High memory voltage on Phenom...

[justapost]

How many B2 owners will try this to get an B3 via RMA once it is available?

[Pianist]

so are you saying that I shouldn't run my geil 4-4-4-12 @ 2its rated 2.4 V? Ive been leaving it all on auto since I read this thread lol, which is 1.8, and cant get it anywhere near 1066 speed.

[dehx]

How many B2 owners will try this to get an B3 via RMA once it is available?

I know I am going to try to.

[su35bm]

Count me in.

[saaya]

wait, so k10 has problems with 1066, has problems with certain mem timings, and it cant even run more than 2.2v vdimm?
not to mention the tlb bug and nb/l3 cache bug... and it cant clock as high as c2quad... man... this is awfull

everytime i convinced myself to buy phenom again i hear yet another bad thing that convinces me not to buy one

I know I am going to try to.

you plan to rma a processor in case you kill it with high vdimm?

[LowRun]

everytime i convinced myself to buy phenom again i hear yet another bad thing that convinces me not to buy one

I'm in the same boat

you plan to rma a processor in case you kill it with high vdimm?

More like after using is ram @ default

[AlabamaCajun]

I would say stay below 2.2V, you are not going to gain enough from a slightly lower latency with these early Phenoms anyway.

Now that Bingo13 toasted his I wonder if the Egg will consider by RMA

[Aussie FX]

What are some of you guys going on about?

It was posted earlier that with the new IMC you don't need to run high volts to get performance.
2.2v seems to be plenty to get 1100mhz where before 2.4v was needed for the same frequency.
It is more efficient (IMC) so I would think that would be a good thing.

Seems like a bit of commonsense has flown out the window and been replaced with FUD.....

[saaya]

common sense?
this is xtremesystems!
we want the best possible memory performance and not "just about enough"

memory needs high vdimm to be maxed out, 2.2v is by far not enough...

[KTE]

I can run 2.2V at 1200 4-4-4-4 PL5 just fine. But these RAMs are the best I've had and most other 1066 RAM do not even run 1066 4-4-4-4 PL5 at 2.3V let alone those wanting to go bench 1300 and requiring 2.35-2.8V. So to bench latencies/maximum bandwidth/performance and show the Phenom platform in the best light as is done with all packages you would need to have high VDIMM available and higher speeds supported when overclocking.

Right now I don't even know if its just running the memory as high as 1000 which makes the system shutdown when oc'ing or what since it will not even boot at those speeds and requires a CMOS clear to start with 1066.

[saaya]

any updates?
so we have 2 people who had several phenoms die on them.
Can we get any more details about what exactly happened?
What vcore? what vdimm? what nb voltage? what boards?

I killed my 9600 in a matter of minutes at 2.38V real while running under load, now the 9500 took a couple of days

On K8/K9 all you had to do to make sure you dont fry the imc with high vdimm was raise vcore.

2.9 V -------- 1.125 V (min)
3.0 V -------- 1.175 V (min)
3.1 V -------- 1.225 V (min)
3.2 V -------- 1.285 V (min)
3.3 V -------- 1.335 V (min)
3.4 V -------- 1.375 V (min)
3.5 V -------- 1.425 V (min)
3.6 V -------- 1.475 V (min)
3.7 V -------- 1.535 V (min)
3.8 V -------- 1.585 V (min)
3.9 V -------- 1.625 V (min)
4.0 V -------- 1.675 V (min)

couldnt it be that we just have to raise vcore or the nb voltage to get phenom to work with high vdimm?

[informal]

couldnt it be that we just have to raise vcore or the nb voltage to get phenom to work with high vdimm?

Yep,could be,BUT ie. MSI's AM2+ mobos in the present BIOS versions don't have NB voltage control(essentially they do,BUT it in fact raises the chipset voltage instead of NB on-die... )
So,we need people who are willing to try it out on GB or DFI.The bad/ugly is that they could still fry the CPUs if the theory is not true...

[justapost]

I know I am going to try to.

Count me in.

Hmm, the B3's will have a new naming 9x50. I assume they will have a big amount of B2's on stock, so the chances are good that you'll just get another B2.

[Lightman]

I was running 2.42VDIMM for at least 8 hours so far..... (1.2Vcore)

Might be that 9500 are stronger than 9600 or simply it depends not only from VDIMM but also from amount of sticks filled in.... I'm running two.

Will try more during weekend...

[BadNizze]

Well I dont have my Phenom running know, but i did bench and play aorund for houres at 2,43 on my gigabyte board. To 8h+ but one houre there and another the next day, will try with my DFi board after the hollidays,

[justapost]

@BadNizze & Lightman: 9500 needs more time.

I killed my 9600 in a matter of minutes at 2.38V real while running under load, now the 9500 took a couple of days but that bastard needed to die.

[Blacklash]

Glad I purchased ram for my Agena rig that only needs 1.85-1.90v.

[Lightman]

@BadNizze & Lightman: 9500 needs more time.

I will give it more time during Christmas

[KTE]

I doubt they'd die that quick because even I've had it 2.3V at 1.232V VCore. But if they can get the dreaded 1.:1.266 divider working then I can test higher VDIMM. Right now I don't need any more till DDR2-1200 which is no where near what I can run with these low divders.

[bingo13]

I killed these CPUs on both the MSI and Gigabyte boards. 1.344V or so on the CPU, 2.38~2.42V on the memory voltage, NB speed around 2000, and the key it appears is that I was using the 1066 setting and clocking up from that point. This has occurred with a couple of other people. AMD is looking at one of the dead CPUs at this time, but they continue to advise us not to run memory voltages above 2.2V for any length of time.

The good news is my 0744 9600 has not died after a week of trying to kill it, but I have noticed a little degradation in memory performance now. It could be I had a couple of weaker than normal CPUs (understand the last production batch is really strong). The main reason for the potential kill problem is something we will discuss shortly.

[Brother Esau]

@bingo13...............if I were to opt to get one of these chips would you say the 9600 is the better move?

[JumpingJack]

can anyone link me to an article or somthing that explains this more in depth? i dont wana bust my cpu!

http://www.jedec.org/

Memory is kinda loosey goosey ... there is a consortium organization, whose membership of course includes most all the major memory makers as well as Intel, AMD, nVidia etc. You can find the JEDEC standards for DDR2 memory at that web site.

The voltage for JEDEC standardized DDR2 is 1.8+/-0.1 Volts for DDR2-800, as I recall (and I may be wrong), JEDEC never ratified any specs for memory above this speed... for DDR2 anyway.

Anand has summarized the JEDEC specs for DDR2/DDR3 in this article:
http://www.anandtech.com/printarticle.aspx?i=2989

This of course does not stop memory makers from trying to woowooo your wallet with promises of faster memory or lower latency but outside of the standards -- you will commonly find low latency DDR2 but speced out at 2.0 or higher volts (operating voltage). nVidia even skirts this by the SPD to store special 'overclocked' settings for the memory used in it's SLI certified memory brand.

Nonetheless, Phenom, with the IMC, was likely built to withstand x% above and beyond the JEDEC spec, however it appears that limit is at or below 2.2 volts... so if you are looking for 1067 memory, or even low latency DDR2-800 ... pay attention to the voltage spec.

[JumpingJack]

How many B2 owners will try this to get an B3 via RMA once it is available?

AMD will know if you ran it out of spec I suspect ... they will decline the RMA most likely.

[JumpingJack]

so are you saying that I shouldn't run my geil 4-4-4-12 @ 2its rated 2.4 V? Ive been leaving it all on auto since I read this thread lol, which is 1.8, and cant get it anywhere near 1066 speed.

I would not run it at 2.4 V given the information kindly passed in this thread by various users.

If you leave it in auto, the memory will default to the JEDEC standard voltage and speeds that will operate at that voltage -- if you want the faster capability, then a) you need to manually set it, or b) if you have an nVidia chipset and the memory is SLI capable, then let the nVidia BIOS SLI option set it for you.

[KTE]

I killed these CPUs on both the MSI and Gigabyte boards. 1.344V or so on the CPU, 2.38~2.42V on the memory voltage, NB speed around 2000, and the key it appears is that I was using the 1066 setting and clocking up from that point.

bingo13, Which MSI board let you run 1066 divider?