[MPlayer-dev-eng] [PATCH] fix for -srate bug

Giacomo Comes comes at naic.edu
Thu Oct 21 16:26:49 CEST 2004


On Thu, Oct 21, 2004 at 01:22:20PM +0200, Diego Biurrun wrote:
> D Richard Felker III writes:
> > On Tue, Oct 19, 2004 at 01:37:02AM +0200, Diego Biurrun wrote:
> > > Giacomo Comes writes:
> > > > 
> > > > This should go in the documentation, or mencoder should be fixed and be able to use
> > > > -af resample without -srate.
> > > 
> > > What is the problem, -af should work with MEncoder.
> > 
> > it doesn't. when the broken af was first committed it was designed for
> > the output end to be overridden by the negotiated parameters between
> > the codec and the ao driver. mplayer was fixed but mencoder never was.
> > so even if you specify resample, mencoder will reconfigure your
> > resample filter to resample to the original rate... :(
> 
> OK, let me try to distill this into documentation.  Currently in the
> man page we have
> 
> resample[=srate[:sloppy][:type]]
>   Changes the sample rate of the audio stream to an integer srate in Hz.
>   It only supports the 16-bit little-endian format.
> 
> IIUC I would just have to mark this MPlayer only, right?  Like so:
> 
> resample[=srate[:sloppy][:type]] (MPlayer only)
>   Changes the sample rate of the audio stream to an integer srate in Hz.
>   It only supports the 16-bit little-endian format.
> 
I agree. Could I suggest to change de description of -srate?

Currently it is:

Selects the given output sampling rate, resampling if necessary.  MEncoder passes this value to lame for resampling.

I suggest something like:

Selects the given output sampling rate, resampling if necessary. The default resampling is optimized for speed (see -af-adv).

The reference to lame isn't accurate anymore because mencoder can use libavcodec too.

Giacomo




More information about the MPlayer-dev-eng mailing list