Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: opusenc applies replaygain from FLAC while transcoding (Read 9435 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Re: opusenc applies replaygain from FLAC while transcoding

Reply #25
Thanks greynol. Being confined to foobar2000 it's of no interest to me. While maybe more precise, if it doesn't have widespread adoption it's pointless (unless somebody just uses fb2k on Windows). Needing a solution that works on both Windows and Android, replaygain stays with me. Thanks again. :)

Edit: by searching some more I now see that foobar2000 has been using R128 for a very long time now 2011), then writing RG tags with the results. I've been using foobar2000 for a longer time, though... so now... I would probably need to rescan everything to make it even. Does anyone know the exact date when fb2k started using R128 for calculating replaygain values? I'm off to ask on fb2k forums anyway.

Re: opusenc applies replaygain from FLAC while transcoding

Reply #26
foobar2000 doesn't really have anything to do with it.

Try this instead:
http://wiki.hydrogenaud.io/index.php?title=ReplayGain_2.0_specification#Loudness_measurement

So...

do you know which algorithm was used to generate the ReplayGain tags in your files?

...it's somewhat rhetorical; intended to make you think a little harder about what I gave in response to your question about R128.

Re: opusenc applies replaygain from FLAC while transcoding

Reply #27
Yup, I edited my post. I'm pretty sure I have my collection partially scanned with one method and partially with the other.

I've been ripping my CDs since 2003 (when I bought my first Squeezebox) so the above it's pretty much a given... :(

Re: opusenc applies replaygain from FLAC while transcoding

Reply #28
Something similar happened with me, though I also used the metaflac.exe to do the calculation which uses the original algorithm (maybe it has also since been updated, I don't know).  Anyway, close need not only apply to horseshoes and hand grenades.

Sorry for the diversion.  :-[

Re: opusenc applies replaygain from FLAC while transcoding

Reply #29
Opus was specified not only to use its own special gain control header, but to explicitly use the R128 standard and apply it in new R128_* gain tags.  All very sensible if you're defining a brand new codec from scratch, but inevitably poorly supported at first and full of incompatibilities especially for dedicated music software already supporting a different set of tags.  "Album" gain control in particular is a tricky area that was barely supported in the original draft spec, now with its own tag but still uncertainty over how it will actually be used when more Opus gain tools come along.  Use of the original REPLAYGAIN_* tags is explicitly discouraged to avoid confusion between gain tagging schemes, although the wording has been toned down since early drafts and they are now allowed for compliant implementations in circumstances where "there is guaranteed to be no such confusion".

Re: opusenc applies replaygain from FLAC while transcoding

Reply #30
With the R128 standard, I presume it is still based off the -23 LUFS reference as opposed to the -18 LUFS reference that is equivalent(?) to the reference of single channel pink noise signal with an RMS level of -20 dB relative to a full-scale sinusoid being reproduced at 83 dB SPL, but also with 6dB of gain applied at the input?

Feel free to correct any errors in the tedious details which I attempted to regurgitate.

 

Re: opusenc applies replaygain from FLAC while transcoding

Reply #31
With the R128 standard, I presume it is still based off the -23 LUFS reference as opposed to the -18 LUFS reference that is equivalent(?) to the reference of single channel pink noise signal with an RMS level of -20 dB relative to a full-scale sinusoid being reproduced at 83 dB SPL, but also with 6dB of gain applied at the input?

Feel free to correct any errors in the tedious details which I attempted to regurgitate.

The Opus standard doesn't say explicitly, only that the gain should be calculated with reference to the EBU-128 standard.  Which of course to normalise to -23 LUFS, forgetting all the stuff about loudness ranges, multiple channels, peak levels, etc.  The pink noise file is just an aid for testing and setup of equipment, intended to produce 73 dB SPL.  There are other files with defined loudness levels for ensuring compliance.

 
SimplePortal 1.0.0 RC1 © 2008-2021