🚀 go-pugleaf

Edwin Steele <ed@zeta.org.au> writes:

> In <bsnqps$t6r$1@yoda.apana.org.au> Kralizec Craig wrote:
> > Chris Morgan <cm@mihalis.net> writes:
> 
> > Sun part numbers for them are 501-5740 and 501-5741. I can only assume 
> > the second one was a result of an engineering bug in the first version... 
> > Anyone with any knowledge of their history?
> 
> I rememeber that there was an engineering bug with the 400mhz CPUs in
> the first few batches - for memory Sun weren't very responsive in 
> replacing
> them. My old employer had an E450 w/ 4x400mhz cpus that was bought just 
> after
> the 400mhz chips started shipping - we got hit by a problem tho I can't
> remember all of the details.

I think you are referring to the US-II chips with the E$ (cache) bug
where it would fail under periods of zero load. This was terrible for
enterprise machines, however this discussion refers to a different
chip, the US-IIi, which was quite a good cheaper variant for volume
systems. I've got ... too many :)

Chris
-- 
Chris Morgan
   "Post posting of policy changes by the boss will result in 
    real rule revisions that are irreversible"

		- anonymous correspondent