Would you be interested to do some discussions about ideas or extensions in future?
I'm currently waiting for some parts to arrive from China, so that will take a while. As I'm waiting I studied the low-level ARM code on github, which looks very good to me. Not sure how "filled" the 1MHz cycle is. I would hope there are still some spare cycles left when moving to assembly level optimization of that inner code.
One thing I was maybe thinking of, is if the arm could optionally run a 1541 emulator instead of a cartridge emulator. I'm unsure if 168MHz is fast enough for that.
Also I was thinking if it would be possible to avoid the internal flash by using page swapping of SDcard file blocks to the 192kB internal RAM of the ARM. Where a page-swap might be done using the DMA pin to stall the 6502 while the sdcard is accessed during a bank switch.
Ah these ideas. Can take a lot of time to implement though.