Subject: Re: LSI Logic
To: None <port-pmax@netbsd.org>
From: Peter C. Wallace <pcw@mesanet.com>
List: port-pmax
Date: 05/17/2000 12:14:06
On Wed, 17 May 2000, David Evans wrote:
> Peter C. Wallace wrote:
> >
> > Looks like (with a quick ohmmeter check) that most if not all
> > MIPS bus signals go through ASIC to the daughterboard connector...
> >
>
> How well does the MIPS bus run over connectors like this? Perhaps that was
> the reason for the creaiton of a "new" protocol between the CPU mdoule and
> the motherboard. SGI did a similar thing on the R4000 Indigo and subsequent
> machines, allowing one to stick, say, an R5000 in an Indy that came with a
> cacheless R4600. They didn't do it with Crimson, their first R4x00 design,
> so maybe they learned from that. I haven't disected an Onyx/Challenge to see
> how the IP19 CPU boards are set up.
>
> --
> David Evans (NeXTMail/MIME OK) dfevans@bbcr.uwaterloo.ca
> PhD Student, Computer/Synth Junkie http://bbcr.uwaterloo.ca/~dfevans/
> University of Waterloo "Default is the value selected by the composer
> Ontario, Canada overridden by your command." - Roland TR-707 Manual
>
Its actually a good design, all high speed local signals confined to the
daughter-card, once the signals get on the main card, you probably have
much longer signal lengths.
Peter Wallace