a little bit maybe from my personal observations regarding the new heater tuning algorithm:
We have here a small anycubic-i3-mega that got as new mainboard the duet-pcb, it sits in a wooden chamber with the standard heatplate used as chamber heater and 1 additional fan in the top of that chamber to "cool" the chamber by sucking out air (there are some holes in the bottom plate of the chamber to let new air in when that happens)
When using the new heater tuning algorithm, I leave the door of the chamber open and put the fan of the chamber to max rpm, BECAUSE first I did it the other way round with door closed and chamberfan off and after tuning and saving the tune with M500 I got a heater fault manually heating up, because that time the door was open and the fan on and that was too much for the state it was tuned in (heating too slow/temperature rise too slow).
So long story short: I would recommendate from my observations that you tune with "hardest" surrounding conditions for the heater. With hardest what I mean is, that it has to work hardest. If you tune it with the easiest condition, you might later get a sporadic heater fault when conditions get harder.
A 4 point mesh will always be flat as the controller connects the 4 points with straight lines. It does not represent the true shape of the bed.
I would be VERY happy with a bed as flat as the one shown with 36 points !!!
@dc42 I am running a chimera and I think my main concern is having the x axis the closest. I set my nozzle heights after I get dialed in. Its been working really good. It also seems like if I had a single nozzle it would be better if I could get it super close in 1 axes and close in the other. Rather then really close in all axes.
@lui2004 On the basis that my extruder stepper motors have about half the the torque of yours and I run them at 80% of max current then try 700mA and see how you get on - you'll soon know if you need to use higher current.