Music: The race to be the loudest

It's all about the equipment

Moderators: Mr Awesomer, JesseMiner, CafeSavoy

Locked
Message
Author
SoundInMotionDJ
Posts: 154
Joined: Wed May 30, 2007 8:27 pm

Music: The race to be the loudest

#1 Post by SoundInMotionDJ » Thu Aug 23, 2007 11:11 am

Here is an excellent article on the loudness trend in music mastering.

http://spectrum.ieee.org/aug07/5429

The first page of the article shows a representative waveform from 20 years ago, and one from today. In editing music for couples over the years, I have seen this first hand.

In the 1980's (and before) the average level of a song was -18dB to -15dB. Today, the average level is somewhere between -9dB and -6dB. If the current trend holds, music will be on average -3dB by 2010. A change of 6dB is perceived to be "twice as loud." So, in the extreme case, if the average level of a song went from -18dB to -6dB, the song will seem to be "four times as loud" (i.e. a 12dB change).

The only "good thing" about this is that all of the dynamic range that is being used can be represented in a compressed (i.e. mp3) format. So, shortly the music industry will get to a point where there will be no loss of "quality" from the CD to an mp3.

Here is a great video on youtube that illustrates the problem....

http://youtube.com/watch?v=3Gmex_4hreQ

Here is the article on wikipedia...scroll down to the "Remasters" section for an example of the waveforms for an original song, and the remastered version.
http://en.wikipedia.org/wiki/Loudness_war

The trend to have the louder songs is one of the most difficult aspects of setting up a sound system.

If I add a compressor/limter to the effects chain, then the current songs will be over-compressed (remember, I will be compressing most songs by 12dB to 18dB to get the kind of correction that I want) and will sound "dull" and "muddy." Basically, I would be doing what FM radio does. Every song is gained up to about -6dB from line, then compressed to fit in FM's bandwidth, then played. This has the effect of turning music into mud. But, a lot of people seem to use the radio as their standard of what a song "should" sound like.

If I set the volume to be "not too loud" using a modern recording, then all the older recordings (i.e. Al Green, Aretha Franklin, etc, etc) will be too soft to really hear and dance to.

If I set the sound system to be loud enough for a song like Aretha Franklin's "Try Matty's", then it will be "too loud" for any modern songs.

If I trust the individual DJs to handle the volume, as a group they will succumb to "slider creep" and steadily push up the volume as they get use to the overall sound level. Within an hour or two, it will be "too loud" and it will keep getting louder all day and all night long.

--Stan Graves

Locked