On Wed, Jul 06, 2016 at 07:18:08AM -0400, Ganesh Ajjanagadde wrote: > 06.07.2016, 04:03, "Carl Eugen Hoyos" <ceho...@ag.or.at>: > > Ganesh Ajjanagadde <gajjanag <at> mit.edu> writes: > > > >> > No question, it would be better if tests would be added quicker ... > >> > >> I do not doubt this, but at the moment we do not enforce it. > >> Do you see any trouble in enforcing this requirement from > >> major release to next major release? > > > > I am against adding such a "hard" requirement. > > I believe we have filters that are impossible / very > > difficult to test. > > I do not understand what is so difficult about testing filters. > Yes, if floating point is used, we can't simply test bit-exactness, > but one can simply use some similarity meaure, right?
yes, would require a reference to compare agaist in the fate samples though [...] -- Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB Avoid a single point of failure, be that a person or equipment.
signature.asc
Description: Digital signature
_______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
