I suspect at least part of this has to do with the pots’ effective curves and dead zones in the mixers’ firmware.
https://community.enginedj.com/t/slightly-enlarge-left-side-of-sweep-effects-knobs-deadzones/
Think of it like a joystick axis for a flight sim: you don’t want the joystick causing any input when there’s no pressure being applied and you also don’t want the effect to be excessively-large when you’re just applying a comparatively small amount of pressure. In the case of the Sweep FX pot, replace pressure with distance from center. I keep harping on this stuff, but on digital gear there needs to be certain credos for these sorts of variable control inputs:
-
There must be no input when no user input is applied. If inputs are being applied when the user isn’t actually moving the control, you need either better parts or better software reading and utilizing those parts.
-
All movement inputs must be detected no matter how small. If you can move it a tiny amount and that didn’t pass some minimum threshold, you’re not using a good enough pot or ADC for the pot for that application.
-
And the control’s sensitivity must have good usability in the curve shape at the foot and high enough amplitude possible at the shoulder. In the case of the Sweep FX pot, the foot of the curve is around the 12 o’clock and the shoulder is to the far left & right. In the case of a jog, it’s in terms of velocity, with the foot side of the curve being no rotation. You need that area at the bottom to be gradual and allow fine low-level inputs, but you also need a high enough maximum possible. That last one is particularly important for controls where velocity is being detected. Preferably, the shoulder will be so high that people would rarely ever turn the jog fast enough to reach that limit.
Analog-like response from digital gear. This applies universally to knobs, jogs, faders, etc.