Many years ago, I requested that Squeezebox server's replay gain be compatible with J River MC's audio analysis. The Squeezebox developers complied and it has always worked well for me. Now I am reading about changes to audio analysis in version 19, and I'm worried that it won't be compatible with the Squeezebox server. Are there any MC19 and Squeezebox users that can speak to this?
There are two scenarios:
a) MC is serving the audio track via Whitebear to LMS in native audio format (e.g FLAC)
b) MC is serving the audio track via Whitebear to LMS in PCM format
In MC v18, in case a) the file is served "as-is" which means that LMS can read / act upon any replay gain tags therein. And in case b) the file is converted by MC to PCM which means that MC can read / act upon any replay gain tags therein, and LMS will not see any such tags. i.e. in case a) any eventual replay gain adjustment could be done by LMS, whereas in case b) it would be done by MC.
Now, (please correct me if I am wrong), in MC v19, if replay gain adjustment is turned on in MC, then both cases a) and b) will involve transcoding by MC. Now from what MrC says, I understand that in case a) the transcoded file might have a replay gain adjustment applied by MC and also have the same replay gain tag embedded in the output file. So it seems possible that both MC and LMS may be able to apply the same replay gain adjustment in series to the output audio. This potentially giving a double whammy. (However I have not tested this since I don't have the test files to do so...)
BTW this situation would not be unique to Whitebear / LMS. The same argument would apply to any other UPnP renderer that has its own capability to apply replay gain...