Le duodi 12 floréal, an CCXXI, Claudio Freire a écrit : > Ok... I had to forfeit my right to sleep, but here it is. Managed to > make both the fast and 2loop methods work reasonably well (albeit I > don't know why anyone would use fast, because it's not faster. I > assume it's an experiment?). > > I don't have access to AAC standards, though, since they're paywalled > (at least the ones I found), so most of the changes were done blind, > assuming from what I know of audio codecs, and by trial and error. > > In particular, fast has some magic numbers there that I got from trial > and error, and nothing more... profound. > > In any case, attached it is, git diff.
Thanks for the patch. I believe you should submit it to ffmpeg-devel, as it will receive more attention. Also, you should use git format-patch rather than git diff, so that author information is included in the patch, and so would be the commit message, where you can explain the changes for future reference. A small style comment; I know nothing about AAC so I can not tell about the validity of changes. > + float lowlambda = av_clip(120.f / lambda, 0.25f, 1.f); > + float rlambda = av_clip(120.f / lambda, 0.3f, 2.f); > + const int minq = av_clip(2 * log2f(120.f / lambda) + 150, 100, 218 - > SCALE_MAX_DIFF); > + const int maxq = minq + SCALE_MAX_DIFF - 1; Do you have a specific reason to use float instead of double? Regards, -- Nicolas George
signature.asc
Description: Digital signature
_______________________________________________ Libav-user mailing list [email protected] http://ffmpeg.org/mailman/listinfo/libav-user
