[Ffmpeg-devel] [BUG] regression tests failure in r7867

[Michel Bardiaux]

Guillaume Poirier wrote:
> Hi,
> 
> Diego Biurrun wrote:
[snip]

> That version number can't be right. Current version of ffmpeg is 7950.
> Did you use IPOT (IP over Time) to fetch your svn copy of ffmpeg? ;o)
> Ok, I guess you meant 7865? ;o)
> Anyway, I think you're right about the fact that it's the WMA regtests
> that "broke" regression tests on PPC.
> I agree with Michael's analysis, it's gotta be a rounding problem.
> For the record, the regtests also fail on x86-64:
[snip]

Well, this leaves me in a tight spot. I was trying to incoprporate the 
regression changes due to the addition of the FACT chunk to non-PCM 
WAVs, and something didnt smell right, so I temporarily disabled the 
FACT chunks, and make test fails:

--- ./data/ffmpeg.regression    2007-02-12 18:25:24.000000000 +0100
+++ 
/besdev01/home/people/michel/internet/ffmpeg-svn-fact/tests/ffmpeg.regression.ref 
   2007-02-12 18:22:41.000000000 +0100
@@ -200,5 +200,5 @@
  stddev:2099.31 PSNR:29.88 bytes:1048576
  ef725cb9c13b9437a41428eabadeba52 *./data/a-vorbis.asf
  35686 ./data/a-vorbis.asf
-41cc3916b3447dfc181acdaa9b1c4e6e *./data/out.wav
+e46368ef818f2ec3abcc3f8390d14df5 *./data/out.wav
  stddev:3882.46 PSNR:24.54 bytes:1056768

I've posted the patch for the FACT-related changes only, which have a 
clear fingerprint since they add 12 bytes.

-- 
Michel Bardiaux
R&D Director
T +32 [0] 2 790 29 41
F +32 [0] 2 790 29 02
E mailto:mbardiaux at mediaxim.be

Mediaxim NV/SA
Vorstlaan 191 Boulevard du Souverain
Brussel 1160 Bruxelles
http://www.mediaxim.com/