On Monday, May 8, 2023 at 6:56:33 PM UTC-4 Mac Doktor wrote:

I recall reading on multiple occasions that a SCSI driver was a non-trivial 
exercise. Sort-of like "if you don't appreciate how non-trivial it is don't 
even bother".  


Yup. It was bad enough when there was an actual controller chip, but at 
least the chip handled some of the handshaking internally. Doing it with 
parallel ports alone was *quite* exciting. Fortunately the original design 
only needed* to support the Xebec S1410 SASI/ST-506 bridge. Even that was 
fraught with peril - the SyQuest SQ306 drive used a servo "wedge", so an 
incautious format operation would destroy** the cartridge servo information 
and render the cartridge unusable. To make matters worse, the field test 
drives had matching cartridges - interchange between drives didn't yet 
work. Ah, the bad old days...

* Of course, right after release a customer showed up wanting to use a tape 
drive
** Xebec eventually released a new S1410 firmware EPROM that prevented this 
- if you happened to have a controller with that EPROM  (it wasn't the 
default)

My company had lots of experience with oddball storage - we had a Vertimag 
5MB(ish) floppy disk prototype. And I was involved with the Evotek ET-5540 
disk drive. That was "interesting", but that's a story for another time. We 
eventually went with CDC Wren drives. DEC (Evotek's other large OEM 
customer) had a bit more inertia and had to scramble for available drives - 
that's why the DEC RD52 might be a Quantum Q540 (more common) or an Atasi 
3046 (rather rare).

-- 
You received this message because you are subscribed to the Google Groups 
"neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/6862d798-a2b9-486a-ba29-9d68fb21235dn%40googlegroups.com.

Reply via email to