HydrogenAudio

Lossy Audio Compression => MP3 => MP3 - General => Topic started by: The diamond geezer on 2008-01-18 08:48:25

Title: Replaygain in LAME 3.97
Post by: The diamond geezer on 2008-01-18 08:48:25
Hi

I'd like to have LAME compute and apply the ReplayGain values during encoding. After checking the LAME homepage, I don't know which switch to choose: --replaygain-fast or --clipdetect? I'd say --clipdetect but I'd like a confirmation...And where in the command line shall I enter this value? 

Thanks!
Title: Replaygain in LAME 3.97
Post by: Jens Rex on 2008-01-18 08:59:59
--replaygain-accurate does clipdetect and additionally computes the RG value more accurately. See also lame.exe --longhelp.

But, as far as I know, hardly any software uses the replaygain values stored in the LAME tag, so you might want to use --noreplaygain and use software like foobar2000 to compute the replaygain values instead, which stores the values in tags. Disabling the LAME replaygain calculation will probably give a small speed increase during encoding, but I haven't tested that. I usually use --replaygain-accurate just for completeness sake.

It doesn't matter where in the command line you enter these arguments.

Edit: By the way, computing replaygain during encoding only computes track gain values, not album gain, so if you want that, you have to use other software to do that.
Title: Replaygain in LAME 3.97
Post by: The diamond geezer on 2008-01-18 10:54:25
OK thanks, I was just looking for a way to streamline the whole ''open EAC - Encode - Close EAC - Enqueue newly encoded files in Foobar - RG'' process, but I'll keep doing that then 
Title: Replaygain in LAME 3.97
Post by: Northpack on 2008-02-20 10:28:11
But, as far as I know, hardly any software uses the replaygain values stored in the LAME tag, so you might want to use --noreplaygain and use software like foobar2000 to compute the replaygain values instead, which stores the values in tags. Disabling the LAME replaygain calculation will probably give a small speed increase during encoding, but I haven't tested that. I usually use --replaygain-accurate just for completeness sake.


Do you know which software actually does support the LAME Replaygain header? How how it is supposed to handle Replaygain info if there is one in the header aswell as in the tags? Does the tag-info get priority over the header? I'm just wondering because foobar's Replaygain should give more accurate result (like --replaygain-accurate) as is computes the mp3 data. Things would go nasty when applying mp3gain to the files: say a track is gained by 6db. If the player now would take the unchanged value from the header (which still is at apr. at +6db), it would be gained twice and thus become much too loud!