i'm making a little progress, but i didn't realize how similar so much of this is...

there's really three parts, with the third doubled and split l/r, but the drum is panned hard to the right so the effect is not commutative. the sound dominating in the headphones is on the left channel, with the right channel adding subtle levels of colour.

the first has some reverb and a little delay, the second is a heavy chorus and the third/fourth are a kind of heavier shadow delay, in that they have almost no dry signal. but the colouring is dramatic.

so, what i'm noticing is that each combination, if held static otherwise, is only minimally different on the right channel. which might seem like it would reduce possibilities - but not all the way down to 27 - but in fact actually requires very carefully listening to isolate the differences.

but i am making progress. splitting the set into three really helped a lot.

this round robin thing isn't working; i'm not comparing apples and apples. rather, i'm going to have to break it down into sets: some renders are "bright", others are "dark" and still others sound very compressed, almost "tinny". comparing them against each other isn't producing a total order. once i can break it into sets, i can compare apples and apples and then figure out which one i prefer as a base sound.

i can't isolate it as a single factor, either. that's why i couldn't get it right. if i take the model down on one and up on the other it will have a similar effect as if i do the opposite thing....