26 May
1999
26 May
'99
12:51 p.m.
ard(a)p850ug1.demon.co.uk (Tony Duell) wrote:
Most later workstations had SCSI interfaces for the
hard disk (even if
they then broke that by insisting on an ST506 drive on the other side of
a SCSI->ST506 interface, as ICL and Torch both did). Some older
Sun did this too. It wasn't 'til SunOS 4.0 that the SCSI-disk driver
would actually talk to SCSI disks. Before that, it wanted to talk to
an Adaptec ACB4000 or an Emulex MD21 (for ESDI disks).
I think I remember reading somewhere that this was done because the
SCSI-to-whatever interface had the intelligence for bad-block
remapping. But I wouldn't be surprised to find that the cost of the
drives had something to do with it; I remember Amiga folks scheming to
use ACB4000 boards with their SCSI interfaces because it was cheaper
than buying a SCSI disk, and I've opened a few Mac SCSI hard disk
boxes to find the same sort of thing inside.
-Frank McConnell
26 May
26 May
1:18 p.m.
New subject: non-SCSI disks on a SCSI disk interface (was Re: Space, the next frontier)
At 02:51 PM 5/26/99 -0700, Frank wrote:
Sun did this too. [scsi to MFM] ...
I think I remember reading somewhere that this was done because the
SCSI-to-whatever interface had the intelligence for bad-block
remapping. But I wouldn't be surprised to find that the cost of the
drives had something to do with it;
Cost has more to do with it than bad blocks, the 4.1BSD disk driver knew
how to remap bad blocks but with Adaptec and Emulex solutions you could put
_two_ cheap drives behind a SCSI interface (logical unit 0 and 1) and when
you did that the costs were significantly less for the scsi+ESDI solution.
Of course Sun was a huge proponent of IPI, claiming it would wipe SCSI off
the planet.
--Chuck
6:27 p.m.
New subject: non-SCSI disks on a SCSI disk interface (was Re: Space, the
On 26 May 99 at 14:51, Frank McConnell wrote:
ard(a)p850ug1.demon.co.uk (Tony Duell) wrote:
Atari ST hard drives had a weird similiar configuration driven by cost and
proprietary considerations I'm sure. The ASCSI (?) output from the ST was
converted by an ICD(company name) adaptor to SCSI and then an Adaptec
board converted that to either MFM or RLL. Dont want to open it up but ISTR
the main chip on the Adaptec board was a WD 1004S (?)
ciao larry
lwalker(a)interlog.com
Let us know of your upcoming computer events for our Events Page.
t3c(a)xoommail.com
Collectors List and info http://members.xoom.com/T3C
Most later workstations had SCSI interfaces for
the hard disk (even if
they then broke that by insisting on an ST506 drive on the other side of
a SCSI->ST506 interface, as ICL and Torch both did). Some older
Sun did this too. It wasn't 'til SunOS 4.0 that the SCSI-disk driver
would actually talk to SCSI disks. Before that, it wanted to talk to
an Adaptec ACB4000 or an Emulex MD21 (for ESDI disks).
I think I remember reading somewhere that this was done because the
SCSI-to-whatever interface had the intelligence for bad-block
remapping. But I wouldn't be surprised to find that the cost of the
drives had something to do with it; I remember Amiga folks scheming to
use ACB4000 boards with their SCSI interfaces because it was cheaper
than buying a SCSI disk, and I've opened a few Mac SCSI hard disk
boxes to find the same sort of thing inside.
-Frank McConnell
9185
days inactive
9185
days old
2 comments
3 participants
participants (3)
-
cmcmanis＠mcmanis.com -
fmc＠reanimators.org -
lwalker＠mail.interlog.com