45nm Intel CPU Failure Resulting From Loadline Calibration Enabled

[SiGfever]

Just slightly above what the real vCore is, if you get 1.41v idle/load by sensors, aim for around 10mV-20mV (0.01 - 0.02v) above the actual vCore. I know the P45 boards don;'t have vTT monitoring which makes determining it a little harder but if you are using LLC I would set the vTT to around 1.38 - 1.40v max and assume a slight overvoltage due to LLC which is safe to say. Otherwise if no LLC enabled set it around 1.44v. The Rampage Formula overshoots by about 40-50mV with LLC enabled. It's safer to assume this is true when you can't monitor it via sensors. vPLL would probably want to be around 1.60-1.64V. vPLL has a very easy method to gauge whether it's right or not with Prime95. All you need to do is monitor the difference between both threads for either core and how long apart they take to complete each loop. Dual core makes it a little tougher since OS related usage can effect the accuracy, but close what you don't need running while Priming and you shouldn't get too much error in ur results from outside system resource use.

on my 65nm Q6600 G0 I use 1.4375v vCore in bios which is 1.41-1.42V real with LLC enabled under load. vTT is set to 1.42V in bios which is 1.46V real, and vPLL is set to 1.62V in bios which is 1.68V real on my board. GTLREF is set to 0.65x for CPU, and 0.63x for NB. vNB is set to 1.47v in bios which is roughly the same real. Heres a quick Everest sensor paste of mine with my voltages. Southbridge PLL = SB Sata voltage in case you don't know. CPU is running on 8x multi and 3.45GHz, FSB at 430MHz, RAM at DDR2-1149 5-5-5-15 ( 4 x 1gb sticks PSC chips I believe) and PL 7 with Phase Pull ins on A1, A3, B1, B3 to PL 6. DRAM Static Read enabled, AI Clock twister at Stronger. DRAM Voltage at 2.26 or 2.28V in bios, and DRAM Controller REF +20mV

Motherboard Name Asus Maximus Extreme / Maximus Formula / Rampage Formula

Temperatures
Motherboard 28 °C (82 °F)
CPU 21 °C (70 °F)
CPU #1 / Core #1 31 °C (88 °F)
CPU #1 / Core #2 34 °C (93 °F)
CPU #1 / Core #3 34 °C (93 °F)
CPU #1 / Core #4 38 °C (100 °F)
North Bridge 38 °C (100 °F)
South Bridge 40 °C (104 °F)

Voltage Values
CPU Core 1.42 V
+3.3 V 3.18 V
+5 V 4.94 V
+12 V 12.04 V
+5 V Standby 4.82 V
FSB VTT 1.47 V
North Bridge Core 1.49 V
South Bridge Core 1.14 V
South Bridge PLL 1.63 V
DIMM 2.30 V
DIMM VTT 1.15 V

So if one thread is slower than the others, say core 1 finishes after cores 0,2,3 what adjustment would you make?

[mikeyakame]

I'd probably increase vPLL one notch and if necessary adjust NB or CPU clock skews + or - 100ns.

[Zeus]

LLC on my P5Q-E was a misery.

Disabled the Vcore dropped and drooped quite a lot but with LLC enabled Vcore actually went up under load!

All nice and dandy, it was real stable, running Prime, Orthos, 3D, everything i threw at it at relative low Vcore.

But then, out of the blue, while surfing the i-net, reading mail or other minimal load on the cpu it just bluescreened, sometimes even just at the end of loading windows, it already gave the desktop and then it bluescreened.

Hell, it even bluescreened while installing windows!

It took me a while to find out LCC was the culprit.

On this Rampage Extreme i have now, LCC is set to auto and Vcore is solid as a rock, hardly drooping and definitely not going up under load.

On my P5K-E wifi, LCC worked just fine, it reduced Vdrop and Vdroop, it never bluescreened at idle once because of it.

And...it never killed one of my cpu's either.

[mikeyakame]

Could be older revision of board or broken-from-the-box voltage regulating components. 1 faulty or dying capacitor could be enough to trigger such unpredictable behaviour

[BenchZowner]

Only board that gave me problems with LLC enabled ( performance ) was the P5E64 WS Evolution which is known to have a problematic LLC implementation.
Other than that, I had no single glitch, no instability with every single motherboard listed here:
Asus P5K Deluxe ( LoadLine Calibration Enabled... it was called CPU Voltage Damper back then )
Asus P5K3 Deluxe ( LoadLine Calibration Enabled... it was called CPU Voltage Damper back then )
Asus Maximus Formula
Asus Blitz Extreme
Asus P5Q3 Deluxe
Asus Maximus II Formula
Asus Striker II Extreme
Gigabyte GA-X48T-DQ6
Gigabyte GA-EP45T-DS5R
Gigabyte GA-EP45-DS4

No problems at all, no reboots, no bluescreens, nothing, nada!
And for Christ's sake, you can't blame LLC for everything. Especially if you're running a nFORCE 680i/780i/790i motherboard because their f* buggy and hard to tweak for most people.

[Leeghoofd]

I only had issues with LLC on the Maximus Formula SE and the P5Q-E... I think that AUTO voltages when OC'ing are way more dangerous to use than LLC enabled or disabled...A digital multimeter is your best friend when OC'ing and discovering a mobo.

In my honest opinion loads of users are obsessed with running the lowest Vcore possible... what's the difference between 1.2 volt and 1.1975 ? Secondly like Benchzowner state each mobo has it quirks and can be a total dog to setup...

Easiest to rule out LLC is to run your stability test with it enabled... then let the pc run at idle and if it BSOD's all of the sudden, easiest thing to do is to up the Vcore and put LLC off...

Regarding damaging the CPU I don't think many users will keep their CPU's for years to notice it and hopefully most will keep the OC based at respectable voltages and temps... as I've seen some wanting clocks way out of reach : due to excessive volts combined with super high temps not aided by their poor air cooling setup...
Some users learn from their mistakes, some will never learn and always will blame something else but themselves... and ofcourse there's always the bad luck to have a dodgy CPU or mobo... but not like some users that have tested dozens of CPU's and can't even get a normal OC stable...

Live and learn... Oc'ing has become more of an art lately, then just whacking up volts... more is not always better....

[nugzo]

If you set vTT higher than vCore then LLC will never damage a CPU. vTT acts as a transient overshoot buffer to protect the CPU, especially during power on when it spikes the highest. Problem is everybody wants to run low vTT without actually knowing what the hell it's designed to do. Even ANANDTECH I would have thought would have known this when they made these ridiculous claims. 45Nm chips are more delicate when it comes to instanteous overshoot, if you are running a good PSU that has excellent voltage output then you should never fry a CPU up this way. A poor PSU and low vTT could easily result in damage to a 45Nm chip at higher than 1.45V vCore.

My vtt is 1.12v in windows. If i raise it any higher, i get data corruption. Nero 8 wont install, RAR files have issues, installers wont work etc. Why is this?

[mikeyakame]

gtl reference voltage i guess. Must be using at least a 0.80x multiplier otherwise it would explain the problem if its using < 0.75x, as such a low vTT needs at least 0.70-0.75x reference voltage of vTT for clocks to be valid and data also

[SergioRoadster]

What I don't get is those warnings about voltage jumping with LC enabled. Everytime I stress the CPU the voltage doesn't change not even a percentage on CPU-Z. Or does it only show if you're using a voltmeter?

[dinos22]

i have questioned this outlandish claim by Kris as soon as this article went online

i dont know what is going on over there but in the past year there have been some articles written which clearly dont represent the reality. I guess you could say that theory and practice still have a little while to catch up to each other heh....if the theory is even right

[platojones]

What I don't get is those warnings about voltage jumping with LC enabled. Everytime I stress the CPU the voltage doesn't change not even a percentage on CPU-Z. Or does it only show if you're using a voltmeter?

Happens way too fast to register on CPU-Z...it is jumping every time you load or unload the CPU, but for a very short period of time.

[platojones]

i have questioned this outlandish claim by Kris as soon as this article went online

i dont know what is going on over there but in the past year there have been some articles written which clearly dont represent the reality. I guess you could say that theory and practice still have a little while to catch up to each other heh

Nothing outlandish about it...unless you are claiming Intel doesn't know anything about how their CPUs and Chipsets are designed to work and why.

[dinos22]

Nothing outlandish about it...unless you are claiming Intel doesn't know anything about how their CPUs and Chipsets are designed to work and why.

yeah that's what it is
a theory that doesnt work in practice the same way
same as high VTT which clearly doesnt do anything on dual core CPUs as BZ has showed

[platojones]

yeah that's what it is
a theory that doesnt work in practice the same way
same as high VTT which clearly doesnt do anything on dual core CPUs as BZ has showed

Well, you are wrong. It's to compensate for the highly inductive effects of a CPU...acts like spring. It's obvious you have no clue what you are talking about..and BZ is some dude who could claim he's running 10,000 v thru his CPU 24x7 and you would believe it...I'm buying his schtick like I'm buying stock in Lehmann Brothers on Monday...geez. Just because you can't see something doesn't mean it doesn't exist...this stuff is basic Electrical Engineering 101...and know what you are talking about before you speak:

http://en.wikipedia.org/wiki/Electromagnetic_induction

It's a LAW...not a simple theory...and it's been around a few years before you were born...so it has a bit of a track record behind it...

[dinos22]

what can i say i am practical and if it doesnt "show" in tests i have no reason to believe that the thresholds stated are correct

is that really that difficult to comprehend

you understand that you should be able to "prove" any theory in practice with at least 95% confidence

where is the proof

[platojones]

Yes, I do understand you...so when you close your eyes, the world ceases to exist...got it.

[dinos22]

Yes, I do understand you...so when you close your eyes, the world ceases to exist...got it.

excellent finally we put this matter to bed
now do you run your setup with LLC enabled or disabled?

[platojones]

excellent finally we put this matter to bed
now do you run your setup with LLC enabled or disabled?

So we did ...LLC disabled, of course...and no, I won't bother asking you the same question...I'm a pretty good guesser

BTW, I don't care what anybody here does to their CPU or MB here...in fact, it's fun seeing people go beyond the specs...but there are a few people here that want to know the risks as well as the rewards of OCing...and claiming stuff that is just not true doesn't help anybody. I think as long as everybody is informed about the risk/reward equation involved in massive VTT overvolts or enabling LLC, no worries...just don't try to claim that these things are perfectly safe when they aren't...if you don't care what happens to your CPU and MB because you love buying a dozen CPUs and MBs at a time and re-installing Vista 4 times a day just to prove you can OC your Q9450 to 5GHz for 10 seconds before your rig crashes and dies...then just say so. That's cool with me.

So no, it's not cool to say "any OC is unsafe, so why not disregard all of the science behind it and go for broke"...there are lines and there are LINES that are crossed while doing this stuff...it helps to know the what's and why's behind those lines....more knowlege is always better than less knowlege.

[moloko]

platojones you're obviously a proffessor of computer science and electrical engineering so I definitely wouldn't want to argue with your theories, the same way that netburst was meant to theoretically reach 10GHz. But for the sake of not looking like an utter douche in the eyes of the experienced overclockers, have some respect for Dinos22 and also BZ.

We, the overclocking community, operate on known values that exist within our testing and any limitations on this testing. As such, the testing which has been done shows little to no degradation from use of high VTT, PPL, vCore etc (keeping in mind cooling is sufficient for the value) and testing to date has returned much the same state of affairs for the use of LLC.

Theories are great, don't misinterpret what's been said in that regard, but if the world was to operate on just theories alone... well we'd be lacking a lot from the present.

[dinos22]

So no, it's not cool to say "any OC is unsafe, so why not disregard all of the science behind it and go for broke"...there are lines and there are LINES that are crossed while doing this stuff...it helps to know the what's and why's behind those lines....more knowlege is always better than less knowlege.

well you are on Xtreme Systems where we go beyond manufacturer's specifications right

please check your browser again <<< NOT Anandtech forum mmmmk

[platojones]

well you are on Xtreme Systems where we go beyond manufacturer's specifications right

please check your browser again <<< NOT Anandtech forum mmmmk

As I said...you can do whatever you like to your CPU/MB. But, then again, where is the charter that says you have to go beyond the functional specs to achieve your stable OC and to have an opinion here? I didn't see it...maybe you could point it out to me

[dinos22]

my point is that you are already going beyond manufacturer specifications but overclocking your system. You are guilty of that too. What some of us do here is show people that would not want to test on their system what happens after that. You insulted BenchZoner who contributes a lot of very useful info for the overclocking community. good on you . Why dont you teach us some more so that we dont make fools of ourselves....maybe drop a PM to fugger to close down the forum too while you'r at it

[platojones]

Over-react much?

Did you even read my posts? I think not. As I said, I don't care what you do to your system, just know what you are talking about...and no, I haven't insulted anybody...only questioned their methods...and how much they have helped is a very subjective matter...helping by displaying a highly unscientific series of vague, uncontrolled tests (and that is being charitable..we have no real proof that these tests are actually being conducted, though I am sure he will assure that they are)...Guess we can't have that in toyland, can we?

[dinos22]

right so please explain to me why a guy would step in here that has built a dozen machines past year say that he's used LLC on all his builds without a single issue. This is an extremely small example of the enormity of usage of LLC on this forum alone right. Can you please teach us more about the effects of LLC and what will happen to our systems then and how we would be able to properly determine/test the effects because we are all trying to learn something new every day. What would benchzoner have to do with his 5xrigs that have been hammered now for 2 months straight with different setups including high VTT, high VCORE, high PLL waaaaaaaaaaay beyond Intel specifications and what any user on XS dares use for 24/7 and some even for benching machines, without side effects of any kind........what do you suggest he does different.

we would love to learn something new from you if you actually have something to contribute. Afterall this is what xtremesystems is about>>> contributing right

[edit] better yet, since you are yet to make a single thread here why not do something constructive with your time and maybe do some testing we can take something from and apply in our machines

[[XC] 4X4N]

Just read through this entire thread and just thought I'd add my 2 cents. As you can see by my sig, I'm a cruncher. I run all my rigs 100% 24/7. I overclock all of them. I run with LLC enabled all the time. I've had an E6300, 7x500 at 1.5vcore with LLC enabled on an Asus P5B Deluxe for well over a year, maybe 2, can't remember when I got it. I have had the rig in my sig running for over a year also with LLC enabled. I'm currently running an ES Q9650 since June at 9x445 1.3vcore with LLC enabled. That is just a few examples. I've built dozens of pc's, overclocked them all, used LLC where possible, and have never killed a chip or had any degradation.