On Fri, 2014-02-07 at 16:21 -0800, Ralph Giles wrote: > On 2014-02-07 3:57 PM, [email protected] wrote: > > > Title : Ogg Encapsulation for the Opus Audio Codec > > Filename : draft-ietf-codec-oggopus-03.txt > > I've posted a new version of the Ogg Encapsulation draft for Opus. >
> (2) We haven't addressed Calvin Walton's questions about the various > gain headers. I just want to try to restate and clarify my opinion on this a bit: The ReplayGain tags are application-specific metadata tags defined by a specification that is managed externally to the IETF[1]. As such, I don't think it's appropriate for the Ogg Opus specification to state whether or not the REPLAYGAIN tags should be present in the file, or their specific semantics. Any required changes for handling ReplayGain in Ogg Opus should probably be done in the ReplayGain specs themselves. - One question I'm not sure about. Is the proposed R128_TRACK_GAIN field supposed to be a generic vorbis comment standard that can be used in multiple formats? (e.g. Ogg Vorbis, Flac), or is it intended only for use in Opus? If the former, it should probably be proposed in a more general way. Inasmuch as there is no real standardization organization for multi-format vorbis comment fields, a discussion on the HydrogenAudio forums may be a good place to see if there is any interest in adopting the comment format. Having a uniform set of tag names for vorbis comments is advantageous to people building music players and decoding libraries, since it means that vorbis comment parsing code can be shared between multiple formats. 1) It's currently published on the HydrogenAudio wiki. Current standard (David Robinson's original calculation algorithm) http://wiki.hydrogenaudio.org/index.php?title=ReplayGain_1.0_specification Proposed revision (updated algorithm based on EBU-R128 with gain offset) http://wiki.hydrogenaudio.org/index.php?title=ReplayGain_2.0_specification -- Calvin Walton <[email protected]> _______________________________________________ codec mailing list [email protected] https://www.ietf.org/mailman/listinfo/codec
