Nico Tonnhofer

I've done this some years ago. But never finished or took it to master. https://github.com/Traumflug/Teacup_Firmware/commit/308bac5e875ca8ef11e84da6d09c02d3535694e8

Hi, check also at the same time that the pin support pwm. When not, software pwm will be used, which shouldn't work.

I found this pitfall also some years ago with Marlin on a RAMPS where I want to use pwm for some LEDs and I was wondering why this don't work....

Not that much slow as it seems. You normally print over 200°C. That are just 4 steps. On a normal binary you need more. Maybe for a print bed. Hmmm...

Ah ok. Was a little bit confused because we start with j = 1. On Marlin, they uses a lookup table to calculate the delays for the stepper timer. This...

We could save the reciprocal as fixpoint. Just made a small test with my local temptable. Any delta of `(x1 - x0) is 42. 1/42*2^13 ~ 195`(error is 2^-4). Now`(((temp...

> (error is 2^-4) This is the error between 195 >> 13 and the division by 42. Not accuracy. Just speed.

> If you're into speed, avoiding this calculation alltogether is likely the best option. Instead of running PID against Celsius values it can run just as well against raw readings....

Hopefully I'm doing anything right with the simulAVR. So here we see some numbers. TEMPTABLE_FORMAT 1 ``` LED on occurences: 710. LED on time minimum: 401.76 clock cycles. LED on...

I moved the debug led to temp.c, this was maybe wrong. I will test your idea, too.