also @faultier re: hatz
here is a terrible explanation of my hat processing:
-start with a pretty clean closed and open hat sample (I sample a lot from old roots 7" records), program basic downbeat and counter rhythm in midi
-bounce to audio, duplicate several times
-I always pitch the audio around (transpose to keep things in time) on one or two of the layers
-At this point its really about distorting the various layers in different ways to fit the vibe, my go to are the drum room presets in Nomad Factory's Bus Driver, then push the output gain up on the compressor to volume match. As the textures and levels change, try to reblend the layers as I go; usually adjust the pitches of the layers quite a bit here
-After im sort of happy with the tone, I'll usually introduce sidechain compression to one or all of the layers. The sidechain source sometimes comes from one of the layers, the kicks, or a muted click track (whatever works for the riddim) I dont really know shit so I just adjust to taste here. One thing that works well with this compression is to start moving some of the layers to wide stereo via M/S EQ or delay. Really can get some nice subtle trippiness when the stereo image is being sidechained improperly haha.
-Print the whole group to a single stem or two stems (closed and open hats). The first is nice because you can get some strange interactions between closed and open hats. I usually do separate stems, then a combined bounce) Then I usually chop the stem(s) at a later date while im actually using them for a tune.
-I often run groups of stems to cassette after the tune has been written too, usually overlay those with the original sound, typicially with another sidechain compressor and/or gate going to remove the excess tape noise in between hits and add extra moves.
I think the key is really changing the pitches on the layers and subtle use of distortion+eq.