[Mplayer-users] AMD K6-2/500

Arpi arpi at thot.banki.hu
Sat May 5 15:56:57 CEST 2001


> So I did some tests on my PC and I`ve got interesting results.
> I played video with -vo null twice, with -nosound in first case and
> with -afm 1 in second case.
> I wrote down VIDEO codec cpu usages at same times in video stream
> for both cases. There were big differences between these values!
> Is it possible? Is A-V sync so hard work? Or there is something
> what I don`t know?
Hmm. These values don't seem to be normal.
A-V sync alone doesn't use any CPU, but it can modify fps to keep video
in sync with audio. And of course, playing at lower/higher fps makes
difference in video CPU usage.
But in normal way, it shouldn't change fps. I know only 3 things which
force fps changes:
- bad file (bad fps or samplerate value in header or something else)
- bad audio playback rate (44khz instead of 48, etc)
- audio driver problems - no working select(), etc.

Is audio playback jerky (non-fluid) or is is perfect?
Is it played at the right speed (no pitch etc)?
Is this bug related to 1 or a few files, or everything?

What are the A-V: and ct: fields in teh status line after some
minutes playing?

A'rpi / Astral & ESP-team

mailto:arpi at thot.banki.hu

Mplayer-users mailing list
Mplayer-users at lists.sourceforge.net

More information about the MPlayer-users mailing list