[info] rusEfi Scheduler Performance & Accuracy

[ptnicu]

If CPU is busy, is it possible to put 2 STM32 by ECU ? They are cheap, and some values can be send via canbus, I don't know how many resources take canbus, but there is not a lot of important value.

The CPU is not busy. If it can do this https://youtu.be/7yXxhvKmVb0 it should generate ignition timing without a glitch, and a lot more.
Multi-core system present other challenges, add complexity not needed to the system.

The RAM consumption is at critical level. There should be some optimization in that area.

[Gepro]

Okay
Thanks,

There is some H7 without dual core and with a lot of ram.

I see that microrusefi come back from F7 to F4, i don't know why, but the ram is the same, so the frequency is certainly useless.

The power of a stm looks completely crazy compared to others.
And the frequency of an engine is stupidly slow, a V12 does 6 sparks by turn, so at 10 000RPM it does 1000 sparks by second, that's is only 1KHz...
I am certainly wrong to compare this, but I'm not a software guy.

That is really nice mck1117 ! Great job <3

[AndreyB]

My preference would be to pause the discussion here until more participants have a chance to contribute their first code pull request

[mck1117]

5800(rpm)=96.667(rps)=0.0348(deg/usec)
.0348(deg/usec) x 2.3(usec) =.08 (deg), twice as good as you're claiming and quite excellent!

Are there some points hiding further to the right? Or is my math wrong?

there are a few points down in the tail:

image.png (11.98 KiB) Viewed 3147 times

But you're correct, 99.8% of all measured samples were in a 0.1 degree wide window, with only 0.22% outside, the worst of which was 0.16 degree later than the earliest.