Replaygain variable

[baja]

Request for the following new variables:

%rgt% - Replaygain track
%rga% - Replaygain album
%rgg% - Replaygain which is actually applied

...for custom tags in "Now playing".

If the value is not present in tags, they shall return "n/a".

Thanks :-)

[GoneMAD]

i can add it to the request list. replaygain is currently only handled in the low level audioengine and doesnt get exposed anywhere that the UI can get it which is why its not currently available

[baja]

Thank you :-) Seems there also lies the problem why Opus-encoded files *appear* unhandled by replaygain -- they all sound the same volume level (Android 4.2.2, Huawei P6-U06). It would be very nice if you could find a way to get ReplayGain to work with Opus files as well.

[GoneMAD]

opus doesnt use replaygain.. it uses R128_TRACK_GAIN and its supposed to be built into the decoder according to the documentation.. so the replaygain setting in GMMP will do nothing because its always being applied

If you use replaygain (REPLAYGAIN_TRACK_GAIN) tags they will be picked up and handled by gmmp

[baja]

Hmm... can't get this to work.

Method A: Use all player defaults, except apply track_gain checked —> Replaygain/R128_gain won't be applied.

Method B: Use EQ preamp to +12dB won't help either.

Problem is, foobar2000 will only write the R128_TRACK_GAIN tag to opus files. My device's opus decoder does not seem to apply this gain, leading to volume jumps between opus and other (mp3/ogg/aac) files.

Xiph.org's Todo list says there still needs a tool to be written that does R128_TRACK_GAIN analysis and tagging. Can't find a tool that would write traditional replaygain_track_gain tags to opus files...

Question: Is there no way to "translate" the R128_TRACK_GAIN value (looks like just another vorbis comment?!) for use in gmmp? Would be great for early adopters. Or an alternative: Have the gmmp scanner scan opus files and write its own replaygain info to the file/database?

[GoneMAD]

gmmp will never do scanning of files to calculate the replaygain.. thats beyond the scope of the player

and yea i dont know i just remember reading the decoder was supposed to already handle it.. meaning regardless of waht the GMMP settings are.. the gain is going to be the same.. because the decoder already processed it before it gets the gmmp.. whether this is true or not.. i dont know.. opus is not a format i use for more than just simple testing

i planned to eventually read the values and make it work alongside replaygain but its not a very high priority at the moment

btw xiph tags are freeform.. you can put in whatever you want so there is nothing stopping you from just adding in the replaygain tags manually. R128 is on a different scale from replaygain it seems.. so some conversion would have to be made (i think they are just based off a different max gain value)

[GoneMAD]

do me a favor and in gmmp compare an opus file with the R128 tags and then one with the r128 tags stripped and see if they play at the same volume.. if so then i guess the decoder doesnt handle the r128 processing

[baja]

With today's release of foobar2000 beta (v1.3 beta 5) and re-encoding my opus files (that have been previously created using older fb2k binaries and were used as starting point of this thread), the issues appear to be gone.

Looks like a bug in fb2k has been fixed, that had *nothing* to do with GMMP. I sincerely apologize for misinterpreting my observations.