[FFmpeg-devel] [PATCH] PCM regression tests

[pross at xvid.org]

On Sun, Aug 03, 2008 at 04:28:17PM +0200, Michael Niedermayer wrote:
> On Sun, Aug 03, 2008 at 01:44:46PM +1000, pross at xvid.org wrote:
> > Hi,
> > 
> > Test case for all PCM codecs. Some muxer modifications were required to
> > complete the encoder-decoder loop for PCM Zork and S24DAUD.
> > 
> > add-pcm_zork-to-riff.diff
> >   Support muxing of Zork PCM audio into .AVI format.
> >   (alternatively, we could drop Zork PCM audio encoding. Is anyone
> >    seriously using this?).
> 
> not ok, see below

> [...]
> > @@ -254,6 +255,10 @@
> >      } else {
> >          bps = 16;
> >      }
> > +    if (enc->codec_id == CODEC_ID_PCM_ZORK) {
> > +        enc->codec_id = CODEC_ID_ADPCM_IMA_WAV;
> > +        bps = 8;
> > +    }
> 
> Iam against this, CODEC_ID_PCM_ZORK is not CODEC_ID_ADPCM_IMA_WAV

Ouch. Thanks for spotting this. Revised patch enclosed.

The regression test gives different Zork and Daud checksums on
my BE and LE computers. I'll hold off on it until I find the cause.

Also, I noticed that the Zork file in the MPlayer samples library
sounds really really awful when played back in FFplay and Mplayer.
Michael, you are listed as the author of the Zork code, is this
how it is supposed to sound?!

Cheers,
-- Peter
(A907 E02F A6E5 0CD2 34CD 20D2 6760 79C5 AC40 DD6B)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: add-pcm_zork-to-riff-r2.diff
Type: text/x-diff
Size: 988 bytes
Desc: not available
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20080804/9283f801/attachment.diff>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20080804/9283f801/attachment.pgp>