On Sun, Dec 21, 2014 at 08:38:40PM +0000, Carl Eugen Hoyos wrote: > Carl Eugen Hoyos <cehoyos <at> ag.or.at> writes: > > > Michael Niedermayer <git <at> videolan.org> writes: > > > > > * commit '6fd91fa11909f27902498648680dbb3d13f1f175': > > > h264: increase MAX_SLICES to 32 > > > > You had fixed this bug in 716c1f7b > > Does this patch not introduce a speed regression for > no gain? > Or do I misunderstand?
the problem is that we do not know if there is a gain IIRC the bug author claimed in the report that this fixes some artifacts, the sample provided does not require MAX_SLICES 32 IIRC but that was IIRC also not claimed This is not really a technically awnserable question. Its a question about the existence of files in the wild. We could revert this and demand a reproduceable testcase or we could belive the reporter has seen sufficient cases with either value to have had a statistically sound basic to identify that the change in value is the cause of the artifacts either way a reproduceable testcase would be nice also benchmarks would be usefull as well to decide if its better to revert or not [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB The real ebay dictionary, page 2 "100% positive feedback" - "All either got their money back or didnt complain" "Best seller ever, very honest" - "Seller refunded buyer after failed scam"
signature.asc
Description: Digital signature
_______________________________________________ ffmpeg-cvslog mailing list [email protected] http://ffmpeg.org/mailman/listinfo/ffmpeg-cvslog
