Intel i960 evaluation board
That's odd ... I remember the i960 being touted as the cat's meow for laser
printers and other highly demanding and high-speed embedded applications. I
was never interested enough to investigate without a contract,however, since
I am not an Intel enthusiast. Last time I saw one, it was on a MYLEX
3-channel SCSI RAID board. Could I be confusing this with something else?
Dick
----- Original Message -----
From: Eric Smith <eric_at_brouhaha.com>
To: <classiccmp_at_classiccmp.org>
Sent: Friday, April 21, 2000 6:56 PM
Subject: Re: Intel i960 evaluation board
> I wrote:
> > The i960 family was the stripped-down commercial version of the
ill-fated
> > Gemini (P7) 33-bit (not a typo!) processor, a collaboration between
Intel
> > and Siemens. The two companies created a workstation company called
BiiN
> > to sell the workstations, and although they shipped some prerelease
> > machines, AFAIK they never offered any for sale.
>
> Allison wrote:
> > It also grew from the 8089 (20 bit) and the 8751(8bit) for embedded
> > processing
> > tasks like engine controls.
>
> Perhaps you're thinking of the 8096. The i960 has no architectural or
> design similarity to the 8089 or 8751, was not designed by the same
> engineering organization, and was not originally intended for the
> embedded market. As I stated previously, it was originally developed
> for high-reliability workstations and servers.
>
> Intel's ended up pushing the i960 for embedded use only after it failed
> to be accepted as a workstation-class product.
>
> The 8089 was not particularly intended for embedded applications; it was
> supposed to be a channel controller, i.e., a smart DMA controller able
> to perform functions similar to the channels on IBM mainframes. It is
> very poorly suited to general purpose (non-DMA) use, even for embedded
> systems (unless the embedded system needs a fancy DMA controller).
> The only thing 20-bit about the 8089 is the address, but at least it was
> a flat 20-bit space unlike that of the 8086.
Received on Fri Apr 21 2000 - 20:08:38 BST
This archive was generated by hypermail 2.3.0
: Fri Oct 10 2014 - 23:32:42 BST