I mounted the circuit into the final enclosure, this thing is going to look awesome when it's finished, however...

The ADC is acting weird, it seems to be working, but the value also seems to trend towards 0, even though the output voltage is steady. There are a few things in the firmware that I need to double check before I start to get worried about it not working, as the biggest suspect is the acquisition time. There is a capacitor in the uC that gets charged to hold the sample data. If I am not waiting long enough on each pass, it could be a bit less charged, hence the progression down hill. Of coure acquisition time is specified in clock cycles, so the fact the clock is a bunch quicker now means I probably need more cycles of time...

The buttons are acting weird, I have only tested the knob, and it's really temperemental, there is not a lot of difference between clockwise and counter clockwise. In fact both seem to register as CCW movement, with the occasional CW thrown in. This is suspect is again due to the processor speed, but may be related to the difference in how interrupt on change (IOC) works on this processor. The rotary encoder has two switches configured to give quadrature output, so I use an interrupt on one of the switches to detect movement, and then I sample the other switch to determine direction. One the previous processor the IOC could be specified to provide an interrupt on a positive and/or negative change and I was using only one polarity. The new processor only knows "it changed" and I don't remember if I changed to code (and if I did _how_ I changed it) to handle this. I will have to revisit it...

The display is not showing what the code says it should be. I was partially expecting this, there are a whole bunch of timing specific bits in there, which I suspect are now off based on the faster processor. I will attach the logic analyzer and see what it going on with it.