Friday, October 3 2008
Nehalem does promise to be a processor to look out for, it would be Intel's next installment, post the successful Core 2 series processors. This time however, Intel made a core modification with the way the system handles memory. The Bloomfield processors house a massive 192-bit wide memory controller for supporting tri-channel DDR3 memory. It however was found that the controller could bring in limitations to the DIMM voltages that the system could support.

The retail version of ASUS P6T Deluxe OC Palm Edition motherboard was unboxed by XFastest. Being the retail product, as usually, it comes with precautionary labels attached to parts of the motherboard. The one that covers the 6 DDR3 DIMM slots reads:

According to Intel CPU SPEC, DIMMs with voltage setting over 1.65V may damage the CPU permanently. We recommend you to install DIMMs with voltage setting below 1.65V.


It could have implications on the current DDR3 memory market as well as you, if you happen to have DDR3 modules, which you plan to retain for use in the future platform, that operate above the said voltage. It also means that in the near future, we could be seeing memory sticks that facilitate overclocking at much lower voltages. From a technology standpoint, companies such as Samsung, Elpida, Micron, etc., are working on releasing DRAM chips based on newer silicon fabrication technologies, that operate at lower voltages.
posted by btarunr - 4:36 PM |  Related News

User comments
51 to 76 of 89 | Go to Page 1 2 3 4    Previous | Next
by Wile E (October 4th - 7:19 AM) - Reply
This doesn't mean Vdimm and Vcore are the same. I mean think about it. DDR3 Jedec calls for 1.5V. There's no way Intel is gonna run these new cpus at a minimum of 1.5V.

It's just like AMD cpus, if you crank too much vdimm on an AMD setup, you can damage the cpu. And it's not just 939, It will still happen on an AM2 cpu. It's just that DDR2 runs on much lower voltages than DDR1, and gave us much more headroom.

While 1.65V is not the optimal voltage for this to occur for an overclocker, the whole issue is still being blown out of proportion.
by Mussels (October 4th - 7:29 AM) - Reply
by: Wile E

While 1.65V is not the optimal voltage for this to occur for an overclocker, the whole issue is still being blown out of proportion.
Yeah i think you have it right.

The CPU voltage and ram voltage arent hte same, the ram CONTROLLER voltage and ram voltage are. That just means theres a max safe ram voltage of 1.65v - it doesnt mean the CPU will be stuck there.

1.65V aint so bad, new, lower voltage ram will come out. To those who complain that their already bought DDR3 is useless... suck it up. you chose to blow your money on an unknown variable, it was your own decision to buy that ram.
by btarunr (October 4th - 7:30 AM) - Reply
That's an overclocker-grade motherboard, and it says >1.65V for the DIMMs = CPU "permanent damage". Asus doesn't usually sound this stern on any of their high-end motherboard labels or manuals. At most they go on with the usual "will void warranties, will damage hardware" drill.
by Laurijan (October 4th - 7:40 AM) - Reply
by: phanbuey
I agree. :toast:. Nvidia chipsets FTW!
Not FTW if Nvidia stops making chipsets: http://tech.slashdot.org/article.pl?sid=08/08/02/1749213
by REVHEAD (October 4th - 8:30 AM) - Reply
This is were I think AMD are going to comeback, they are the kings of the onboard memory controller and have been for a few years now, Intel are going to have to fight hard to overcome there own implimentation of this and this seems to be one of the early roadblocks, and I am sure there are going to be a few more.

I am not a AMD fanboy I just like to run whatever is best bang for buck at the time, I have a Intel setup at the moment and have had so for a year or so now, but I do feel when AMD move to DDR3 AM3+ they will be level pegging with Intel once again, its been a longtime comming, but non the less its heading this way..
by btarunr (October 4th - 8:38 AM) - Reply
This is where I think Intel makes real sure that the Yorkfield stocks that are left, are properly milked. AMD doesn't even have an iota of chance against Bloomfield as of today.
by Solaris17 (October 4th - 6:37 PM) - Reply
you know i have to agree with how stupid this is but at the same time i just want to let everyone know that i think its all hype.....i think in all honesty they are just making this a big deal to stop OC'ing. Im saying this because my M2n32 SLI deluxe and my NF4 Ultra II M2 said the same thing....when you went into the bios and went to change the ram voltage on the right hand side in the little panel it would say adjusting ram voltage over default settings could lead to severe damage to your processor and memory!....and you whant to know what? nothing ever happened i think the chance has always been their intel is just sizing it up for some reason i honestly dont think it will be anything to worry about.
by Laurijan (October 4th - 7:39 PM) - Reply
by: Solaris17
you know i have to agree with how stupid this is but at the same time i just want to let everyone know that i think its all hype.....i think in all honesty they are just making this a big deal to stop OC'ing. Im saying this because my M2n32 SLI deluxe and my NF4 Ultra II M2 said the same thing....when you went into the bios and went to change the ram voltage on the right hand side in the little panel it would say adjusting ram voltage over default settings could lead to severe damage to your processor and memory!....and you whant to know what? nothing ever happened i think the chance has always been their intel is just sizing it up for some reason i honestly dont think it will be anything to worry about.
Soon we will know if thats true..
by phanbuey (October 4th - 10:10 PM) - Reply
by: Scrizz
NVidia can't help you at all :slap:
the mem controller is in the CPU :laugh:
since the the MB controls the voltage... Im pretty sure theyre linked artificially, CPU and memcontroller are not relevant to ram voltage from what ive read.

If a Mobo maker wanted to, they could design their board with unlinked voltages. I think "chipset" was the wrong word to use there. But someone like Nvidia or Ati could make a nehalem board that uses whatever trichannel ram fits in the slots.

thats what i meant by "nvidia chipsets FTW" - not that theyre great MBs, but rather that alternative companies making mobos will put an end to intel's bullS%^ pretty quick.

EDIT: yeah i read that their division was one the way out. Which sucks... because i think the ddr3 linkage is in the inherent design of the X58 (if it is true) and that intel will continue to put limitations on the OCability of the cheper processors to make us splurge on the higher end models. bastards. My next build will be AMD out of sheer priciple if that happens.
by npp (October 4th - 10:48 PM) - Reply
by: REVHEAD
This is were I think AMD are going to comeback, they are the kings of the onboard memory controller and have been for a few years now, Intel are going to have to fight hard to overcome there own implimentation of this and this seems to be one of the early roadblocks, and I am sure there are going to be a few more.

I am not a AMD fanboy I just like to run whatever is best bang for buck at the time, I have a Intel setup at the moment and have had so for a year or so now, but I do feel when AMD move to DDR3 AM3+ they will be level pegging with Intel once again, its been a longtime comming, but non the less its heading this way..
How do 27,4ns latency and 19000MB/s sound to you? I don't think that AMD's memory controller is capable of that right now. And this is what you get with a Q965, check it out for yourself @ ocxextreme.org. By the way, the guy was running DDR3-1600Mhz and almost 2V voltage, so all this hype about some linkage between RAM and CPU voltage is pure BS.
by phanbuey (October 4th - 11:11 PM) - Reply
by: Mussels
Yeah i think you have it right.

The CPU voltage and ram voltage arent hte same, the ram CONTROLLER voltage and ram voltage are. That just means theres a max safe ram voltage of 1.65v - it doesnt mean the CPU will be stuck there.

1.65V aint so bad, new, lower voltage ram will come out. To those who complain that their already bought DDR3 is useless... suck it up. you chose to blow your money on an unknown variable, it was your own decision to buy that ram.
why can't they put a resistor(or something) between the ram and the cpu mem controller that will lower the Vdimm for the controller?
by CDdude55 (October 4th - 11:17 PM) - Reply
by: Laurijan
Not FTW if Nvidia stops making chipsets: http://tech.slashdot.org/article.pl?sid=08/08/02/1749213
They will never stop making chipsets.(but then again the Core i7 boards are saying otherwise)
by kid41212003 (October 5th - 12:02 AM) - Reply
by: Wile E
This doesn't mean Vdimm and Vcore are the same. I mean think about it. DDR3 Jedec calls for 1.5V. There's no way Intel is gonna run these new cpus at a minimum of 1.5V.

It's just like AMD cpus, if you crank too much vdimm on an AMD setup, you can damage the cpu. And it's not just 939, It will still happen on an AM2 cpu. It's just that DDR2 runs on much lower voltages than DDR1, and gave us much more headroom.

While 1.65V is not the optimal voltage for this to occur for an overclocker, the whole issue is still being blown out of proportion.
I agreed.
AMD cpu has memory controller, and they still not support DDR3 yet. (Only In the future with AM3 CPU)
Intel cpu has memory controller too, but It already support DDR3.
I think both of them is pretty much the same, It just Intel is ahead of AMD about this.
by Mussels (October 5th - 3:19 AM) - Reply
by: phanbuey
why can't they put a resistor(or something) between the ram and the cpu mem controller that will lower the Vdimm for the controller?
they could. lets hope they do.
by phanbuey (October 5th - 3:27 AM) - Reply
by: Mussels
they could. lets hope they do.
Those thieving bastards.
by Mussels (October 5th - 5:23 AM) - Reply
by: phanbuey
Those thieving bastards.
wait, what
by buggalugs (October 5th - 5:49 PM) - Reply
I think people are getting their knickers in a twist for nothing. It just means the memory makers are gonna have to develop ram that runs at lower voltages than what we have today. Although if you already spent a boatload of cash on some DDR3 sticks hoping to use them with i7, that could suck a little.
by Wile E (October 5th - 9:46 PM) - Reply
Let's also not forget that Elpida has already developed chips with Buffalo that will do 2000Mhz CAS9 on just 1.5V.

As I said, I really don't think this is anything to worry about.
by Fitseries3 (October 6th - 7:24 AM) - Reply
whats the mem voltage...



thats an intel board that INTEL was displaying for demo's. im beginning to doubt...
by hayder.master (October 6th - 8:08 AM) - Reply
that's mean lower overclock for ram's, dummm
by InnocentCriminal (October 6th - 10:23 AM) - Reply
2volts is still too high for me. I'll be happy with 1333MHz modules at 1.5volts, I want to see 1600MHz at 1.65 at the very most. All this 2+ volts pisses me off, doesn't it defeat the point of DDR3?
by DaMulta (October 6th - 1:45 PM) - Reply
The trick will be balance. You need to have plenty of cpu voltage going on before you pump up the memory voltage.

Other wise you could have a dead cpu/or/defective cpu that kills ram
or just dead ram.

What sucks is the DDR3 that they have been selling that is 2ghz requires 2v.

Then this new Intel system is triple channel not dual channel. So I don't know yet if it will even work in dual channel mode. And if they put 6 slots on the motherboard if you install 4 chips will that take it back down to single channel/dual channel mode that's something that could really be interesting. That's the information that I would like to learn.

by: InnocentCriminal
2volts is still too high for me. I'll be happy with 1333MHz modules at 1.5volts, I want to see 1600MHz at 1.65 at the very most. All this 2+ volts pisses me off, doesn't it defeat the point of DDR3?


What about people with a 500usd set of ddr3 that runs on 1.9v 2.0v already in their machine?? Because Intel wanted to have the latest and greatest memory speeds with current chips......
<---system specs
by Mussels (October 6th - 4:28 PM) - Reply
i've seen it confirmed, and i can answer your channel questions.

2 sticks WILL run in dual channel.

4 sticks could either run triple channel for the first 3, or dual for all 4. I would assume intel would choose dual channel, due to the balanced performance. (current 775 chipsets can run 3 sticks, where only two are dual channel and the third runs slower, so its easy to assume the same tech could exist in core i7)
by DaMulta (October 6th - 4:36 PM) - Reply
Do you have any links or did you talk/see with someone that has one.



And are you sure about 3 chips in current 775 chips?
by Morgoth (October 6th - 4:48 PM) - Reply
Nehalem supports single, dual, triple
51 to 76 of 89 | Go to Page 1 2 3 4    Previous | Next
Post your own comment
3598 Users online, 2.61 mbps
Quick Search
Already a member?
Username:
Password:
Register Here!
New Forum Posts
09:09 by HammerON
[WTB][US] LGA 775 Quad (0)
09:04 by kurosagi01
Razer Pro Click v1.6 Mouse (12)
09:02 by eidairaman1
Socket A Club (489)
09:02 by sinar
4ghz Club (701)
08:58 by eidairaman1
Very dissapointed in Palit! (22)
08:54 by athenaesword
*GTX 260/280 Unofficial TPU! Thread* (1741)
08:53 by fairytale00
Good black friday sale 2009! (6)
Last Articles

Popular Articles