i find learning polymeters / polyrhythms a bit of a strange concept, it's like trying to calculate a groove in advance, how do you know if something will groove until you play it / hear it? it might be a goal in itself, having something technically challenging, but for most such pieces while some changes might seem random they all fall within some higher meta pattern. though the distinction might not matter that much because our brains will find patterns everywhere anyway.
from that pdf @jasonw22 posted:
"A tablature notation system has been used to make the interlocking of the different drum parts visually understandable. This is important because, even though each piece is written under a time signature, e.g. 6/8, the actual playing of the piece will tend to fluctuate somewhere between 6/8 and 4/4. This is due to an entrainment effect that occurs between the drummers.
Entrainment may be considered vital to liturgical drumming. The Dutch scientist, Christian Huygens, published his research on the subject in 1665: “If 2 rhythms are nearly the same and their sources are in close proximity, they will always lock up, fall into synchrony, entrain.”
i think we are capable of getting entrained by any rhythm really, i guess what i'm trying to say is - the techniques described can serve as a starting point (as an example, i often will cut one loop in ableton shorter and line it that way so that it starts shifting against other tracks) but then it's up to you to discover that inner logic in complexity and follow it.
as a tangent, i've also been thinking about changing the grid itself, instead of being a rigid thing imagine your typical xox patterns on a grid, but this grid is a piece of rubber and you start stretching it... so you still have the grid structure, but it's elastic now, it's just a way to keep beat to beat relationships. as one example, i'm fascinated with how different implementations of clock multiplication (which is based on you having to guess when the next beat will occur) deal with tempo changes - intentionally confusing them can make for some interesting results. another thing i've been trying to do is start with different loops that seem out of sync but at some point somehow come together and reveal the meta pattern.
and i wouldn't completely discount 4/4. there is such a hard cold hypnotizing beauty about it.
since we're talking about polyrhythms i should mention orca again, since polyrhythms are the very foundation of it... as an example: