Agemixer (07:29:37)
i give up

Agemixer (07:29:08)
Problem, i don't have any idea how to timing should be in that case, what is allowed and well... it turns out of module programming and DMA buffers in that case. but. it.will.be.fastest.

Agemixer (07:27:16)
what the heck, c64 module port has a DMA signal line for 1MHz transfers.

Agemixer (07:26:26)
buffering does not speed up things in that case.

Agemixer (07:24:37)
...of 16bit 44kHz raw sample.

Agemixer (07:23:52)
while there, cycle exactness is important. Problem: bad lines....ok. turnoff the screen for sampling then!

Agemixer (07:21:29)
..which then leads into more hardware to get solved like that xD the syncing to c64 reads of course. the 1MHz line must be read, and once in 22 cycles it should reset the counters itself.

Agemixer (07:18:01)
just read the ever counting counter into eternity... and fixup the delta values later. Problem! The c64-wise 22 cycles must be then hardcoded to the sampler device somehow. wellwell. 2 extra counters needed. and the first initian reset line of course

Agemixer (07:14:26)
how to no reset need? Oh yes! no resets at all.

Agemixer (07:13:23)
if cia no need reset then theres a chance for loop nicely at 44kHz (inx, bne for loop of 256 samples)