replaygain plugin query

hi

am creating an audio CD but would like to normalize the volume before burning.

i have 2 questions about its usage. a sample song i chose to analyse has replaygain of -6.8db

  1. is the analyse and normalise function mutually exclusive under replaygain? my understanding is that if replaygain normalises the song, then there would be no need to amplify the signal subsequently by -6.8db.

  2. would opening and normalising 1 track at a time result in any difference than if i load all 20 tracks for the album at the same time?

thanks

I could take a good guess (or you could try it yourself… :wink: ) But, let someone else answer your questions and just throw a couple more steps at 'ya.

ReplayGain is meant for an unlimited number of tracks, so if you’re making a CD you may be able to do better—

After applying ReplayGain, listen to the tracks and if necessary make any additional tweaks by ear. (Be careful not go over 0dB.)

After they are all equal in perceived volume open the Amplify effect on each track and note the peak levels.* DO NOT APPLY the Amplify effect yet.

If any of the tracks peak at 0dB, that’s it and you can stop. (If the loudest track doesn’t hit 0dB, you can bring up the levels of all the tracks.)

If the loudest track peaks at less than 0dB, Apply the Amplify effect to bring the level up to 0dB. Amplify all of the other tracks by the same amount.


\

  • The Amplify effect scans the track and defaults to whatever adjustment is need to normalize (maximize) the peak at 0dB. So for example, if Amplify defaults to +3dB, your peak is currently at -3dB.

No, you don’t have to apply amplify individually or note the values.
Amplify takes always the highest peak over all selected tracks, i.e. it guarantees that no track exceeds 0 dB and only the one with the highest peak reaches this limit (minus your desired head room).

Robert

No, you don’t have to apply amplify individually or note the values.
Amplify takes always the highest peak over all selected tracks, i.e. it guarantees that no track exceeds 0 dB and only the one with the highest peak reaches this limit (minus your desired head room).

YES, YOU DO!!! If you don’t apply the same gain to all of the tracks you will completely un-do the ReplayGain.

ReplayGain matches the perceived loudness, and the additional steps insure that the volume of the overall CD is maximized while maintaining the perceived-volume matching.

Thanks for the replies

  1. Why is the default value for the slider “attenuate normalized output” set at -0.1 db and not 0? That means after replaygain normalizes the file, it then reduces it by 0.1db again? So before normalizing the output have to adjust the slider value to 0 each and everytime?

  2. Is there any downside to applying replaygain twice, once when creating the 16 bit pcm wav file from the 32 bit sample, and then again when the tracks are being finalised to be added to the album? Is there any loss on the 2nd application when working with the 16bit file compared to the original 32bit sample?

  3. Will the function “preserve difference in volume between tracks in album”(i’m not sure if this is in replaygain) be saved in the .cue file that is subsequently created, and if so, will it work on a 1980s manufactured CD player?

Thanks again

  1. If you’re using the same plug-in as me, the slider would be called “Adjust normalized Output”.
    The last setting will be stored, thus, once “0.0” is applied, it will be reused next time.
    The slider is there because ReplayGain is often used with two different reference levels (I think something around 87/89 dB SPL).
    Again, it will be better to ignore this setting and do the normalization by the usage of “Amplify” (but not normalize, which does defeat the replay gain result) afterwards.
  2. It is always better to apply effects on 32 bit data as this guarantees thousands of edits without apparent degradation. However, 16 bit can as well be amplified dozens of times if not hundreds.
  3. I can’t follow, are we still in Audacity? Replay gain is permanently applied when the files are being exported. The procedure we gave makes all tracks on the album approx equally loud.
    You could adjust softer tracks manually with the gain slider afterwards.
    There’s no easy way to keep the relative loudness for an entire album unless you import it as a single track and do the analysis then.
    Makes not much sense for self-compiled CDs.

Robert

Thanks

  1. I am new to audacity so i was under the mistaken impression from initial google search that replaygain possesses the same functions inspite of different software implementations

How is the gain slider function different from the amplify effect? If i wanted to adjust the volume for tracks which are way too far out with the others after replay gain normalization, i should first use gain slider on the problem tracks and then apply amplify effect on all tracks in the album with the lowest amplitude track as reference for 0db?