The Inquirer reports that at least some current nForce3/nForce4 motherboards are incompatible with Revision E Hammers, and that it might not be fixable.
This may well be a big deal, it may well not. However, under such circumstances, the smart thing to do until we know for sure is to follow the rule, “When in doubt, don’t.”
One way or the other, we most likely will get a better picture of this problem within a week, so in the meantime, wait (and maybe look a bit more closely at some Via boards).
I suspect this won’t turn out to be all too big of a deal, since apparently some nForce boards do work, but then again, neither I (nor anybody else) is going to buy you a new mobo if that assessment is wrong, so wait.
This is as good a time as any to note another oddity in AMD’s behavior lately.
AMD hasn’t updated its datasheets for a long time. Outside of a bare mention of the first 90nm Athlon 64s last October, they haven’t mentioned anything new. Nothing on 90nm Semprons. Nothing on Revision E Opterons, or revision E anything.
Normally, AMD publicly updates the datasheets when a product is announced/becomes available. They’ve stopped doing that. Why?
As we’ve noted before, no talk about strained silicon either, and little talk about Revision E itself. Since the operative Prime Directive for AMD seems to be, “When the going gets tough, the tough get quiet,” you have to wonder if there’s trouble in Dresden.
Maybe it’s just this nForce mobo problem. Or maybe besides that, AMD’s having some problems implementing IBM’s secret strained silicon SOI sauce at 90nm. Sorry, AMDroids, but this is an important matter to those wanting 3GHz from their Hammers.