For those considering the new Sandy Bridge chip: READ THIS!!!

[Peter.Cort]

Intel expects that over 3 years of use it would see a failure rate of approximately 5 - 15% depending on usage model.

^^ from the anandtech article.

 

[aspire]

Well I just hit up Intel chat support and was told that they were still in the process of investigating how widespread the issue was but that they would let me know if I would need to return my motherboard or anything like that.

Was also told that using it as normal for the time being would be just fine.

 

[PP Mguire]

Intel expects that over 3 years of use it would see a failure rate of approximately 5 - 15% depending on usage model.

^^ from the anandtech article.

I would expect that to be a total guesstimation honestly. The stuff just got released and they found a flaw that quickly, meaning that some cases have shown as early as now.

 

[Peter.Cort]

On its conference call to discuss the issue, Intel told me that it hasn't been made aware of a single failure seen by end users.

^^ first part of the article.

That would lead me to believe that the issues have only been seen in server applications or in testing where massive amounts of data are being transferred and the errors will be seen in a much greater scope

 

[Slaymate]

Well if anyone can over-stress a component I'm your man. My new 2600K and Asus P8P67 Pro are scheduled for delivery on Thursday and I'll probably start abusing it by Saturday. If I have any issues I'll be posting them

 

[PP Mguire]

On its conference call to discuss the issue, Intel told me that it hasn't been made aware of a single failure seen by end users.

^^ first part of the article.

That would lead me to believe that the issues have only been seen in server applications or in testing where massive amounts of data are being transferred and the errors will be seen in a much greater scope

I only read a little of the TPU article, so if what you and the other article says is true then normal people should be fine -.- Big deal over nothing.

 

[Peter.Cort]

wanna send me your i7 and mobo for a hug ? i'll pay for shipping

and PP it's more the idea that I think overtime it will become a more prevalent issue for end users. Better than how microsoft handled the RRoD. If you notice the issue, might as well get the issue over and done with as soon as possible.

and the main problem is that it's a physical electrical issue in the engineering of the circuit so it's not like it's only a couple of isolated issues.

 

[Slaymate]

Didn't see this when I posted. Oops.

I added the link from your Thread on this subject to the 1st post in this Thread

 

[Vizier]

The issue is a silicon-chip oversight, and as mentioned before, this would degrade your SATA performance to the point where it would eventually not recognize the drive anymore, or disappear entirely from BIOS.

Since this is a hardware issue, no amount of programming, third-party drivers, or firmware updates will solve the issue: thus why Intel has sought to issue a total recall.

From what I can gather, Manufacturers and OEMs found out about this the same way we did: through the press. To this end, this means that you will have a mixed reactions from companies, ranging from sweeping this under the rug to going above and beyond the call of duty to make things right for the customer. Trust me when I say NOBODY is happy with this bombshell intel dropped on us.

The news is important because for those of us who game hardcore and do heavy video and digital editing, our rate of failure is more around 15% and upwards. That means quite possibly one out of five of us will go "wtf, my rig failed."

While you got to give Intel props to catching this early and taking the recall and PR hit, you also have to wonder who took a coffee break during Quality Control testing to let this one slip by.

 

[Slaymate]

Personally, I'm thinking this was probably found once the enthusiast market started overclocking the crap out of them. And Intel maybe assuming a "worst case scenario" posture now so that they can downplay the issue to their stockholders. If the failure rate is 10% but they've projected 15% then they look "better" than they would if they initially said 10% and it did reach 10% or higher.

And I'm wondering if maybe this is part of the reason why these cpu's hate raising the BCLK.