RomRaider

Has anybody reverse engineered the STFT logic in so far as being able to pull out the tables to tune things like PID gain, damping etc ?

Trying to tune out some idle hunting issues... it would be helpful to turn the gain down some what...

That will help big time on the mystery hesitation that disappear when you switch to openloop as well.

But for your case have you tried to adjust the injector firing? if your idle is between 2 different firing angle values it will hunt.

I might have found it, a long time ago actually, but I haven't made any effort to verify it.

The logger definitions lead to the AF Correction value (SSM parameter 3, extended parameter 81), and the closed loop fueling target (extended parameter 121). There's a function that refers to both of those and does a bunch of math. If you import this IDC file...
https://raw.githubusercontent.com/Legac ... WC522N.idc
...and look at the code at address 0002BAF2, I think that's where the PID stuff is.

The function starts by pushing a bunch of registers onto the stack, then it loads a pointer to an array of floating-point values (starting 4 bytes after the AF correctionv value), then it does a loop which I think is adding up the last 15 lambda values from the AF sensor, weighted by the values in the floating-point array.

I'm not clear on what happens next but there's some speculation in the comments.