David Buezas
David Buezas
The 9kB flash consumption from TMC is crazy considering that it only needs to send some bytes over spi or serial once. I took a quick look and the library...
Looking forward to use this! BTW, have you considered using the median (median != average) instead of > _then the values fartherest from the middle will be removed_? This is...
I see now, I had missed the fact that the average os used after removing outliers, makes a lot of sense. Thank you
[deleted wrong equation, will write down later] But we need to confirm the classic motion system still works, and likely find a compile time alternative. The easiest way would be...
I agree, the thing is that the same hal applies to many MCUs in the stm32 case, all would need to be handled correctly from the get go
It isn't? I think one would need a runout block too then
We could call `ftmotion.plan_runout_block()` after synchronising there, that would add enough time for all smoothing and all shaping to finish stepping.
Why is setting the position important at all? That position is not even used inside ftmotion. I think it is all about adding a runout block so all echoes and...
I don't get why this would make any difference: ```cpp AVR_ATOMIC_SECTION_START(); count_position = planner.position; AVR_ATOMIC_SECTION_END(); ``` The layer shifts come from axis synchronisation suddenly changing how much it delays axes,...
I think what we could do is scheduling a runout block after synchronising the planner, and then waiting for it to finish.