mixing down very similar frequencies

[manzatour]

hi

i'd like to have some advice about mixing more tracks with similar frequencies.
i'm mostly able to manage this for the low end by cutting everything not absolutely necessary with a sharp eq
but i tend to have problems in the mid and hi range.
i can't eq everything in this domain because some frequencies are shared by several sounds.
how can i try to get som order there?

- assign a limited range of frequencies to every sound? sometimes it is not possible
- pan? it has only a limited impact
- try to solve it with compression? (sidechain also for higher frequencies?) it works for a track or two, then ends in a soup
- select / get rid of concurrent sounds in the composition process? i already do, but sometimes it is not possible
- multiband compressors? really not an expert... would this help?

is the key a combination of all these techniques?

thanks

[hydrogen]

change the music so the sounds don't actually overlap.

[Phase Ghost]

Panning will help. Also, changing the volume of some of these sounds will probably help the most. You don't want too many dominant sounds or it will, as you described, end up like soup.

[Phurniture]

"- select / get rid of concurrent sounds in the composition process? i already do, but sometimes it is not possible"

I don't know what kind of music you're dealing with, but why would it not be possible to change the composition? If you really must have those sound together then I would suggest the following:

1. Prioritize your sounds: decide which sound is more important than the other sound. Cut some frequencies from the other sound.
2. Volume/EQ/Filter automation: leave both sounds as is, but when the two play together automate the volume/eq of the other to cut the offending frequencies.
3. Key change: if you're dealing with two melodic elements, then change the octave or key of one of them.

Just like too many sounds with bass frequencies take the impact away from a kick/basslines, too many high frequency sounds take the sparkle away from the sounds that really need the hi-freq content. I'm sure that some of these sounds could use some low-pass filtering and still work well.

[manzatour]

thanks for the inputs
it is difficult to explain, but i rarely compose using "classic architecture" (with kick, hat, bassline, chords, melody percussion).
i use to put together strange experiments where noises take the function of standard instruments.
so it happens that the bassline is a glitching noise with prevalent frequencies in the low end, the harmony is another noise based in the mids, and there is no melody but another noise with a higher register... and so on.
i manage to solve this not too bad in the low end.
the problem comes because, let's say harmony, melody and some rythmic parts share frequencies that i can't cut, and the track is based on these sounds. so i can't replace them with something else. i do when possible, but working this way it is not always possible.
i already equalize and try to strip down conflicting parts, but i would really like to know if there are other techniques that could help me to avoid this lack of brilliance and clarity in the mixdown.
sometimes after the mastering process it does sound much better, sometimes not.
i imagine that when the mastering helps it is because multiband compressors are applied, and i was wondering if i could make something similar during the mixdown, or even before...

[trak660]

Is it possible to sidechain?

[ethereal]

Youre working too much with frequency and not enough with timbre.

[kdgh]

boost the frequencies that are next to your sweet spot freq.

example : 2khz is the sweetspot, boost 1,5khz and 2,5 khz a little. With this it 'seems' you boosted the 2k. This is the same way how your ears 'repeairs' themselves. pseudo-stuff