6 Jul
2010
6 Jul
'10
10:36 a.m.
There's no standard on-disk format for SMD drives,
unfortunately.
Within the DEC-compatible SMD world, many controllers (especially the Massbus-emulating)
offered lowest-common-denominator compatibility with DEC RM03/RM05/RP05 packs
in CDC 9762/CDC 9766/Memorex 677-51 drives.
There are some delicate hooks between the software device drivers, Massbus registers, and
the low level format that made it desirable to keep this lowest-common-denominator there
even for SMD drives that had geometries nothing like the 9762/9766/677-51.
But in the Sun/VME world I know of no such lowest-common-denominator standard.
Tim.
6 Jul
6 Jul
10:42 a.m.
On 7/6/10 1:36 PM, Shoppa, Tim wrote:
Yes, I remember this being a great benefit when using an Emulex QD32
to control a CDC 9762; that combination worked with RM02/RM03 packs. It
was EXTREMELY handy at work.
There's no
standard on-disk format for SMD drives, unfortunately.
Within the DEC-compatible SMD world, many controllers (especially the Massbus-emulating)
offered lowest-common-denominator compatibility with DEC RM03/RM05/RP05 packs
in CDC 9762/CDC 9766/Memorex 677-51 drives.
There are some delicate hooks between the software device drivers, Massbus registers, and
the low level format that made it desirable to keep this lowest-common-denominator there
even for SMD drives that had geometries nothing like the 9762/9766/677-51. But in the Sun/VME world I know of no such
lowest-common-denominator standard.
That's the problem. :-( Worse, I've not heard of any such attempts at
compatibility in any other minicomputer ecosystem.
-Dave
--
Dave McGuire
Port Charlotte, FL
11:03 a.m.
On 7/6/10, Dave McGuire <mcguire at neurotica.com> wrote:
I've written the list several times about the SMD drives we had on our
11/750 via a Systems Industries SI-9900 controller. The SI board in
the VAX was register-compatible with a Massbus controller, but had a
pair of 40-pin cables and was not cable-compatible.
For our boot drives, we had a 160MB Fuji drive with a transparent lid.
It appeared as a pair of RM03s to the system - DRA0: and DRA1:, IIRC.
The Fuji Eagle was *not* the same size as an RM05, so to use all the
tracks, we had to wait for the SI patch for each version of VMS
DRDRIVER.EXE to update the drive geometry table with an entry for an
Eagle in place of a real RM05. We couldn't boot from it, but it made
a great data drive. The only drawback was being stuck at the old
version of VMS until we could get up-to-date driver patches.
Fortunately, the 160MB boot drive really did look just like a pair of
RM03s and required no special handling to "just work".
-ethan
There are some
delicate hooks between the software device drivers, Massbus
registers, and the low level format that made it desirable to keep this
lowest-common-denominator there even for SMD drives that had geometries
nothing like the 9762/9766/677-51.
Yes, I remember this being a great benefit when using an Emulex QD32
to control a CDC 9762; that combination worked with RM02/RM03 packs. It
was EXTREMELY handy at work.
5064
days inactive
5064
days old
2 comments
3 participants
participants (3)
-
ethan.dicks＠gmail.com -
mcguire＠neurotica.com -
tshoppa＠wmata.com