On 12/28/2012 05:05 PM, Fred Cisin wrote:
The NEC 765, and presumably all of the
"compatible" alternatives do a
"reset" on seeing the index pulse. That makes many "crowded"
formats,
such as Cromemco, TRS80, etc. difficult to read on PC.
If feasible, PLEASE consider having that overrideable in software, so that
it is possible to read immediately after the index.
WD 179X has a "track read" command.
NEC does not. "read multiple sectors" is nice, but it doesn't let you see
intersector gaps, address marks, and MFM without "IBM/WD" sector headers.
Is it feasible to implement?
Yes, it's annoying to be working with a WD17xx-created format without
the IAM and have a 765-type FDC "go blind" during the "PLL sync"
period. There was a NEC uPD7265 that omitted the IAM for so-called
"Sony Mode" recording, but those are exceedingly rare.
NEC has sort of a "Read Track - read multiple sectors without looking at
the IDAM" command; it starts after the index and reads subsequent
sectors starting at each DAM. You can have an unreadable IDAM and still
recover the sector data if the DAM is readable.
You also can feed the 765 a "Read Track" with a phony datalength
specified. The 765 will suck up all of the data after the first DAM
after the index until the transfer count is exhausted. Thus, you get
gaps, AMs, headers, all "in the raw", meaning that the 765 does not
resync during the transfer. Great if you're using original floppies
made using a Formaster that writes an entire track in one swell foop,
not so much for tracks where the sectors are written individually. The
WD17xx and 27xx type controllers, OTOH, resync at every address mark
during a "Read Track".
Harvard Presentation Graphics used this technique to read their "HGC"
stashed in the gap bytes between sectors.
Also, consider implementing some of the 82078 added commands,
particularly "Format and Write" (can make copying go very fast) as well
as the PM functions.
Personally, I'd do the board as a USB card, given the difficulty of
finding ISA-equipped motherboards or dealing with the nuances of
PCI/PCIE, etc.
You should also be aware that Andrew has been testing prototypes of an
ISA FDC that can use either the I82077AA, PC8477 or DP8473 chips.
--Chuck