Article View: comp.arch
Article #10799Re: COMPAQ PROPOSED SCALABLE I/O ARCHITECTURE
From: Keith Smith
Date: Tue, 22 Dec 1992 15:35
Date: Tue, 22 Dec 1992 15:35
49 lines
3075 bytes
3075 bytes
In article <id.IO0W.TLE@ferranti.com> peter@ferranti.com (peter da silva) writes:
>I've seen two versions of this: one successful, one less so. The Convergent
>Technology bookshelf unit is a very nice design, with a positive mechanical
>lock and no arbitrary limits to the bus length. The Amiga expansion port is
>less successful, because it has no lock and the signal definition was never
>designed for multiple piggybacks. The original Amiga design would have been
>much better: basically stacked units with a bus running vertically up the
>back of the stack, using gravity for locking. It was abandoned because it
>would have added a several dollars to the cost of manufacturing a $1500 PC.
A modular block design is going to be slightly more expensive due to
case/connection cost. I'd like to see the Convergent box sometime.
I've never encountered one before. Is the BUS architecture closed?
But imagine if it was standardized. You could buy a pre-fab block with
standardized mounting points for a circuit board and connectors, and
plug a project board into it. Drive goes out, snap it off, and snap
another on. This of course would require that:
>I'd like to see the bus specifics decoupled from the software, so you had
>some sort of interface that you told "go to unit so-and-so and send this
>control/read this data". Then you could use a low-cost hookup with ethernet
>type cables, or a high-speed bus type option, without impacting the software.
>You could even plug a low-speed adaptor into a high speed bus to hook up
>your laptop when you get home.
>
>If course this would logically be a LAN (um, CAN: Computer Area Network?).
Yes, exactly. You could even make the thing somehow know how far off it
is, like contact the device 3 hops from the mother block, etc. The
ethernet idea of unique Umpteen bit addresses somehow scares me.
Perhaps use one decoupled bus pin, as a serial address marker that is
gobbled, incremented and passed down to the next device on the same pin,
sent every second or so until a block is connected and acks the address,
which would then pick up it's address, increment it ...
>> <SIGH> not in my lifetime I suppose.
>
>Go get a Burroughs/CT bookcase unit. UHaul uses them here in Houston, or
>call your local Unisys rep.
^^^^^
Hehehehe :) :)!
I'm sure It's not an expensive box :) :)!
--
Keith Smith uunet!ksmith!keith 5719 Archer Rd.
Digital Designs BBS 1-919-423-4216 Hope Mills, NC 28348-2201
Somewhere in the Styx of North Carolina ...
Technical Details
Message-ID:
<1992Dec22.153519.8512@ksmith.uucp>Path:
arc1-ca.pugleaf.net!sparky!uunet!rde!ksmith!keithReferences:
<1992Dec16.205908.4826@dvorak.amd.com> <1992Dec20.212105.13649@ksmith.uucp> <id.IO0W.TLE@ferranti.com>Show Full Headers
Complete Article Headers
Newsgroups: comp.arch Path: arc1-ca.pugleaf.net!sparky!uunet!rde!ksmith!keith From: keith@ksmith.uucp (Keith Smith) Subject: Re: COMPAQ PROPOSED SCALABLE I/O ARCHITECTURE Organization: Keith's Computer, Hope Mills, NC Date: Tue, 22 Dec 92 15:35:19 GMT Message-ID: <1992Dec22.153519.8512@ksmith.uucp> Keywords: I/O, Point-to-Point, High Performance, Low Cost, Bus References: <1992Dec16.205908.4826@dvorak.amd.com> <1992Dec20.212105.13649@ksmith.uucp> <id.IO0W.TLE@ferranti.com> Lines: 49