mastering and volume

[crosstudio]

when mastering an EP wherein you have 5 songs of varying tempo and attitude how best to set the volumes such that the tracks run well together?

what i have been doing is multiband-compressing and maximizing the individual stereo mixed tracks as necessary, then using the volume faders to adjust the relative volume between songs.

since we are all so concerned with slamming the meters from stem to stern for every track how do we get the volumes of the individual stereo mix tracks to lay together for a cohesive unit?

so i started thinking.

maybe i should take the 5 songs (with no eq, compression or peak limiting) and just play with the volume levels until i get them to gel and take note of the varying volume levels.

then

set all of the tracks back to 0db, start adding on the multiband compression and peak limiting to get the songs blazingly loud like everyone else in the industry, then adjust the gain on the peak limiting such that the volume level deltas between the individual songs are preserved.

for instance, lets say track 1's volume was set to 0db while track 2 was set to -2db when i was just trying to get the relative volumes correct.

then

once i get the loudness maximized just right for each track, i would back off of the gain on the maximizer by -2db for track 2 to make sure that the relative loudness between the 2 tracks is preserved.

is it possible to love and hate something at the same time? there's no such thing as bad sex, but bad music is limitless.

 

[Shailat]

If you are already going that way, then I would do it the opposite.

Load the songs into a burning program and set the levels in there. Change the gain by 1dB untill you get the average level you want. Then use a mastering compression to actually get to the levels you had set in the burning program. Again - don't make the gain changes in the software and then burn ! just mock it in the software but the actual gain should be done in the compressor. This will eliminate any overloads.
Remember - If mixed correctly, you'll only want to tame the levels
and bring up the overall level by 2-3dB. You don't have to really slam them to wazoo. By now they should sound pretty loud and healthy. But if you feel it's needed you can boost them more. Some people would normalize them. Your decision.


As for the cohesiveness.... check each single song by checking the start middle and end and decide on a average level. Then find a single song (or two) to use as a "marker" and compare them all to that one. Usualy it will be a song with "higher" levels. Songs that will seem unaturaly loud like without drums or simply quiet songs will need to be lowered.

Before the compression you might have to EQ the "attitude" a bit to get a more cohesive picture but be extra carefull here.

 

[masteringhouse]

when mastering an EP wherein you have 5 songs of varying tempo and attitude how best to set the volumes such that the tracks run well together?

One of the things that may help is to find some good metering software. Waves and Spectra Foo are very good. Spectra Foo uses Bob Katz' K-System which is becoming popular among MEs in order to set reference levels.

The dominant frequencies of a piece also play a role in the percieved loudness due to the non-linearity of our ears. The Fletcher-Munson effect explains this. See


http://www.allchurchsound.com/ACS/edart/fmelc.html

Assuming the frequencies are pretty much the same between songs, you will want to use some compressors for bringing up the average level and others (most likely limiters like the L2) for control of the peaks.

I wouldn't just throw a multi-band compressor on as a rule of thumb. Use this to control stray frequencies that an EQ doesn't take care of properly. For example a bass track that "pokes out" when certain notes are played, but where you feel the balance on the bottom is fine otherwise. Also possibly for things like de-essing if you don't have a dedicated unit.