Print

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

[whatwhat]

strangely enough, for example, on peak level normalize of adaptive volume if i put five music files in a new playlist, volume level of all files is added to 12~15db.
but if i put the same five files in a playlist which has many mp3 and flac files exist,  volume level of all files is added just to 5~6db regardless of file formats in it.
i suppose mc21 checks all files in a playlist and decides an adaptive volume level for them.

if so, i don't like this volume leveling because some playlists get higher volume level and others get lower if i turn on peak level normalize of adaptive volume. if this issue is not a bug, i think this is kinda improper for those who manage many playlists.

[RD James]

That's because you're combining adaptive volume with volume leveling.

If you want all tracks to play at an equal level, use volume leveling on its own.
If you want each track to play as loud as it can, use adaptive volume on its own.
If you want the playlist to have all tracks leveled and then boosted to play as loud as it can without clipping, enable both.

Sounds like you want #2.

[whatwhat]

thank you for your lucid advice
of the three, which does output the best sound without quality loss or with minimum noise and clip?

[~OHM~]

I would say the one that sounds best to you!

[RD James]

thank you for your lucid advice
of the three, which does output the best sound without quality loss or with minimum noise and clip?

Volume Leveling on its own sounds best, as it means that all tracks play at the same level.

[blgentry]

of the three, which does output the best sound without quality loss or with minimum noise and clip?

None lose quality.  None generate noise.  None allow clipping.  They should all be pretty equal in those respects.  I think, based on what you've said, you want just Volume Leveling, by itself.  Using just VL all playlists will have the same volume target level.  So all playlists should have a very similar average volume level and that average volume level will be the same for all songs in that list.

Brian.

[stewart_pk]

If you want each track to play as loud as it can, use adaptive volume on its own.

I have Adaptive Volume on and Volume Leveling Off and Audio Path is reporting that peak level normalise (Fixed) is adjusting volume by -5.0dB so that's not "as loud as it can" in my book.

[glynor]

In my opinion, enabling both Volume Leveling and Adaptive Volume (in Peak Level Normalize mode) provides the "best quality".

This provides a good listening experience in a "do no harm" kind of way. It levels the volume of playback so that you don't have wild swings in apparent loudness as you play different content. And, it boosts the volume of playback to compensate for the volume reduction this causes, but does so in a way that preserves intra-track dynamic range on an album (or an entire playlist).

For the details, Matt explains it quite well here:
http://yabb.jriver.com/interact/index.php?topic=82023.0

None lose quality.  None generate noise.

Well, that's not quite true.

It is true if you mean when Adaptive Volume is in Peak Level Normalize mode. That's generally what has been discussed here. But Adaptive Volume also has Night and Small Speaker modes which certainly do impact quality (but sacrificed at the altar of usability depending on the situation). Everyone was using Adaptive Volume interchangeably with Peak Level Normalize, so I thought it was worth making clear.

I have Adaptive Volume on and Volume Leveling Off and Audio Path is reporting that peak level normalise (Fixed) is adjusting volume by -5.0dB so that's not "as loud as it can" in my book.

Yeah, that's because this:

If you want each track to play as loud as it can, use adaptive volume on its own.

That quote gets the "spirit" of the setting fairly well, but isn't quite right.

Matt explains:

Adaptive Volume without Volume Leveling will adjust to the peak of the playlist in 'Peak Level Normalize' mode, not each track.

The other modes are more adaptive so it's a blurry line if they're adjusting the track or playlist.

It considers the entire Playing Now list, not individual tracks. I also don't think Matt ever said it goes absolutely right to the edge. You can search his posts if you want, but it probably has some conservative "fudge factor" built in.

[RD James]

I have Adaptive Volume on and Volume Leveling Off and Audio Path is reporting that peak level normalise (Fixed) is adjusting volume by -5.0dB so that's not "as loud as it can" in my book.

It will play as loud as it can without clipping.
The track that you are playing must be a highly compressed lossy track. The lower the bitrate, the lower the volume level playback typically has to be to avoid inter-sample clipping.
Check the Peak Level R128 tag for that track (or album) and it will be around +4dB.
There is an additional -1dB reduction on playback because the analysis is not perfect.

In my opinion, enabling both Volume Leveling and Adaptive Volume (in Peak Level Normalize mode) provides the "best quality".

This provides a good listening experience in a "do no harm" kind of way. It levels the volume of playback so that you don't have wild swings in apparent loudness as you play different content. And, it boosts the volume of playback to compensate for the volume reduction this causes, but does so in a way that preserves intra-track dynamic range on an album (or an entire playlist).

If you don't mind adjusting your volume when you make a change to the playlist, this is fine.
Every time you add or remove an album, it can shift the whole playback level up or down. (or it might not affect it)
If you use Volume Leveling alone, it uses a fixed playback level instead of a variable one.

Volume Leveling on its own means that you set the volume as a preference.
Volume Leveling + Adaptive Volume means that you have to adjust the volume in response to what is being played.
It's still far better than just playing albums without any leveling at all though which can be a lot more variable.

If you just set up a playlist and go, without making changes to it, then you might not see the difference, because everything within that playlist is still being leveled.

Matt explains:
It considers the entire Playing Now list, not individual tracks. I also don't think Matt ever said it goes absolutely right to the edge. You can search his posts if you want, but it probably has some conservative "fudge factor" built in.

I did some testing.
Peak-level normalization still behaves on a dynamic track/album basis like volume leveling does.
So if you play an album, it is based on the loudest track in that album. (on an album where tracks are +0.2dB and -7.0dB, the whole album plays at -0.8dB)
If you play random tracks, it operates on a per-track basis. (when mixed with tracks from different albums, the -7.0dB track plays at +6.0dB)

It only treats the playlist as a whole if you have Volume Leveling enabled at the same time.
This is the way it should be. It's a very smart system.

The only thing I can think that it's missing is a way to tag an album that forces it to use per-track leveling/normalization for those rare albums (usually compilations) where that would be better.