Zitat von
hap
Is this new? (In other words, did I accidentally break it when I added Simultano?)
No, this problem already existed before.
Zitieren:
Or, is it only when you exit MAME without pressing the Stop button first? If so, then the solution is simple: for the newer Saiteks with NVRAM, always press Stop to power-off before exiting. Also applies to RISC 2500 for example.
Aah, that's the answer to this riddle - I didn't know that the Stop button has anything to do with the NVRAM!
So I've now made many tests with all these Saitek engines (with NVRAM), and actually only the newer Turbo King (ver. D) (tking) shows this problem, for all others it doesn't matter whether you use this Stop button before exiting or not (also the RISC 2500 doesn't need it, as you wrote).
But also for this Turbo King (ver. D) it's NOT necessary to use this 'Stop' EVERY time - it's enough that this button has been used only ONCE before closing, then this problem disappears (also for all following starts).
I guess that this Stop button (after used at least once) initializes any byte(s) in the NVRAM to make it 'valid', and only then the NVRAM is used or read correctly by tking in following starts.
Indeed a very confusing and inconsistent behaviour of some Saitek devices, which otherwise are very similar or almost identical.