I'm betting these are not truly-separate recordings. The only way to get effective "subtraction" is to invert a copy and mix.
For example, if you record yourself saying "hello" (or record a guitar part) then make a copy, invert the copy (or invert the original), and then mix at the same volume level with no time-shifts, you'll get complete subtraction and total silence. If you change the level of one track, you'll get partial subtraction.
With a time shift you'll get comb filtering (addition and subtraction at different frequencies) and in some cases that can lower the volume too. With a time shift, inversion will give you "different" comb filtering than non-inverted mixing.
However, if you record yourself saying "hello" twice (or record the guitar part twice), invert one recording and mix, it will sound exactly like regular mixing without inversion.
Quote:
I raise them both to a level that the rhythm guitar can still be heard the master volume peaks at around only - 16
|
If it
sounds OK I wouldn't worry about it. You don't really
need meters during mixing. Mix with your ears! Meters are important during recording, mainly to make sure you're not clipping your ADC.
And they
can be useful during finalization/mastering, but IMO a loudness scanner that analyzes the whole file is better and more accurate than watching a meter...