Print

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

[sKiZo]

Interesting discussion going on another board ... apparently Windows built in Media Player's volume control is considered lossy, in that it enables an addon to digitally reprocess the data stream - that addon is otherwise bypassed when the volume is all the way up. Same apparently holds true if the master volume setting isn't also defeated by leaving it up all the way.

Not sure if that's specific to DirectSound or WASAPI, both or neither. Don't know how much truth there is to that as it doesn't seem to be a well documented "feature", but makes me wonder about MC's internal volume controls?

PS ... I don't use Windows media player, and I do already leave the system volume all the way up, but I DO use the MC volume setting to tweak playback as needed - too lazy to use the analog controls available on my system. MC's internal volume is usually at 93%, but varies from 85% - 100% depending on the album selection.
 

[blgentry]

MC's internal volume is done in a 64 bit lossless domain.  So there should be no real "degradation" of the signal; it's about as good as it gets for digital volume control.

As long as you are using WASAPI, ASIO, or another method that bypasses the OS touching the sound, JRiver should be the only component in the processing chain.

Brian.

[sKiZo]

Darn! There goes my new years resolution to get more exercise! <G>

Same hold true with any other internal processes like loudness? I also use the convolution kernel and adaptive volume by default.

PS ... sort of on topic. Still happy with v19 here ... any improvements in adaptive volume and volume leveling since? As mentioned earlier, I still have "issues" here with quite a wide range of volume adjusting required going from album to album ... I tend to play full sides at least and understand those processes work differently there than they would with a mixed playlist. Current solution is to normalize in Audacity to smooth the results when I get one that I know is gonna scream at me if I forget to turn the volume down.

[blgentry]

Same hold true with any other internal processes like loudness? I also use the convolution kernel and adaptive volume by default.

Everything in DSP Studio is done in a 64 bit lossless domain.  So yes. 

PS ... sort of on topic. Still happy with v19 here ... any improvements in adaptive volume and volume leveling since? As mentioned earlier, I still have "issues" here with quite a wide range of volume adjusting required going from album to album ... I tend to play full sides at least and understand those processes work differently there than they would with a mixed playlist. Current solution is to normalize in Audacity to smooth the results when I get one that I know is gonna scream at me if I forget to turn the volume down.

I never used 19, so I'm not all that sure.  I think MC19 uses R128 volume analysis, so that's the current "standard" even in MC21.  I don't do a lot of album playback; mostly mixed playlists for me.  With that, Volume Leveling plus Adaptive Volume (peak level normalize) works pretty well for me.  Some volume differences are to be expected of course.

Have you run analyze audio on all of your audio files?

Brian.

[sKiZo]

That's actually a very interesting question, according to some recent reading ...

The answer is (drumroll please) yes ... and no.

Been here since v16 if I remember right, and from what I gather, any ANALYZE done prior to v19 needs to be redone to conform with the "improved" R128 standard. I should probably do that before I continue to complain ... <G>

Thank the gods I can do that with one click ... just point to the library, select all, say go, and walk away. I imagine it'll take a while to run thru the entire collection ...