - what do you think ?
I think that scythe42 mentioned something about that (for Win8) a while ago, so it may be a 'feature' of Win8
Do you mean a float (as you had decimal before with the Win32 API) ?
MPAR really needs a very accurate measurement of refresh rate (i.e. the one that EVR presenter estimates), otherwise it can create worse dropped/repeated frames (instead of making it better). So it's better to force MPAR bias to 1.0 (no correction) unless we know for sure that the video is running on the primary monitor (and MP was started there etc.). We need to disable EVR presenter vsync correction for the same reason in the 'no primary' situation as well. Of course this hopefully doesn't apply to Win8
I'm happy to make the changes to dshowhelper (for MPAR and vsync correction control), but I think it might be better for MP C# to handle the 'logic' of this - what do you think ?