"Rob Young" <[log in to unmask]> wrote in message
news:lGELmlXAGUYz@eisner.encompasserve.org...
> In article <[log in to unmask]>, "Bill
Todd" <[log in to unmask]> writes:

...

> > Sorry - UltraSPARC III.  You know, the processor line Rob always
denigrates
> > for being such a pig that Itanic will quickly usurp Sun's entire
customer
> > base as its start on conquering the world.  That's why it would be so
funny
> > if McKinley turned out to be slower after all.
> >
>
> Is that an "official" prognostication, or a wish?

I made a quantitive prediction (based on publicly-available information
which I described) some time ago that McKinley would weigh in at 600 - 700
SPECint2K (and most likely in the middle of that range), and I stand by it
(unless it gets delayed so long that Intel can skip the first release
entirely and go directly to what would have been the first major
post-release process tweak - which is looking increasingly possible).  The
best USIII SPECint2K currently is 610, but since such values do tend to rise
as new configurations appear I'd say there's a distinct possibility, though
perhaps not a better-than-even chance, that if McKinley doesn't in fact wait
for a process-tweak before releasing it will at introduction have a hard
time equalling USIII's SPECint2K speed.

- bill

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *