On 9/10/22, Paul B Mahol <[email protected]> wrote: > On 9/10/22, James Almer <[email protected]> wrote: >> On 9/10/2022 4:11 PM, Paul B Mahol wrote: >>> On 9/10/22, James Almer <[email protected]> wrote: >>>> On 9/10/2022 3:27 PM, Paul B Mahol wrote: >>>>> Patches attached. >>>> >>>> This looks like it would fit in adpcm.c as an ADPCM codec. >>> >>> Nope >> >> nibble, step, prediction. All terminology taken straight out of adpcm, >> down to the decoding process. >> The entire decoder can surely be implemented in adpcm.c and named >> adpcm_msc4, and unless you have a good reason not to I'd really like you >> doing so, because once the AVCodecID is added renaming it becomes a PITA. > > It have parser, and if you look actually carefully different > layout in packets, not sharing anything with adpcm.c code. > > So please refrain from non-constructive comments.
Also it is not really ADPCM codec as it have dynamics weights and use dot product too. So next time look at code more carefully. _______________________________________________ ffmpeg-devel mailing list [email protected] https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email [email protected] with subject "unsubscribe".
