MAP sampling: what's your timing ?
Turbo Mini Forums

You would not ? Avoid TPS ?
I'll try. With a small manifold that could work.

std 998 A+, g295, MD266, RHF4, 109hp @0.8bar/5400rpm

I like to have as much information as possible available to the ECU so map and TPS as inputs. But if you approach from the perspective that, without knock sensing you may end up with a safe retard of 3-4 degrees anyway, then the picture starts to look slightly different.

Hopefully next year will see the B&G ion sense circuitry available for a play. If it works that could be very interesting, especially on the A series as you can have different advance inner and outer cylinders to compensate for any mixture issues you have.

I've played with the advance on my N/A 1000. Listening for the know through 2 knock sensors connected to a modified walkman. The engine is very unsensible to the advance, it pings say at 45, but setting 25 or 35 does not change things, at least I can not feel difference. I did not do measures, but humans are usually sensitive to the torque changes.

With such a poor chamber design as on A-series the combustion cycle dispersion is high and I doubt it's possible to set the optimal advance.

I believe there is more to gain with a good engine cooling design, otherwise cyl 3 and 4 are always hotter.

std 998 A+, g295, MD266, RHF4, 109hp @0.8bar/5400rpm

If a 10 degree change in advance doesn't change the power then IMO there is something wrong somewhere.

But happy to be proved wrong.

std 998 A+, g295, MD266, RHF4, 109hp @0.8bar/5400rpm

I changed the code. Now I'm sampling every 1.2 ms and calculating an arithmetic average value for every intake phase (excluding overlap period).

std 998 A+, g295, MD266, RHF4, 109hp @0.8bar/5400rpm