From: MagerValp (MagerValp_at_cling.gu.se)
Date: 2005-11-09 09:13:43
>>>>> "s" == silverdr <firstname.lastname@example.org> writes: s> And you end up using the regular GCR nybble decoding tables as s> found in the 1541's ROM but you also wrote it was faster than the s> 1571 decoding routines (I think I found them already in the 1571 s> ROM but didn't yet have time to analyse)? Or was it so that the s> 1571 alike routines would be faster but wouldn't fit in the RAM and s> the best [out of what could still fit] was to do it this way? Yes, I haven't looked at the 1571 routines, but if they have full decoding tables I'd expect them to be faster than my routine. This was the best that I could fit in the 1.25k available (roughly 256 bytes for the GCR decoder, 512 bytes for the irqloader, and 512 bytes for the nybble buffer). It's my first attempt at writing a floppy speeder so I'm sure there's plenty of space for improvement. E.g. Krill's loaders are much more flexible, and much faster. The main improvement in my routine is the partial nybble decode during read, which speeds up the decoding pass. -- ___ . . . . . + . . o _|___|_ + . + . + . Per Olofsson, arkadspelare o-o . . . o + MagerValp@cling.gu.se - + + . http://www.cling.gu.se/~cl3polof/ Message was sent through the cbm-hackers mailing list
Archive generated by hypermail pre-2.1.8.