Two days ago I managed to write a proof of concept that runs on C64 emulators and dumps the whole contents of a disk sector by sector. This time around I am using a custom sector read approach, based on lft’s on-the-fly GCR decoding routine.
I didn’t manage to fix the communication between the 1541 and the C64, as I was getting misaligned data, but I was able to lower the dumping times for a whole disk down to less than 30 seconds. Now that might not seem impressive compared to 22 seconds taken by WarpCopy64, but it’s my first iteration.
I am going to have to revise my calculations after I fix my transfer code over the IEC bus, but in the best scenario it might be possible to break the 20 second threshold. This would probably be impossible to do with a 1541 attached to a C64 due to the fact that they run clocks at different speed. However IECHost can be set to run at a speed that I choose and therefore it might be possible to get away by doing less synchronization and acknowledgement on the IEC bus.
Stay tuned as this might turn out into an awesome one next year.
Yesterday evening and this afternoon after work I worked at porting my DC2N5 framework to a different STM32 MCU which has more room for future expansion in terms of program memory and RAM. I plan to drop development on my other boards and focus on what I refer to as DC2N5 Low Cost version 2. I didn’t have to do much really as I already had the whole framework running on a similar MCU, but I stumbled on a couple of things that took more time than I had hoped for to get fixed.
I plan to incorporate IECHost into DC2N5 Low Cost version 2 in the long run, but I will probably focus on getting a few prototypes out without the IEC bus interface.
This one might turn out into an even more awesome one next year so definitely stay tuned!