At your suggestion, I took a day off.
There are certainly options, but it sounds like there is plenty of time before the FW3A will start production. The most straightforward option is to splice thermal regulation into NarsilM. The thermal code from the D4 would probably work with only minimal changes.
I’m probably going to try improving the thermal regulation algorithms soon though, because I had some new ideas.
And although it was “just some thoughts” a few days ago, now I’ve got a completely new from-scratch firmware almost done. So that might be an option. Or I’ve been wanting to try the FW3A driver’s light-sensing capabilities, which could lead to something interesting.
Things can get bad with too many cooks in the kitchen though, so I’m looking for guidance from Tom and the other FW3A chefs before I try to put anything in the oven.