PDA

View Full Version : Am I a victim of the Loudness War?



tahel
10-04-2010, 12:55 PM
I've been reading up on the Loudness War the Loudness War (http://en.wikipedia.org/wiki/Loudness_war).

I have a lot of remastered CD's and I'm staring to wonder how many of my discs have reduced dynamic range due to de-ranged producers.

Is there a way to use dbPoweramp batch converter to detect and report dynamic range and possible clipping on CDs? I'm thinking I could use the test converter together with a plugin that detects dynamic range and percentage of time at max values for songs.

Since I have duplicates of some discs with both originals and re-issues, this could be a great starting point to determine which discs to do A/B testing of.

Spoon
10-04-2010, 02:37 PM
ReplayGain values are a good indication.

Teknojnky
10-04-2010, 03:02 PM
to elaborate, (album) replay gains closest to zero are generally considered the best, or least compressed.

IE if you have 2 copies of the same album, one @ -6 db's, the other @ -11 dbs, then the one with -6 would probably be the least compressed.

tahel
10-04-2010, 07:13 PM
Great,
Just downloaded the ReplayGain utility and running it on 15 000+ FLAC files.

And tips as to what would be the best utility to view this over a large file set?

Basically I'd like to sort all albums by Replay Gain value.

Are you sure -6 dB is better than -10 dB?

According to the Wikipedia article, the albums have higher loudness and thus poorer dynamic range the closer they are to 0.

Spoon
10-05-2010, 03:11 AM
ReplayGain values are reduction values, so -10dB is reduced more than -2dB, 0dB and the file is left untouched.

garym
10-05-2010, 06:43 AM
According to the Wikipedia article, the albums have higher loudness and thus poorer dynamic range the closer they are to 0.

WRONG!
Also, mp3tag would allow you to load up all your files and sort on the replaygain tag info. (and it handles FLAC, etc.). Or you can use dbpa dsp to export tag data to spreadsheet and sort as you wish).

tahel
10-05-2010, 07:32 AM
WRONG!


Thanks for the constructive criticism. :p Is it me or Wikipedia who is wrong?
Is the dbpa dsp the same as audioinfo?

@Spoon: Thanks for the info. I thought the value might be the average dBFS level, but it makes more sense that it is the value the program will use to normalize the sound level.

garym
10-05-2010, 08:14 AM
Thanks for the constructive criticism. :p Is it me or Wikipedia who is wrong?
Is the dbpa dsp the same as audioinfo?

I didn't look at the wiki, but if you interpreted it and reported it correctly, the wiki is wrong.

tahel
10-06-2010, 12:02 PM
I didn't look at the wiki, but if you interpreted it and reported it correctly, the wiki is wrong.

Since you haven't bothered to check the facts before making bombastic statmenents:

The article talks about dBFS, not pure volume.
dBFS (Wikipedia (http://en.wikipedia.org/wiki/DBFS)) is dB Full Scale, or relative to the maximum of the medium. Since max is at 0, all measurements will be negative.

And I see nothing wrong with letting a system like ReplayGain state the average dBFS value instead of the adjustment value. Then the playback software could make it's own decisions on how to adjust volume. A car player might then for instance adjust to a more even level than a HiFi that instead could leave in some differences.

I've now learned how ReplayGain actually does it. But more due to the constructive input of Spoon.

garym
10-07-2010, 07:08 AM
Since you haven't bothered to check the facts before making bombastic statmenents:


I did not intend to make a bombastic statement so I'm sorry you interpreted it that way. But the original statement you made in regard to replaygain reported adjustments was in fact wrong. I didn't need to read the wiki you referenced because the statement you posted here was clearly wrong. I wasn't attacking you personally, just clarifying that the statement was wrong for the benefit of other readers of this thread. And the main reason I posted was to point out an additional program that might be useful to you in sorting on replaygain values....i.e., I was trying to be helpful.

edit: and to be clear, I'm talking about the statement regarding replaygain adjustments and difference from zero being wrong, not dBFS differences, which are negative in reference to zero as you indicate.