Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Library Eye]

Been aiming to convert playlists to normalized files.

In effort to normalize volume for consistent playback, Volume Leveling and Adaptive Volume> Peak Level Normalize work differently. From what I understand, Peak Level Normalize aims peak of all files toward 0db; this does not result in consistent listening experience if input files vary in dynamic range, since for instance a brickwalled song taken to zero will be loud throughout while a very dynamic file will only peak on occasion. Volume Leveling takes this into account, and uses R128; you can see if viewing converted file waveforms, highly dynamic files tend to get a boost, if there's room for it, and less dynamic files get attenuated.

Both Leveling and Peak Level Normalize are meant to leave each track's dynamic range intact. (I have seen at least when used together, Peak Level Normalize can alter the forms a bit besides an overall volume increase.)

Because Leveling can reduce overall volume of playlist, and quite a lot at times, JRiver dialogue suggests enabling Adaptive Volume to compensate, further explaining in that option's window that these two features work intelligently together.

However, when both are enabled what ends up happening is, the difference between tracks yielded by Volume Leveling is altered, with instead the peak level of all files raised to 0db, effectively giving a result you might expect if Adaptive was not enabled at all. (This does vary a bit across playlists, sometimes not all peaks go to zero.)

To get around this, I have resorted to three step process...
1. convert with Volume Leveling on
2. import resultant files into Audacity, select all, and get an amplification level from the Amplify effect
3. go back to original files in JRIver, and convert with Volume Leveling on and with Parametric EQ set to adjust volume by dB reading provided in Audacity, or just shy of that if you wanna leave a little headroom

Is there no setting in JRiver that can achieve this? Take advantage of the Volume Leveling achieved by R128 implementation, and then raise volume of playlist as a whole to compensate for the volume drop without changing the Volume Leveled difference between tracks? Or have I got something wrong here?

Thanks

[mwillems]

So volume levelling and adaptive playback do work intelligently together during normal playback and you can confirm this by looking at audio path.  There was another thread recently where someone was getting unexpected results when using both options during file conversion (as opposed to playback), which is a less common use case.  This suggested to me that maybe these DSP blocks don't work the same when converting files vs. during playback?

If you take the same playlist of files and play it in JRiver with both volume levelling and adaptive volume enabled and then look at the audio path, do you see each file being boosted to 0 dB?  Or something else?  When I play files with both enabled I see the loudest file getting boosted to 0dB, but the rest vary depending on their relative loudness.  It's important that you test with the same playlist of files in both cases as the specific amount that each file will be altered will depend on what other files are in the playlist.  If you're converting files individually, then each file will always be recentered to peak at 0dB.

[Dawgincontrol]

Personally I would do without adaptive volume.  Just use volume leveling.  Adaptive volume inhibits certain passages.