I’m attempting to mix a song and am trying to create a sense of depth.

The delays are working well as are the reverbs.

However, my question is more concerned with EQ’ing for a sense of depth. Instruments that are “right in your face” are the brightest, and as you move back (depth) in the mix the brightness falls off – at least that’s my understanding.

Question: what frequencies should be dropping off? For example if run all my drums in a group, place a shelving EQ filter on them, what frequencies should be cut?

A more involved example: if I “imagine” 4 layers of instruments (vocals first and closest, guitars, bass and keyboards next and 2nd, drums 3rd, and finally strings in the back) should each “layer” all be grouped and experience a similar EQ cut within the layer? For example: the strings, which are furthest back in “distance” have a cut at say 6 khz, the next nearest layer, the drums, have a cut at 7 kHz, the 2nd layer cut at 8 kHz, and so forth?

How is it decided what frequencies to cut or to “suppress” to achieve depth?

Is there some sort of “rule of thumb?”

If you are visualizing a mix in 3 dimensions, I think of EQ as more of the vertical, Y dimension.  Your time based effects (delay and reverb) and the wet/dry balance of those, will give you more of the front to back, Z dimension.

I already wrote a brief article on this as it applies to EQ.  It’s called 3D Mixing and the Art of Equalization.

I seriously don’t know many engineers who use EQ to bring something more forward or move it back, unless they are doing more of a special effect.  If you want something to sound distant, that’s usually achieved with reverb and having the wet/dry balance set more to wet, depending on how far back you want it to sound.  The diffusion of the sound through the reverb will also change the overall equalization a bit just by the nature of the effect.

However, I guess if you were going for something really far away, you’d want to roll off the higher frequencies as they are absorbed easier by objects, and usually don’t travel as far as the lower frequencies.  You’d just have to play around with a shelving filter to figure out what sounds best for that particular instrument and song and the desired effect.  There are certainly no rules of thumb on this.

In reality, with today’s very loud and bright music, it seems most mix engineers are boosting the high end on pretty much everything, rather than cutting it.  But, it’s certainly good to have some contrast, and cutting high frequencies on instruments where you don’t really need them makes space for the more up front instruments to cut through.

With the EQ it’s more of making sure there aren’t too many things trying to occupy the same frequency range at the same time, and figuring out which frequency range is the most important for each instrument that is playing at the same time, anyway.  If you’ve got a good arrangement to start with where instruments with similar frequency ranges are playing similar parts at the same time, then you have a lot less EQ work to do to make everything fit together.  It really all boils down to the song arrangement and good musicians that know how to stay out of the way of each other… but, I digress.