🚀 go-pugleaf

In article <2670@decuac.DEC.COM> avolio@decuac.dec.com (Frederick M. Avolio) writes:
> In article <JWG1.89Apr24142605@bunny.gte.com>, jwg1@bunny.gte.com (James W. Gish) writes:
> > According to our local DEC salesman, there are currently no third
> > party drives available for the 3100, i.e., you might call it
> > pseudo-SCSI.
> 
> Of course this is not true.  Or maybe you were tongue-in-cheek?  You just
> need to find a SCSI drive that sticks to the specs.  We've run quite
> a few different company's SCSI devices.  There was a discussion of
> this in comp.unix.ultrix a while back.  Perhaps someone who saved the info
> could send it to you?  I am cross-posting this reply to there.

It is worth pointing out that the SCSI standard is pretty successful at
standardizing the hardware connection level (Apple excepted), but even
with CCS (common command set) the software compatibility has a lot of
problems.  Sure the sequences for reading and writing are fairly uniform,
but the sequences required required for "formatting" the drives, using
bad block substitution, power up initialization vary widely.  Also the
values returned for reading capacity and geometry seem subject to random
intrepretation.

The general procdure for writing a SCSI driver is therefore to take a
crack a writing a "generic SCSI" driver and then special case it until
it works with each of the drives you feel your company is likely to use
or that you have lying around and want to work.  

As a result, you can usually expect the same model drives, with the
same or later ROM revision and hopefully their successors to work, but
expect a new and random drive to work only somebody not only claims it
works, but can demonstrate such from taking it from the box, formatting
and using it.

I suspect that once the "word" gets around about which drives do work
and/or what tricks are required, the third party types will have lots
of them.  It takes a while.

-- 
George Robbins - now working for,	uucp: {uunet|pyramid|rutgers}!cbmvax!grr
but no way officially representing	arpa: cbmvax!grr@uunet.uu.net
Commodore, Engineering Department	fone: 215-431-9255 (only by moonlite)