Mixing Exercise

Posted: Mon Jul 07, 2008 8:48 am

This weekend I did an interesting exercise with one of the songs that I am mixing, and learned a couple of things. I also am curios as to what other people think of this technique.

This is going to sound crazy, but stay with me. In the end the results were good. I tried a different method to find a frequency range for each part to sit in the mix. I started off with just the lead vocal, and put a 24db/octave high and low pass filter on it. I narrowed the band down until I found a small range that conveyed the vocal. It sounded like it was coming out of a little old AM transistor radio, but the vocals were clear. I added one track at a time using bands that had no overlap until I had all the tracks in, and the whole frequency spectrum covered.

The results were as follows:
Bass 20-150
Dobro 150-300
Male Vocal 300-600
Female Vocal 600-2k
Guitar 2k-6k
Mandolin 6k-20k

Even with the hard EQing it didn't sound bad like this. To make it sound good I ended up using these frequency ranges for the instruments, but not as band passes.

I gained a couple of things from this exercise:

1. It reinforced that there is more than one good way to mix a set of tracks.
2. There is a difference between EQing to accentuate parts of the track, and EQing to mix and fit the tracks together.

Posted: Mon Jul 07, 2008 9:00 am

Can you post the song, or a sample of the song?

Posted: Mon Jul 07, 2008 9:33 am

I wish I could post a sample, but can't. Mix is so subjective that this post may be hard to comment on. It's not a critique of this particular mix that I am interested in, more just on the approach to the mix.

Has anyone else done it this way before?
Are their pit falls that as a recording engineer you see with this method?

If there is more interest, and I have time, I'll do it with one of my own works that I post to the internet.

Posted: Tue Jul 08, 2008 8:37 pm

I'm not interested in critiquing you, I'm just curious to hear how it sounds. No dice?

Posted: Wed Jul 09, 2008 5:46 pm

This is a little snip of the chorus.

Here we go:

Be My Lamp Lord

Posted: Wed Jul 09, 2008 9:35 pm

That sounds a lot better than I expected. Maybe I'll mix one of my songs like that and see how it turns out. Thanks for sharing!

PS: I kinda like ol' country gospel.

Posted: Thu Jul 10, 2008 7:22 pm

Thanks for the idea, I am going to try it on my own project shortly.

The style is a cross between rock and funk, very similar to what the Red Hot Chili Peppers might do.

The instrumentation is as follows:

Drums
Latin Percussion (Shakers, Clave, Guiro, Djembe, Cowbell)
Bass Guitar
Acoustic Guitar
Electric Guitar
Old Electric Organ
Lead Vocal
Harmony Vocal
Hip-hop Vocal (kind of rapping)
Shout Vocal (not screaming at all, just short staccato accents)

I know there are many more instruments in my example than yours, but here is my initial thoughts, before I try it out:

Drums: No clue, wide range from bass drum to cymbals
Percussion: Probably a higher frequency, at least above 1000
Bass Guitar: between 50 to 150
Acoustic Guitar: chunky meaty sound, so maybe 150-300, also higher
Electric :i'll try yours at 2k-6k
Old Electric Organ: somewhere maybe the 6-20ish range
Lead Vocal: male, so i'll try yours at 300-600
Harmony Vocal: male, but its higher so i'll try your female range
Hip-hop Vocal: i want a really dry sound, maybe 1k-3k
Shout Vocal: not sure anywhere from 500-3k

my biggest problems are the range of the drums, how the guitars sit with the organ, and the variety of vocal tracks. Taking a look at what I have given you so far, are there any potential roadblocks ahead?

do you think this concept can be applied to panning as well? i mean start with the vocals in the middle, and then work your way out, as instruments become less "important"?

when I am done tracking, i will try and get something posted! thanks

Posted: Fri Jul 11, 2008 10:47 am

Thanks mhutch.

ThirdBird,

There was a recent post from YellYo talking about using panning to separate things in your mix. http://www.recording.org/ftopict-47600.html There was some good stuff in there. The method you mentioned is what I do for the most part.

Another thing to consider is that the background instruments need less frequency space. I think of it in fractions of octaves (like a graphic EQ) on a 24 band eq, the lead vocal may get 3 or 4 bands in the middle, but the dobro fills may only get one band.

I have found that rolling off a little high and low frequency helps push them back, and keeping the front and center tracks pretty flat, and cutting the background instead of boosting the foreground helps.

Posted: Mon Jul 14, 2008 2:35 pm

Thanks for the link!

This topic sponsored by: