On Wed, 12 Dec 2001, Pat Finnegan wrote:
> On Wed, 12 Dec 2001, Boatman on the River of Suck wrote:
>
> > On Wed, 12 Dec 2001, Jeffrey S. Sharp wrote:
> >
> > > > Blatent easy way out is emulate that 11/70 on athlon XP 1900+. :-)
> > > > Smack me if you dare. :-)
> > >
> > > That would probably be his cheapest way to attain the speed he wants.
> > >
> > > <ducks>
> >
> > And exactly how would you speak to UNIBUS?
> >
> > Peace... SRidhar
> >
>
> How would you talk to unibus without the 300 or more wait states that
> would make the 1G processor speed useless anyways? Unless you can hook
> real IO devices up to the processor that are of *reasonable* speed, all
> you have is a 'souped up adding machine' (to quote _Colossus:_
> _The_Forbin_Project)
What I was planning on doing is doing a custom set of bridge sets to talk
to multiple UNIBUSses (way multiple) simultaneously by way of a
channelized I/O system transparent to the software. (We are talking
multiple years of work).
Peace... Sridhar
Received on Wed Dec 12 2001 - 17:21:12 GMT
This archive was generated by hypermail 2.3.0
: Fri Oct 10 2014 - 23:33:38 BST