Subject: Re: strange disks...
To: None <port-vax@NetBSD.ORG>
From: Michael Sokolov <mxs46@po.CWRU.Edu>
List: port-vax
Date: 12/30/1997 11:18:05
Dear Tim,
You write:
>Of the companies that sell Q-bus and Unibus MSCP controllers, I think
>the only ones still intact are Dilog, Andromeda, and CMD.
I know all of these except Andromeda. Do you know any contact
coordinates for them?
You write:
>I think it ought to be completely banished from the disklabel process
>for MSCP disks
A good idea in theory, but is it worth rewriting the relevant OS code
and running into compatibility issues?
>the geometry is so hidden behind the interface
>that it's difficult to use this information for performance purposes
>(i.e. seek ordering.)
You have a good point here. In the case of SMD, the vast majority of
Winchesters split the disk into zones with different angular recording
densities in order to make the arial recording density closer to constant.
This means that no traditional scheme (# of cylinders, # of heads, # of
sectors per track) can describe the true geometry of the Winchester.
You write:
>I've got about 4000 pages of documentation on various SMD drives, so
>unless you want to see a truck pulling up at your door, you might
>want to narrow down your search :-).
When I get to pulling my 8" drives out of those horrible Sun drive
cabinets, I'll let you know what they are (assuming they have stickers).
>(Most of my books are Fujitsu
>tech manuals.)
I have some Fujitsu SMD Winchesters. There is an M2392K, and the two
unknown drives may turn out to be Fujitsu. I also have a Fujitsu ESDI
Winchester (M2246E).
You write:
>Most of the DEC resellers (i.e. Newman, Continental Computers, etc)
>will have used Emulex QD32's/33's available.
Could you give me some contact coordinates for them?
Sincerely,
Michael Sokolov
Phone: 440-449-0299
ARPA Internet SMTP mail: mxs46@po.cwru.edu