On 28/05/2024 23:38, Marton Balint wrote: > > > On Wed, 29 May 2024, Jan Ekström wrote: > >> On Mon, May 27, 2024 at 10:31 PM Marton Balint wrote: >>> >>> >>> >>> On Mon, 27 May 2024, Jan Ekström wrote: >>> >>> > On Mon, May 27, 2024 at 12:37 AM Lynne via ffmpeg-devel >>> > wrote: >>> >> >>> >> apichanges will be updated upon merging, as well as a version bump. >>> >> --- >>> >>  libavutil/channel_layout.c | 4 ++++ >>> >>  libavutil/channel_layout.h | 8 ++++++++ >>> >>  2 files changed, 12 insertions(+) >>> >> >>> >> diff --git a/libavutil/channel_layout.c b/libavutil/channel_layout.c >>> >> index 98839b7250..2d6963b6df 100644 >>> >> --- a/libavutil/channel_layout.c >>> >> +++ b/libavutil/channel_layout.c >>> >> @@ -75,6 +75,10 @@ static const struct channel_name >>> channel_names[] = { >>> >>      [AV_CHAN_BOTTOM_FRONT_CENTER  ] = { "BFC",       "bottom >>> front center"   }, >>> >>      [AV_CHAN_BOTTOM_FRONT_LEFT    ] = { "BFL",       "bottom >>> front left"     }, >>> >>      [AV_CHAN_BOTTOM_FRONT_RIGHT   ] = { "BFR",       "bottom >>> front right"    }, >>> >> +    [AV_CHAN_SIDE_SURROUND_LEFT   ] = { "SSL",       "side >>> surround left"    }, >>> >> +    [AV_CHAN_SIDE_SURROUND_RIGHT  ] = { "SSR",       "side >>> surround right"   }, >>> > >>> > Just as a note, if you want to follow the ISO 23091-3 naming for these >>> > (BS.2051/IEC 62574 calls them just SiL, SiR), then it would be Lss, >>> > Rss ("Left side surround", "Right side surround"). Just a word order >>> > thing. This would then also match the Apple identifier, >>> > `kAudioChannelLabel_LeftSideSurround`. >>> >>> I guess the idea was to be consistent with existing FFMPEG channel word >>> order and short names. I kind of prefer that approach, because the other >>> channel names will not follow ISO 23091 anyway, so there is not much >>> benefit in being consistent with ISO 23091 for these 4 newly added >>> channels only, at the cost of not being consistent with existing ffmpeg >>> way... So I think the patch is fine as is. >>> >> >> If we explicitly go for a non-standardized names then it probably is >> an even better idea to add "///< " comments marking what we actually >> based these new enum entries on with regards to matching ISO 23091 and >> IEC 62574 / BS.2051 identifiers (even if we leave out the approximate >> azimuth and vertical degrees out of it, as that is a whole separate >> mess). >> >> I am planning on attempting to add such for at least some of the enum >> values we have for the existing ones, but for these new ones we should >> have these from the get-go. > > Sure, OK, I have no problem with adding comments about the origins of > the channel. And we should ask MediaArea to add a new column to their > table, the FFmpeg channel names :) > > Thanks, > Marton > >> >> Jan >> >>> Regards, >>> Marton >>> >>> >>> > >>> > Also we might want to start adding comments into the enum like "///< >>> > +90 degrees, Lss, SiL" to note the matching ISO 23091-3 and BS.2051 >>> > identifiers that this channel is supposed to match? >>> > >>> >> +    [AV_CHAN_TOP_SURROUND_LEFT    ] = { "TTL",       "top >>> surround left"     }, >>> >> +    [AV_CHAN_TOP_SURROUND_RIGHT   ] = { "TTR",       "top >>> surround right"    }, >>> >>  }; >>> > >>> > So here the mapping seems to be that ISO 23091-3 calls these Lvs and >>> > Rvs (Left, Right vertical surround) and they are mapped against >>> > BS.2051/IEC 62574 TpLS and TpRS - except BS.2051 doesn't seem to >>> > mention these two, it contains as U+110 Ltr and Rtr. At this point it >>> > would be nice if someone actually has access to IEC 62574 :D . >>> > >>> > as for the enum comment, maybe "///< +110 degrees, +30 degrees >>> > vertical, Lvs, TpLS" ? >>> > >>> > references: >>> > >>> > BS.2051 >>> https://www.itu.int/dms_pubrec/itu-r/rec/bs/R-REC-BS.2051-3-202205-I!!PDF-E.pdf >>> > ISO 23091-3: >>> ISO-IECJTC1-SC29_N19613_Text_of_ISOIEC_23091-3CDAmd1_Information_technology__Coding-independent_code_points__Part_3_Audio__AMENDMENT_1_Headphone_support_SC_29WG_06_N_00045.zip >>> > (used to be public on N-documents, but now that site always requires a >>> > login :< ) >>> > ISO 23003-3: USAC spec - contains the same listing as in 23091-3. >>> > Some D document from 2021 on setting up audio setups that has a table >>> > that attempts to match their names to various actual specifications >>> > (including IEC 62574 where it mentions TpLS and TpRS): >>> > >>> https://www.audiosciencereview.com/forum/index.php?attachments/dolby-atmos-home-entertainment-studio-technical-guidelines-2021-05-pdf.246631/ >>> > MediaInfo's attempt at matching the different specs: >>> > https://mediaarea.net/AudioChannelLayout >>> > >>> > Jan >>> > _______________________________________________ >>> > ffmpeg-devel mailing list >>> > ffmpeg-devel@ffmpeg.org >>> > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel >>> > >>> > To unsubscribe, visit link above, or email >>> > ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe". >>> _______________________________________________ >>> ffmpeg-devel mailing list >>> ffmpeg-devel@ffmpeg.org >>> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel >>> >>> To unsubscribe, visit link above, or email >>> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe". >> _______________________________________________ >> ffmpeg-devel mailing list >> ffmpeg-devel@ffmpeg.org >> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel >> >> To unsubscribe, visit link above, or email >> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe". > _______________________________________________ > ffmpeg-devel mailing list > ffmpeg-devel@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-devel > > To unsubscribe, visit link above, or email > ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe". So which names should I resubmit with?