Core i7/X58 Overclocking Thread

[SteveRo]

This is using the 'SpeedStep/ Turbo Mode"...(of course).....


The batch is from the 3838A series..I know some folks have been only able to get this one to 3.8 or 3.9 max

Keep in mind this is on air..and nope , my load temps are not 70C or 80C or 60C..which for me is unacceptable.

Some reviews seem to state that "Turbo Mode" offers very little improvements..I disagree...also this function is interesting because when idle, the CPU is clocked at 2.4 Ghz... so when not in use the CPU is not pushing wattage......very cool

Cheers, JRM /Hermes

Congrats!! - 4.2 is no easy feat on air even with ht off.
By the way - British Columbia is a beautiful part of the world - i particularly like Vancouver Island - Victoria is one of my favorite places.
The seaplane ride up from Seattle was spectacular.
When you have a moment - show us some screen shots under load - prime, intel burn in test or LinX - your choice - and with everest or realtemp running.

[mojooh]

Anyone to advice me how to run my memory at 1333 without changing the bclk??
is it the qpi voltage that would make a difference?

p6t dlx and 920

sorry for posting again..

[SteveRo]

Anyone to advice me how to run my memory at 1333 without changing the bclk??
is it the qpi voltage that would make a difference?

p6t dlx and 920

sorry for posting again..

good morning, qpi v up to 1.35 and/or mem v up to 1.65 to stay within the intel max. mem multi x bclock to get to mem speed.

and this is important - you need to set you uncore multi to twice your mem multi, actually twice your mem multi plus 1 click up is even better.
However - careful - max uncore speed is 4000 may be just a bit higher. Hope this helps.

[Rammsteiner]

good morning, qpi v up to 1.35 and/or mem v up to 1.65 to stay within the intel max. mem multi x bclock to get to mem speed.

and this is important - you need to set you uncore muti to twice your mem multi, actually twice your mem multi plus 1 click up is even better.
However - careful - max uncore speed is 4000 may be just a bit higher. Hope this helps.

Im so glad I didnt go with those OCZ Blade's. The specs look awesome but yeah, if you Uncore cant run it you'll have to downclock the RAM anyway (right?). Although this might allow for tighter timings but well

Instead I got a nice Mushkin kit which had a 100 Euro discount

Anyway, I dont have the parts yet, but to be sure, VTT/QPI or how you want to name it, is best to stay within 1.35V according to Intel, leaving your Vdimm to be max 1.85V (to be sagfe maybe a notch slower), right?

[mojooh]

thanks for the quick reply

What i have tried so far is 3.2ghz QPI (24x) , 2.66ghz uncore (20x) and 1333mhz memory (10x). Bclk always 133mhz.

I did try up to 1.25 qpi v but still had problems passing linx!

So is it just the qpi voltage i should try to raise to get it stable?

[SteveRo]

thanks for the quick reply

What i have tried so far is 3.2ghz QPI (24x) , 2.66ghz uncore (20x) and 1333mhz memory (10x). Bclk always 133mhz.

I did try up to 1.25 qpi v but still had problems passing linx!

So is it just the qpi voltage i should try to raise to get it stable?

At what voltage are you running your memory?
Memory v might need a little boost also.

[SteveRo]

Im so glad I didnt go with those OCZ Blade's. The specs look awesome but yeah, if you Uncore cant run it you'll have to downclock the RAM anyway (right?). Although this might allow for tighter timings but well

Instead I got a nice Mushkin kit which had a 100 Euro discount

Anyway, I dont have the parts yet, but to be sure, VTT/QPI or how you want to name it, is best to stay within 1.35V according to Intel, leaving your Vdimm to be max 1.85V (to be sagfe maybe a notch slower), right?

The jury is till out on the max mem voltage controversy.
INTEL officially told all the memory makers to keep the max v's to 1.65.
That said, there are quotes from INTEL engineers (well one anyway) that says that you can increase mem v as long as you do not exceed .5v over qpi.
So for example, if you run qpi at 1.35 you should be able to safely run mem v at 1.85v.
From what I can tell though - long term effects of any of this high voltage stuff is unknown.

[Rammsteiner]

The jury is till out on the max mem voltage controversy.
INTEL officially told all the memory makers to keep the max v's to 1.65.
That said, there are quotes from INTEL engineers (well one anyway) that says that you can increase mem v as long as you do not exceed .5v over qpi.
So for example, if you run qpi at 1.35 you should be able to safely run mem v at 1.85v.
From what I can tell though - long term effects of any of this high voltage stuff is unknown.

That counts for a lot of things, in the end those max values are questionable too. Not like Ill go and try stuff

But yeah, in the end it's defined by Intel, the same compagny who isnt really happy with people OC'ing.

For example, I guess 1.4V QPI for 24/7 might be less harmfull then an i7 at stock running 95C 24/7, although IIRC that's safe. Oh well, I got the information I needed thus far, now I only have to wait for the parts

[ewitte]

Heat is my problem (stock HSF) at the moment so don't have hard numbers but

19*211 with HT on will boot and do single threaded apps at 1.2v, 1.225v for untested stable (tested 1 minute because of heat)

Lowest stable vcore I've tried so far is 0.95v. 2.67Ghz. Runs 8 threads in prime, didn't crash.

[chance]

4GHZ 1.22v - PRIME95 64BITS 11H

[SteveRo]

chance

Excellent! What are you using for a cooler?

[zlojack]

Batch # chance?

[chance]

chance

Excellent! What are you using for a cooler?

NOCTUA NH-U12P SE

Batch # chance?

3838B079

http://www.xtremesystems.org/forums/...&postcount=448 - (8 HOURS). Trying to go under 1.30v at 4.2GHZ

[dejanh]

This is using the 'SpeedStep/ Turbo Mode"...(of course).....


The batch is from the 3838A series..I know some folks have been only able to get this one to 3.8 or 3.9 max

Keep in mind this is on air..and nope , my load temps are not 70C or 80C or 60C..which for me is unacceptable.

Some reviews seem to state that "Turbo Mode" offers very little improvements..I disagree...also this function is interesting because when idle, the CPU is clocked at 2.4 Ghz... so when not in use the CPU is not pushing wattage......very cool

Cheers, JRM /Hermes

45C load?! Surely, you must have something wrong there...unless you are keeping your computer outside in minus temps (like -40C) there is no way that a 1.45V, 4.2GHz OC with HT on will give you 45C load. Heck, even stock 2.66GHz will not give you 45C load with a 23C ambient.

No way. Completly impossible. What are you real load temps?

NOCTUA NH-U12P SE




3838B079

http://www.xtremesystems.org/forums/...&postcount=448 - (8 HOURS). Trying to go under 1.30v at 4.2GHZ

If you can get 4.2GHz with HT on below 1.30V that will be a record I think. One other thing...are you running XP 64?

[Kanasatake]

45C load?! Surely, you must have something wrong there...unless you are keeping your computer outside in minus temps (like -40C) there is no way that a 1.45V, 4.2GHz OC with HT on will give you 45C load. Heck, even stock 2.66GHz will not give you 45C load with a 23C ambient.

No way. Completly impossible. What are you real load temps?



If you can get 4.2GHz with HT on below 1.30V that will be a record I think. One other thing...are you running XP 64?

No no..running Vista 64... and no this is not BS.. here is what I'll do,,I'll run Everest and put ,Asus Probe 2, Core Temp, 3 versions of CPUz..what else do you need??
A video clip maybe.. my wife on the video as witness,a judge ?

The Core Temps are definitely way up there..which is what I suspected, but the Thermalright is holding its own..no problem

Cheers, JRM/Hermes

[Kanasatake]

an NO the HT is NOT on.. I didn't say that. Indeed I would expect it to be at least in the 50C range and the Core Temps way up there..... I wouldn't dare do that.


Cheers, Hermes

[ewitte]

If you can get 4.2GHz with HT on below 1.30V that will be a record I think. One other thing...are you running XP 64?

ORLY will have to try that myself

[Kanasatake]

Now that he suggested it why not...ok 1.30V WITH HT on.. I guarantee you it WILL get HOT.

Cheers. JRM/Hermes

[ewitte]

My take is if it gets too hot its time to upgrade the cooling not disable stuff in BIOS

[Kanasatake]

My take is if it gets too hot its time to upgrade the cooling not disable stuff in BIOS

Exactly, for some this cooler works exactly like the old Thermalright Ultra 120 Extreme, but with the review from Hardware Canucks, you got the impression and by looking at their temps , I wasn't impressed...thats where I said to myself , like Dejanh says..somethings not right..my experience with the Thermalright has always been excellent.
I just did a CPU stress test and was able to "pump" the temps up to 65C..it took 6 minutes at 100% all cores...anyhow I stopped it...thats too high for me.
With the LCD poster in games at full bore my temps hit 38C sometimes higher , I have seen it as high as 42C with Flight Simulator X, which is a CPU bound game.
I expect this summer that will change significantly .
So sorry Dejanh, the Thermalright Ultra 120 eXtreme 1366 does and CAN work very well. Like I said, that hasn't been the case for many, I don't know why.

So if this post enough isn't "proof enough" ..I'll have to get a thermometer and place it beside the monitor...and no I don't have the computer "outside" in Aklavik..I live in BC, today its 12C outside and mixed sun, ambient temp inside here is about 20C or so.

The great thing about this processor is I can run the games at CPU voltage at 1.0 V and CPU speed at 3.0GHz..and it still is a beast..thats probably what I'll be doing this summer.

Cheers, JRM

[ewitte]

From what I've been reading its suggested they moved the sensor closer to the center of the core so your not actually not any hotter at 80C than you were at 60C on the other chips I have no problems with 80's for stress testing as that would still be quite a bit less for everyday usage.

[Kanasatake]

Congrats!! - 4.2 is no easy feat on air even with ht off.
By the way - British Columbia is a beautiful part of the world - i particularly like Vancouver Island - Victoria is one of my favorite places.
The seaplane ride up from Seattle was spectacular.
When you have a moment - show us some screen shots under load - prime, intel burn in test or LinX - your choice - and with everest or realtemp running.

Thanks , yes British Columbia is really beautiful. Even with "Flight Simulator X" you can fly around.... I made it on my "flight plan all the way to Cache Creek

Ok here is a "stress test" as you can see the temps go way up....past my comfort zone.
I hope this qualifies?

Cheers JRM/ Hermes

[chance]

NOCTUA NH-U12P SE

3838B079

http://www.xtremesystems.org/forums/...&postcount=448 - (8 HOURS). Trying to go under 1.30v at 4.2GHZ

If you can get 4.2GHz with HT on below 1.30V that will be a record I think. One other thing...are you running XP 64?

Is it my impression or user "Kanasatake" is answering my "Dejanh" question? http://www.xtremesystems.org/forums/...postcount=2789

Yes i am, XP 64Bits. If it does 4GHZ at 1.22v or even less (testing) maybe it can do less then 1.30v to 4.2GHZ, we never know ah?

[dejanh]

Is it my impression or user "Kanasatake" is answering my "Dejanh" question? http://www.xtremesystems.org/forums/...postcount=2789

Yes i am, XP 64Bits. If it does 4GHZ at 1.22v or even less (testing) maybe it can do less then 1.30v to 4.2GHZ, we never know ah?

Indeed, I'd be extremely impressed by that. I loaded up 19x222MHz today with 1.3V but it was not stable at all. No way I could run any stress test on it. You got some nice chip there. My chip does 3.8GHz (19x200MHz) at just 1.18V, about 1.26V for 4GHz using 19x211MHz, but if I go with the 20x or higher multipliers for say 4GHz, 4.2GHz, or higher then the voltages jump way up. At 20x200MHz I need about 1.3V to be stable, or rather that was before I set up my new cooling. I can expect that I can drop that now to about 1.26-1.28 probably but I have not tried. Running 4.2 at 1.36V using 21x200MHz on the i7 940.

@Kanasatake - LinX will push your temps higher still. The TRUE does not exactly hold it's own at 4.2GHz with or without HT. Without HT you can run it ok most likely but with HT on there is no way. Not unless you can get chance's 1.30V 4.2GHz I own the TRUE as well, same one as you, and it's sitting in the closet right now. I just had to ask about the temps because people get the wrong impression about what realistic clocks/temps are when reading this thread and then complain a lot about how they cannot get the same results. That's all. I did not mean to sound mean. If you do not do overly intense work that needs all 8 threads then your temps should be ok for day to day use and you should not be worries. Realistically, you got nothing to worry about until you are way up in the upper 80s or even low 90s.

[chance]

Indeed, I'd be extremely impressed by that. I loaded up 19x222MHz today with 1.3V but it was not stable at all. No way I could run any stress test on it. You got some nice chip there. My chip does 3.8GHz (19x200MHz) at just 1.18V, about 1.26V for 4GHz using 19x211MHz, but if I go with the 20x or higher multipliers for say 4GHz, 4.2GHz, or higher then the voltages jump way up. At 20x200MHz I need about 1.3V to be stable, or rather that was before I set up my new cooling. I can expect that I can drop that now to about 1.26-1.28 probably but I have not tried. Running 4.2 at 1.36V using 21x200MHz on the i7 940.

Are you talking about your 920 or 940?