Traumflug

> You're the best person do decide that Best decisions are based on evidence, performance measurements, code size, such stuff. And by no means I'm an expert in every detail...

> I just created another branch. Thanks. I see so much additional code. Code is expensive. Less code to achieve the same goal is always better and so far I...

One thing a bit unpleasant about temperature readings is that it causes a whole lot of interrupts on AVRs. For every reading recognized by temp_sensor_tick(), about 25 interrupts happen. That's...

Yes, a moving average is a good thing. To my experience it doesn't matter at which frequency it happens, because noise is random. But it has to happen often enough...

> My idea on the STM is to start a new conversion just 10ms or whatelse before the heater/pid/temp-tick. This would be awesome and saving a lot of interrupts on...

No problem, already restored.

> 0ms ADC start for channel, like freerun On AVRs, there is no freerun for multiple channels. Mimicking one, like it currently happens, causes a lot of pointless interrupts. One...

> I say we have achieved our initial goals. Looks like you're looking forward to move this bunch of commits to _experimental_. Excellent! A few things which help to have...

Yes, these three can go. They were on _experimental_ for a day, but turned out to break Configtool. Now they're back on _configtool-refactoring_ (first three commits there).

> HAVE_ANALOG_TICK to HAS_ANALOG_TICK. Or what do you think? I think it doesn't matter. Neither "HAVE" nor "HAS" is used anywhere. Is it necessary at all? Which situations have no...