On Wed, Mar 06 2002 at 00:33:54 +1100, Darren Reed wrote:
> In some email I received from Thilo Manske, sie wrote:
> > On Tue, Mar 05 2002 at 23:57:52 +1100, matthew green wrote:
> > >    
> > >    the SM81 with 2MB cache on the other hand, is a different story.  but then
> > >    those aren't useable in a sparc10 anyway. :)
> > > wouldn't that be SM82?  (i've never heard or seen of them before.)
> > I think "SM81-2". (the page http://mbus.sunhelp.org/modules/ says so)
> > 
> > I guess SM82 would be the name of a double CPU module but read the text the
> > two "name game" links on that page link to see how strict Sun's mbus-naming
> > conventions were ;-)
> 
> No, that'd be an 812.
Well, we can't be sure since it doesn't exist, but this aproach is from the
other direction. 

What I wrote was an answere to the question "If something is called SM82, what
would that be?" and not "What would be the name of a 2*SS@75Mhz module?"

> 502 = 50MHz, 0MB 2nd level cache, 2 CPUs (1 board)
> 512 = 50MHz, 1MB 2nd level cache, 2 CPUs (1 board)
> (http://mbus.sunhelp.org/misc/game52.htm)
Yes, I put a note about the "name games" in my first reply.

Quote from that text: "So did the SM52 module (dual-SuperSPARC @ 45MHz) ever
actually exist? Yes!"

So the only thing that ever existed with an SM?2 name was a double CPU
module, the logical conclusion is that SM82 would be a double CPU module as
well.

> There are no 6*2, 7*2 or 8*2 boards.
(Except 61-2 and 81-2).

I guess we all agree now that Sun's mbus module naming scheme was pretty
confusing/stupid...

-- 
Dies ist Thilos Unix Signature! Viel Spass damit.