We've been chatting to vendors, retailers and discovering just what it is that's broken in Intel's Sandy Bridge chipset. Essential reading follows...
To say that Intel's announcement of a recall and gradual replacement of Sandy Bridge motherboards caused a stir in enthusiast circles is a touch of an understatement. Our story yesterday drew a mess of commentary ranging from outright blasts at Intel to befuddlement as to how this could happen - and our forums were little different.
Regardless of the response, however, one thing remained constant - anyone who has already invested in the platform feels a little ripped off!
At the heart of the problem is Intel's decision to effectively shoe-horn last generation circuits into its new platform. Tech site Anandtech did a great job of explaining exactly what the issue is, so it's worth reading - however, the gist of the matter is thus:
"The problem in the chipset was traced back to a transistor in the 3Gbps PLL clocking tree. The aforementioned transistor has a very thin gate oxide, which allows you to turn it on with a very low voltage. Unfortunately in this case Intel biased the transistor with too high of a voltage, resulting in higher than expected leakage current. Depending on the physical characteristics of the transistor the leakage current here can increase over time which can ultimately result in this failure on the 3Gbps ports. The fact that the 3Gbps and 6Gbps circuits have their own independent clocking trees is what ensures that this problem is limited to only ports 2 - 5 off the controller."
The resulting failure rate is something like five to 15 per cent over a three year time period. No failures have been yet reported.
Locally, it's been difficult to get responses from vendors and manufacturers alike, a situation made even more complex by Chinese New Year, when many of the Taiwanese head honchos that local product managers report to.
However, there is some news. GIGABYTE, for instance, has told us that it's requested that all retailers suspend sales of affected parts immediately, while anyone who has purchased a Sandy Bridge board should look to eventually RMA their purchase - for now, though, the official word is that users stick to SATA channels zero and one. It's been suggested that there may be a straight swap available, but that will of course rely on availability of new parts. It is, at least, good to see GIGABYTE being so straight up about this. We're waiting on responses from other manufacturers, but we expect them to be basically identical - what else can they do?
Distributors, so far, are keeping very quiet - no doubt they're caught in the middle by the need to maintain good relationships with Intel and mobo manufacturers alike. We have spoken to some retailers, and one anonymous store owner has told us he's planning to offer free SATA II PCI cards to affected customers until he here's anything from his importers and suppliers. Good will offers like that are unlikely to sway nervous customers from committing to a new purchase, however.
Interestingly, while you could expect AMD to be crowing over their rival's misfortune, the opposite seems to in fact be the case. At yesterday's Fusion APU launch in Sydney, AMD's Vice President for Worldwide Product Marketing Leslie Sobon suggested that it merely illustrated the complexity of modern computing designs. Regardless, it's hard to think someone at AMD isn't considering rushing its Bulldozer out as soon as possible.
She also suggested that AMD's move to spin-off its fabbing operations into separate, independent companies is something that will help stop AMD making similar mistakes. "It's in their best interest to deliver good parts," Sobon said, "in a purely business sense." With Intel's production entirely internal, it's hard to fault her wisdom.
Atomic's not immune, either - we and our sister title PC & Tech Authority have a pile of Sandy Bridge boards that it's now more or less useless to test. It looks like the Sandy Bridge architecture, which we talk up pretty heavily in issue 122 (on sale next week, FYI), is pretty much on hold for the time being.
NOTE: Apologies Nich... and any hardware enthusiast Gypsies.
Issue: 137 | June, 2012