[FFmpeg-cvslog] r18096 - trunk/tests/seek.regression.ref

[Reimar Döffinger]

On Sat, Mar 21, 2009 at 12:14:39AM -0700, Baptiste Coudurier wrote:
> It wasn't meant to fix them. When diffing I checked the new values and
> they were better/not worse than before, and this case was especially
> better, so updating them was IMHO ok, and was fixing FATE.

I certainly don't complain that you updated them, that's good!
It's just that I always have a bad feeling about such unexpected side
effects, even when it seems to unexpectedly improve things it's quite
often just a bug in hiding.
If you're confident it isn't I'll just stop complaining.