After a few more minutes of GGXX #Reload, some things set in.
-Rapid movement still feels rather weird at the moment. But circle motions actually work pretty well, once you get the hang of the actuation zones. I noticed problems with diagonal jumping, so I tried to compensate for that...perhaps a bit too much. A double-QCF move attempt kept giving me a DP move. The trick? Move the stick just a bit higher so it's resting at 6 (right) when I hit the button, not 3 (down-right). I never did quite notice this 'til I started to listen carefully for the distinct microswitch clicks.
-The angle of my arm to the stick affects the feel a lot. Come to think of it, the general ergonomics of how I'm holding the stick in the first place and how it's positioned do seem to have a drastic effect on the feel when trying to pull off those motions.
Speaking of octo-gates, word is that what makes the JLF and the LS-56 so different in that regard is that the former actually increases the stick's travel relative to the square gate, while the latter restricts it instead. Is that true?
I'm also still left wondering why there aren't different actuator bits for JLFs and their clones. If people are resorting to electrical tape and heatshrink tubing, why not offer multiple actuators like there are multiple restrictor gates? At the very least, a JLF could be tightened up like an LS-56, but the LS-56 can't be made all loose like a JLF.
As for the LS-32...I'm not sure. It's more common, but I might like an octogonal gate better...yet I manage with square-gated Neo-Geo CD pads when playing shmups on said console. Perhaps it depends on what's too loose, what's too tight, and what's just right, which isn't easy to tell without actually getting a hands-on evaluation.
(Speaking of which: Why isn't there a sink list for these things? Something like the Switch Try Board, except they're arcade joysticks!)