overvolted memory my damage corei7 chips.

[alsenior]

According to custom pc intel is recommending that people do not push the memory beyond 1.65v for the new chip as it may cause permanent damage to the chips.

"Overclockers looking to bolster their new Nehalem CPUs with overclocked memory may be disappointed. Intel is telling motherboard manufacturers not to encourage people to push the voltage of their DIMMs beyond 1.65V, as anything higher could damage the CPU. This will come as a blow to owners of enthusiast memory, such as Corsair's 2.133MHz DDR3 Dominator RAM, which needs 2V to run at its full speed with 9-9-9-24 timings."

form Custompc via /.

will you be pushing your new Corei7 processor when it arrives. or has this put you off enough to make you stick to stock timings for the time being?

note: a spokesperson from asus says the warning applys only to memory outside of the JEDEC spec. See the custompc article.

[Agent]

Yeah, this was posted a while back Only time will tell how badly it hits us overclockers.

It's always been the case with overclocking, but I guess with major changes to the design at some point it was going to catch up with us

[dave87]

What has 1.65v on the CPU got to do with the memory requiring 2v?

Or am I being dense?

[alsenior]

What has 1.65v on the CPU got to do with the memory requiring 2v?

Or am I being dense?

you cant push the memory above 1.65v or else it will damage the cpu that is what intel are saying

[Funkstar]

Nehalem uses an integrated memory controller doesn't it? So an increase in memory voltage is going to mean an increase in voltage in that part of the processor to, leading to more heat.

[dave87]

I thought memory and CPU voltage were independent?

My memory takes 1.8v, which is low compared with high speed DDR2?


Or is Nehalem paired with DDR3?

[Funkstar]

Thats just what I assumed when I read about this before, so it might not be right, but to me it's logical that having that 2v in the processor die isn't going to be a good thing.

Nehalem uses DDR3, just as the above quote says

[dave87]

lol, I skimmed the article, and missed that.


Reading becomes more challenging after circa 18 hours awake

[Mblaster]

This was never mentioned as a problem for AMD CPUs that have had the memory controller on die for ages now. All I can think of is that the memory controller (in the CPU) has to use the same voltage as teh memory chips, but that sounds wrong.

Does anyone know the actual reason why Intel are recommending this?

[Funkstar]

This was never mentioned as a problem for AMD CPUs that have had the memory controller on die for ages now. All I can think of is that the memory controller (in the CPU) has to use the same voltage as teh memory chips, but that sounds wrong.

Just means Intel and AMD are doing it different ways.

It's only a guess, no one has said anything to contradict this in anything I've read (which i know, doesn't make it right)

[CooperXS]

This was never mentioned as a problem for AMD CPUs that have had the memory controller on die for ages now. All I can think of is that the memory controller (in the CPU) has to use the same voltage as teh memory chips, but that sounds wrong.

Does anyone know the actual reason why Intel are recommending this?

I guess you never really overclocked A64 then
With extensive voltage on RAM but low on CPU you could fry memory controller.

As for the actual reasons and physics behind Intel's statement one needs to know and fully understand all the schematics and feedbacks of CPU-RAM. Although those lucky ones having i7 on hands haven't killed any as of yet. And these guys do utilize memory on high voltage.

[chilled]

Seeing the JEDEC spec for DDR3 is 1.5V, 1.65V isn't that bad...