On Thu, 22 May 2014 14:55:00 +0200, [email protected] (Niels =?iso-8859-1?Q?M=F6ller?=) wrote: > Luca Barbato <[email protected]> writes: > > > On 08/05/14 10:46, Niels Möller wrote: > > >> I have now tested it with > >> http://samples.ffmpeg.org/A-codecs/DTS/dts/5.1%2024bit.dts. I do > >> > >> avconv -request_channel_layout 3 -i 5.1\ 24bit.dts -c:a pcm_s24be dmix.au > >> > >> and compare the output before and after the change. > >> > >> Maximum sample difference I get is 66 (in 24-bit format), or a relative > >> difference of 1.3e-5 or 2^{-16}. I guess this could be normal, the > >> (absolute) difference between between the tabulated float and fixed > >> point values (D.11 in the spec) is also about 2^{-16}. > >> > >> Ok? > > > > Sounds a leap in the right direction to me. > > Ping?
Sorry for the delay, pushed. -- Anton Khirnov _______________________________________________ libav-devel mailing list [email protected] https://lists.libav.org/mailman/listinfo/libav-devel
