Elpida memory deaths: the plot thickens

JUST AFTER POSTING our article about dead and dying Elpida Hyper memory chips yesterday, we came across an article at Anandtech that only reinforced our impression that Elpida has a problem.

Rajinder Gill of Anandtech had originally meant to write an article comparing Corsair's and OCZ's DDR3-2000+ memory offerings, but he had so many kits die on him during his testing that his linked article warns readers to stay away from either company's Elpida Hyper memory kits until the problems are resolved.

We assume Anandtech informed both firms of the problems it was having, as a quick response about the issue was forthcoming from Corsair along with a slightly slower one from OCZ.

Picture courtesy of xtremesystems.org

As we mentioned yesterday, Corsair quickly 'fessed up to knowing about the issue, saying "through lab testing, we have now been able to reproduce similar failures."

Admirably, the firm said it would continue to probe into this Elpida memory mystery until the exact cause of premature chip death could be determined, adding that "although a relatively small percentage of 'Hyper' ICs appear to be affected, the rate of failure is not acceptable to Corsair or to our customers." Corsair concluded, "Due to these failures, we will no longer sell Hyper-based modules until the issue can be resolved."

OCZ, although a tad slower off the mark with its response, also stepped up to say it would put a halt to sales of Elpida Hyper memory kits, although, somewhat bizarrely, its statement reads almost as though it thinks everything is just hunky-dory and there's no problem as far as OCZ can tell.

"OCZ has not yet seen extended failures in the field, and the Elipda Hyper ICs we have used came from earlier batches so we do not yet know if this is the reason why we are not seeing field issues," says the statement.

"But," it continues, "as a proactive measure based on the latest information from media reviewers like AnandTech we have officially held the use of any Elpida Hyper IC's at this time."

So, just because Anandtech complained, OCZ will freeze all its Elpida Hyper memory sales anyway, merely out of an abundance of caution mind you, but right away. Hmmm. Somehow we are tempted to doubt that was all there was to OCZ's decision, but it stopped shipping the suspect memory kits so it can spin that however it wishes, we reckon.

Either way, the firm writes that "any customers that are seeing issues are encouraged to contact us and we will absolutely take care of all customers with a no questions asked refund or exchange for another OCZ product of equal value, whichever the customer prefers. We want to make sure that our customers have peace of mind and the confidence that we will take care of any and all issues related to modules that utilize Elpida Hyper chips."

Now, OCZ's statement could mean one of two things: either the company knows there's definitely a problem but doesn't want to admit it for some reason - and if that's the case, one might be forgiven for suspecting that Elpida could have had something to do with it - or OCZ really hasn't paid attention to the high RMA rate on these particular high-end memory modules. The latter is not implausible, as rather few such high-end memory kits are sold by any one vendor in any given month, so OCZ might not have noticed that a large proportion of its bleeding-edge, screamingly fast Elpida memory kits were failing.

Still, OCZ has admitted that there might be a problem, which is certainly honourable enough for us.

The only firm that doesn't seem to be admitting there's any problem here is Kingston, which at first unofficially - in online forums and through its tech support - said it would be replacing its Elpida Hyper memory ICs with another brand due to the large failure rate, but then asked the INQ - and other news sites - to issue corrections to the effect that this was not actually the case at all.

So insistent was Kingston that there was no problem with its Elpida Hyper memory kits that it even told Legit Reviews, "The enthusiast DRAM problems some customers are encountering are nothing more than a typical issue that Kingston encounters from time to time and resolves through its normal customer service process. As such, this matter doesn't merit any special attention or consideration."

So it would appear Kingston and Elpida are in denial that they might have a problem, at the very least, and - although we wouldn't go so far as to suggest the two firms are in cahoots to cover up this situation - it seems Kingston's customers shouldn't expect much sympathy from the company, because it obviously doesn't consider these Elpida Hyper memory problems worth "any special attention or consideration."

How charming. µ