You might recall I wrote about the same subject to you a few month back, before releasing LSOS 1.3. I also came to the same conclusion, that something must be quite inefficient (or at least an overkill for our purposes) in the MPU code, at that time I measured 18% of code space alone taken by the MPU related stuff.
Ah, indeed you did. I do recall that now, but didn't get around to actually looking at your new code. I'll have to do that now. Thanks for the reminder and the work you've already done on this matter.
I'm not against offering a DIYino with a more powerful cpu, but it will mean a price and size tag. The size impact I might be able to counter, but the price will go up by ~10$. To go down that path I first need to know if people are really willing to pay for the extras. If we stick to the same architecture, the main advantage of a more powerful cpu would be that all 3 blade types could be united in one, selectable by menu, but who knows if there is a popular use case for that?
Good points. I think the value proposition is a big one; you don't want to price the board super high.