23 Apr
2008
23 Apr
'08
7:54 a.m.
Date: Wed, 23 Apr 2008 09:30:23 -0500
From: Jules Richardson
Sigh.. no. I had it in my head that one of the OMTI
5300 / 5400 boards
would do it - I checked though and they *are* QIC-02, but the interface is
only SASI, not SCSI :-(
Or "brain dead" SCSI, in the case of the 7400. Sigh. I'm not
turning up anything on the web on QIC-02-to-SCSI bridges.
Hazy memory says that one vendor did do a QIC-02 -
QIC-36 bridge, but
including one of those *and* a suitable SCSI bridge in whatever you're
doing might be a bit silly.
I think the QIC-bridges go the other way. I've got a Wangtek QIC-36-
to-QIC-02 bridge board (QIC-36 is the "basic stupid" interface, while
QIC-02 is the "intelligent buffered" interface).
It does seem there were a few QIC-02 ISA tape
controllers around, though -
so you'd think that someone must have done a SCSI bridge!
That's mostly because there's nothing much on a QIC-02 ISA tape
controller; just address-decoding logic and DMA and IRQ handshake;
think something on the level of your basic IDE controller. The
smarts are in the drive.
Ah, well. Time to look for a nice SCSI QIC drive. Personally, I
like Tandberg over Wangtek. Anyone have any other preferences?
Cheers,
Chuck
23 Apr
23 Apr
11:29 a.m.
Chuck Guzis wrote:
I don't think I've ever had any of the 7xxx-series boards, just the 5xxx ones.
"brain dead" in what way? Just the lack of Identify that you mentioned in
reply to Al?
I've found so many "SCSI" devices of that era which lack Identify but appear
to be otherwise SCSI-compatible. I'm yet to find a way (or find anyone who's
done it) of working around that limitation in Linux - it'd solve a lot of
problems (even if I had to somehow hard-code fake Identify responses into the
kernel driver and compile different kernels for talking to different devices!)
I *think* the way it worked was that there was SASI, then SCSI, then SCSI with
the Common Command Set - and it's the latter which dictates Identify support
(and that modern hardware/software expects). Anything from the "just SCSI"
days doesn't quite cut it...
Entirely possible. I've got various QIC drives that I use occasionally, but
nothing that's '02 or '36.
I suppose in theory the protocol/interface might be documented somewhere such
that a suitable high-speed parallel port (of sufficient data width) might be
able to talk to a QIC-02 drive - but that's a lot of effort for something that
then wouldn't be particularly portable.
Date: Wed, 23
Apr 2008 09:30:23 -0500
From: Jules Richardson
Sigh.. no. I had it in my head that one of the
OMTI 5300 / 5400 boards
would do it - I checked though and they *are* QIC-02, but the interface is
only SASI, not SCSI :-(
Or "brain dead" SCSI, in the case of the 7400. Hazy memory
says that one vendor did do a QIC-02 - QIC-36 bridge, but
including one of those *and* a suitable SCSI bridge in whatever you're
doing might be a bit silly.
I think the QIC-bridges go the other way. I've got a Wangtek QIC-36-
to-QIC-02 bridge board (QIC-36 is the "basic stupid" interface, while
QIC-02 is the "intelligent buffered" interface). Ah, well. Time to look for a nice SCSI QIC drive.
Personally, I
like Tandberg over Wangtek. Anyone have any other preferences?
Can't comment on those two, but I've got a pair of SCSI-interfaced Archive
ones (such as Al mentioned) which have always done me proud (plus a 'just in
case' dead one in the junk pile for a source of donor bits)
cheers
Jules
1:08 p.m.
New subject: Tandberg QIC was WTB: Emulex MT-02 controller
Ah, well.
Time to look for a nice SCSI QIC drive. Personally, I
like Tandberg over Wangtek. Anyone have any other preferences?
Can't comment on those two, but I've got a pair of SCSI-interfaced Archive
ones (such as Al mentioned) which have always done me proud (plus a 'just in
case' dead one in the junk pile for a source of donor bits) 
24 Apr
24 Apr
1:05 p.m.
> I think the QIC-bridges go the other way.
I've got a Wangtek QIC-36-
Most of them certainly do. QIC02 to the host and QIC36 to the drive.
I have (and still use) an old Archive Sidewinder drive. It's old enough
that it's 8" form factor, and needs +5V and +24V. On the metal frame is
the drive iteslf and a stacked pair of PCBs that form the controller. The
interface to the host is QIC02, the interface between the drive and
controller is QIC36 (or something very close to it).
to-QIC-02
bridge board (QIC-36 is the "basic stupid" interface, while
QIC-02 is the "intelligent buffered" interface).
Entirely possible. I've got various QIC drives that I use occasionally, but
nothing that's '02 or '36.
I suppose in theory the protocol/interface might be documented somewhere such
that a suitable high-speed parallel port (of sufficient data width) might be
able to talk to a QIC-02 drive - but that's a lot of effort for something that
then wouldn't be particularly portable.
1:47 p.m.
Tony Duell wrote:
The QIC02 interface is pretty simple. 8 data lines and
a handful of
control/handshake lines (I think fewer than 8 of those). The control
lines, from what I remmeber are all unidirecional (obviously some from
drive to host, others from host to dive), the data lines are
bidirectional (but all in the same direaction at any given time).
I think the interface can be bit-banged. If not (i.e. there are some
critical timings)
Hmm, maybe the designers were sensible, and any critical timings are solely on
the device response side - i.e. the host can take as long as it wants to
process things, but the device must respond within a certain time period (and
then just happily sits there until the host acknowledges).
Of course none of this probably helps Chuck, who quite possibly doesn't want
to be messing with designing hardware and writing driver software :-) It's
just nice to speculate that it could be done once QIC interface boards are
unobtainable...
cheers
Jules
6086
days inactive
6087
days old
4 comments
4 participants
participants (4)
-
ard＠p850ug1.demon.co.uk -
cclist＠sydex.com -
jules.richardson99＠gmail.com -
spectre＠floodgap.com