Ultimate Core i7 Overclocking Guide -- We Push Nehalem to its Limits
Posted 02/05/09 at 12:00:00 PM by Gordon Mah Ung
Here’s everything you need to know about overclocking Intel’s new CPU
Overclocking can kill your CPU. It can corrupt your OS, melt your motherboard, and cause you to lose a month’s work or more. Despite those dire orange-alert warnings, however, overclocking has moved on from the Nerd’s Only Club to become practically a mainstream hobby in the last few years.
So why overclock if the risks are so great? For some folks, it’s about bragging rights. Like drag-strip racers who burn up an engine just to set a quarter-mile record, there’s a small community who will overclock a CPU to the brink of destruction just to run a benchmark and take a screen shot of the result.
The bulk of overclockers, however, are more concerned with the cost dividends. If you can take a $300 CPU and make it as fast or faster than one that costs $1,000, the money you save can go toward other components in your system. For these folks, it’s like getting a free high-end videocard.
Whether you’re a cheapskate or a drag racer, you’ll find that Intel’s new Core i7 CPU is unlike any previous Intel CPU, and overclocking this beast requires more tinkering than you might expect. Follow along as we explore what it takes to push this chip hard.
A Brave New World -- The crumbling of the front-side-bus wall means a wholesale change in how you overclock Intel’s top parts
Veteran Intel overclockers know that little about the process changed since the beginning. Like the basics of flying (stick, rudder, ball), all you had to worry about with an Intel processor was the front-side bus, clock multiplier, and core voltage.
Not so with Core i7. With Intel retiring the front-side bus, you’ll need to brush up on your overclocking skills and concepts if you want to get the highest-performing overclock out of the Godzilla of CPUs.
The Base Clock
If you haven’t kept up on current events, you need to know that Intel eliminated the front-side-bus architecture that has connected the CPU to the core-logic chipset since 1978. The memory controller, which used to sit in the chipset, is now integrated directly into the CPU.
Go into the BIOS on a Core i7 and you won’t see any reference to the front-side bus. Instead, it’s now the base clock or bclock. Some BIOSes also refer to this as the host clock or reference clock. On the current i7 procs, the base clock is 133MHz. While it’s not a front-side bus, most overclocking methods will require that you tweak the base clock just as you did with older FSB-based Intel chips and push it beyond 133MHz. One important thing to remember about the base clock is that it is the main reference clock for other components in the CPU— goosing this one setting will also overclock the RAM as well as the “uncore” (i.e., the L3 cache, memory controller, and Quick Path Interconnect, or QPI). We’ll get to more on this shortly.
Doing the Math
One thing that has not changed at all is the multiplier. The Core i7-965 Extreme Edition features a multiplier of 24 and is unlocked so you can move it up or down. The budget Core i7-920 features a multiplier of 20 and is upwardly locked so it will not move past 20. Experienced overclockers can grab a hall pass and skip to the next section, while those who have never done this before will need to stay in the classroom. As it was with the Core 2 and Pentium 4, the overall clock speed of the CPU is derived by multiplying the base clock (formerly the FSB) by the multiplier. For the 965, take 24 and multiply it by 133 to get 3200MHz or 3.2GHz. For the 920, take your multiplier of 20 and multiply by 133 to get 2667MHz or 2.66GHz.
Turbo Talk
One wrinkle to the multiplier is the new Turbo Mode. This mode essentially automatically overclocks a single core of the CPU under certain loads. If you have overclocked a Core i7-920 to 3.66GHz and then you switch on Turbo Mode, the CPU will actually run at 4.03GHz in single-threaded apps. Is it worth it? Frankly, we’re not sure. We are getting to the point where it’s pretty rare to be running performance-intensive single-threaded applications, so the performance boost will be minimal. You do get a 1x multiplier boost in dual-threaded apps so you most games would run at 3.83GHz. Sounds good right?
Our engineering sample Core i7-920 gave us results in line with what others have achieved with engineering sample parts as well as retail parts.
Unfortunately you can’t set your individual Turbo Mode settings on the cheap chips. Intel limits fine-grain Turbo Mode control to the Core i7-965 Extreme Edition. The pedestrian Core i7-920 and Core i7-940 are limited to a single multiplier increase for single-threaded apps, which is of marginal usefulness. It’s also clear that not all motherboard vendors think Turbo Mode is worthwhile. We’ve tested two different Asus boards that don’t implement Turbo Mode the same way Intel does. Instead of letting the user set the individual Turbo Mode settings on an Extreme Edition chip, your only option is to overclock all cores simultaneously.
Turbo Mode is something that should be evaluated based on your needs and the specifics of your overclock. For example, our case study actually found that a moderate overclock with Turbo Mode gave us better benchmark results than a higher-speed overclock without Turbo Mode.
Uncore: Separate but Connected
The Core i7 is a modular design with two main areas, the “core” and the “uncore”. Inside the chip, the actual execution cores that do the heavy lifting are treated as the “core.” The other parts, such as the integrated memory controller, L3 cache, and the Quick Path Interconnect are treated as the “uncore.” Since they’re separate entities, you can overclock the execution cores without overclocking the uncore to the same degree, in theory. This should let you hit higher speeds, since you wouldn’t be running the QPI, memory controller, or L3 quite as hard. In reality, however, it doesn’t work that way. Intel’s non-Extreme Edition Core i7 CPUs offer limited control over the uncore multiplier, so a boost to the base clock boosts the uncore speeds as well.
You need to set your Uncore multiplier to at least twice the multiplier for the system RAM. Also pay attention to the memory voltage. We had to run 1.66 volts for stable performance even at low RAM clock speeds. QPI also needed to be nudged up to 1.3 volts.
One thing to remember as you fumble around the BIOS is that the uncore must run at twice the speed of the system RAM. Here’s where it gets a little confusing. The speed of the uncore is determined by multiplying the uncore multiplier by the base clock. On a Core i7-920 chip, for example, the uncore defaults to 16. The uncore thus is 16 times 133 for a total uncore speed of 2,133MHz or 2.1GHz.
To figure out the RAM speed, you have to take the memory multiplier and multiply it by the base clock. In the case of a Core i7-920 chip, the default memory multiplier is 8. So to determine the main memory speed, multiply 8 by 133 for 1,066MHz. Why aren’t higher DDR3 speeds available? The highest official memory speed of the Core i7 is DDR3/1066. You can overclock your RAM to higher speeds, but depending on the motherboard, the only way to accomplish a memory overclock will be to crank up the base clock for the CPU—unless you own an Extreme Edition CPU.
The take away here is to remember to keep the uncore speed at twice the speed the RAM runs. If you plan to run DDR3/1600, you’ll need to run the uncore at 3,200MHz. On a Core i7-965, you can run that speed without overclocking. On a Core i7-920, you’ll have to overclock the base clock to get the RAM at that speed.
If your machine will...
Submitted by JCCIII on Tue, 11/03/2009 - 7:14am
"If your machine will withstand a couple of hours of Prime95, you’re doing good. If it’ll run over night, it’s bullet resistant. If it’ll run overnight in the middle of the summer, in an 80 degree room, well, you really have something to brag about."
Is not that the truth!
How many times have I wondered if my EE was one the verge of a meltdown this summer?
?'s about motherboard and overclocking
Submitted by sinister213 on Tue, 10/06/2009 - 9:49pm
I recently purchased a dell studio xps 435t with the i7 processor (920) and12 of ddr3 at a major discount through my school. I plan on making it into a capable gaming system with a few upgrades. Efforts to overclock have been futile. Apparently Dell bios negate the ability to change system settings like clock speed. I opened the towere and there is no lead to what type of board dell supplied me with other than 0X501H. Today i bought an evga classified x58 w/triple sli and a corsair 1000 watt power supply (still waiting for gt300).
However, the salesman at comp usa was bashing the evga board and pushing the p6t. Is the p6t the better board and what are the downfalls if any on the classified board? how capable are they in terms of overclocking? Which board will offer better support for a dx11 gpu? I plan to have all mods done by a local company. they have asus signs everywhere so i know their opinion will be biased. any unbiased input would be greatly appreciated. thanks
Which CPU temps do I look at (CPU or CPU Core)?
Submitted by MasterShake on Thu, 09/03/2009 - 3:55am
I am overclocking my i7 920, and am running Everest to measure the CPU temps. Which temp should I be looking at? I get a status of each core and the overall CPU.
The cores run at about 70 degrees, while the CPU itself shows about 60 degrees.
Thanks!
@ Woofa
Submitted by ghot on Wed, 08/26/2009 - 9:37am
Yes, I'm an AMD fan...whats your point? I just got an AMD 955BE to 3.7ghz in 5mins on air...and I haven't even dug into the BIOS, really. Low voltages stock RAM timings...and yes 24 hour Prime95/Blend stable IN an 80 degree room with max load temps at 40-41C. Gimme a month and it'll be at or near 4Ghz on air and 24/7 Prime stable. Yes I do build budget systems...is there something wrong with that?
My earlier comment was just to point out that AMD came up with the architechture, Intel just copied it, threw wads of money at it, and other than ungodly memory benchmarks...it doesn't do much better than an AMD, yet costs alot more.
I could buy Gucci jeans too, but what's the point? Levi's do the same job for a lot less.
I'm really kinda of sick and tired of Intel FAN BOIs and all their BS..to be honest. AMD with their 3 Fabs smacked down Intel badly a few years ago...and they will do it again...why....because they are better than Intel at designing CPU's.
Be my guest Woofa...keep buying Intels...I'll stick with AMD and a vacation instead ;)
Only dead fish....."go with the flow" :/
Take an OS, and edit out all the efficiency, and what you have left is a post-XP Microsoft operating system :)
Congratulations, another
Submitted by marscay on Wed, 10/21/2009 - 10:56pm
Congratulations, another blind AMD fanatic talking absolute BS.
You need a month to get 4ghz stable 24/7??? I was at 3.9ghz 24/7 stable on a Q6600 2 years ago......yes well done AMD and their followers, a truly amazing feat. AMD were well in the lead a few years back but apart from budget rigs they just cannot compete right now.
Anyone who uses their pc for more than reading blogs is nuts to consider PhenomII over i7/i5, for rendering/encoding duties an i7-920 which only costs another £60 verus the top Phenom II chip (where are you vacationing with that saving, basingstoke for 1 day?) absolutely destroys it.
The only reason you're sick and tired on Intel fan boys is because you know it to be true and you're backing the losing horse to save a bag of peanuts.
Are you serious?
Submitted by pianoman6954 on Sat, 10/10/2009 - 3:33pm
Intel's entire wafer process is different than AMD's. The QPI is multidirectional and modular without special implementation or seperate chipset intervention, making it nearly infinitely scalable. So its a little hard to believe the architecture is copied. A similar method to handling RAM is far from a complete architecture copy. AMD hasn't been on top since the Pentium 4 was Intels best offering. AMD didn't get it's reputation for being the fastest or the most efficient, they are cheap, that's all.
overclocking a core i7 920
Submitted by studlymetts on Thu, 04/16/2009 - 1:37pm
i'm going to buy a core i7 920 with 12 GB DDR3 RAM
What other customizations do i need to make my core i7 920 run at 3.2 GHz?
and do I need to buy a better CPU cooling fan after overclocking it to 3.2 GHz?
Re: overclocking a core i7 920
Submitted by vfxcorp on Thu, 04/23/2009 - 4:55am
Re:studlymetts - Yes, to overclock and to deal with the resultant heat increase you will need two more things: a better thermal paste and a better air cooler than the stock items that come with the Intel processor. Through my own experimentation and web research I have found that OCZ Freeze is the near top best thermal paste and performs 1-2 degrees better than the popular Artic Silver 5 and has a much shorter cure time. A better air cooler can make up to a 10-15 degree difference under load compared to a stock cooler. Based on BenchMarkReviews.com, for a 1366 processor, the Thermalright Ultra-120 Extreme and also Prolimatech Megahalems are the 2009 first quarter bests. Ufortunately Newegg is currently not stocking the Thermalright Ultra-120 Extreme or Prolimatech Megahalems, but I found the Thermalright Ultra-120 at Microcenter.com.
Heat trade-off for Hyperthreading ON or OFF
Submitted by vfxcorp on Thu, 04/02/2009 - 2:10pm
Great OC article. I wished they had gone more into the value or lack of value for Hyperthreading being ON or OFF.
I'm overclocking an i7 940 at 176 QPI (4048 Mhz) with some intense Vue 7 renders working up to 72 C. I did notice that by turning off Hyperthreading in the Bios, it saves about 10 degrees! That's a big heat difference. With Hyperthreading ON my Cinebench R10/64 scores are Single 5411, Multiple 21951, and OpenGL 8966. With Hypertheading OFF the Cinebench seems to do a little better on the Single, but does drop a little bit in Multiple: Single 5450, Multiple 19589, and OpenGL 8920.
Would Hyperthreading show more value in video production software I wonder? What software currently takes advantage of Hyperthreading?
Intel’s Desktop Control Center
Submitted by justme on Wed, 03/18/2009 - 12:14pm
where do i find this???
Do you have an Intel DX58SO?
Submitted by Woofa on Mon, 04/13/2009 - 5:09pm
If not then it's of no use to you, at least speaking strictly i7. If you do then simply go to intel.com and figure it out, it's not that hard as you'll find it where you find the other downloads for the DX58SO.
3.3ghz with Aircooling
Submitted by Pball1224 on Mon, 03/16/2009 - 9:30am
With turbo mode still turned on, the best speed I can reach and be stable at is 3.3ghz with a BCLK of 160mhz. Any faster and the temp gets too high and the sys restarts, usualy within 3 minutes of starting a test.
Hmm
Submitted by Woofa on Thu, 02/26/2009 - 11:54pm
I was wondering while reading this why you got such limited results. Not that they were bad, just not noteworthy. Finally there it was, the motherboard you used, "Intel DX58SO “SmackEDover”." I've got nothing against Intel boards, I'm replying from one of my systems that happens to use a Bad Axe 2 with a Q6700 running at 3.6 on air that has been crunching away on distributed computing projects for over a year almost non-stop and zero issues. However their performance "Extreme" boards are not quite up there with many other motherboards when it comes to overclocking and from my experience the DX58SO is not an exception. The Bad Axe 2 I'm using is probably the best of the Intel Extreme category. Actually, strike the probably. I have my hands on a lot of hardware since I build systems and the Bad Axe 2 is the only one of the extreme boards that has been able to get in the same OC ballpark as a number of other makers like Asus, EVGA, Gigabyte and others from the same generation. It's old hardware now and there are other boards I could put this Q6700 in now that would perform better but it's been established so I leave it as there is no sense fooling with this unless the board takes a dive. I built a new i7 920, EVGA X58 on air for myself two months ago and it wasn't even breathing hard at the 3.66 you recorded and there are plenty of other examples like this out there using EVGA X58, Asus P6T, Gigabyte, etc. mbs. Those aren't just bragging number either as they are well documented and I won't even talk about what my 920 is clocked at right now. I have to wonder why you chose this comparison using the DX58SO. Maybe because Intel supplied the whole kit? I don't know. Anyway put that 920 in a P6T or EVGA X58 and you'll get a different OC experience.
As for the AMD fanboy ghot and the ridiculous comments. There's nothing much more entertaining than someone who doesn't have that much knowledge spouting off about the brand they blindly follow. I don't build just Intel systems nor do I blindly follow either cpumaker, in fact for a few years my systems were mainly AMD based. That situation may return someday, but right now if you aren't budget building Intel is the choice. I use and recommend AMD cpus for media centers and they work excellent for that purpose, workstation environs or top gaming machines they aren't the top dog and haven't been for awhile. ANY Core i7 absolutely crushes any other choice out there right now.
Core i7 Water Cooling
Submitted by macumber on Sun, 02/08/2009 - 3:39am
Gordon, are you testing any water cooling, peltier, or other solutions with Core i7? I am one income tax return away from building a system and debating air vs water vs something else. Looked at CoolIt, Thermaltake Bigwater, and Danger Den DIY ... dunno what to do. Was planning on doing some overclocking of a Core i7 920 chip.
My fried hardware from OCing my X58/COREi7
Submitted by Badger1 on Sat, 02/07/2009 - 4:49pm
Nov. 17 I order my Gigabyte GA-EX58 Extreme mobo, Intel 965EE, a kit of 6 gigs of Patriot Viper EE DDR3 1333mhz and a couple of 150gb VelociRaptors. It all comes in by the 21st, I build it, boot it up and it's unlike anything I have ever used. WOW. I run it for a month and start pushing it. It goes to 3.86mhz with the memory running at 1500mhz like nothing and is still under 40c with air. Everytme I get to 4 ghz I lose stability. I see Patriot says the memory is for 1.65v. I am aware of the dreaded 1.65v + on this CPU but I take it to 1.64 and get 4 ghz stable. It goes into standby mode and locks up. I try different and it seems good until it sleeps for 4 hours and blows my HIS 4870 Ice-Q video card. I put an old X1600 I had it and I'm back up and smoking. I think maybe a bad video card. After another 4 hour slep it blows the X1600. It boots up with a real bad looking screen. I next day a Sapphire 4870, install it and it boos up once, says my BIOS is corrupt and it's restoring. That was the last of that system. Gigabyte said power supply. I tried 3 different ones, new memory, a fourth video card, another mobo and another CPU. I went and got an ASUS P6T Deluxe/PalmOC and i booted right up with my old CPU. Being w/o my system for 16 days, the shipping and return charges I paid made this a sad story. My ASUS was flying so I asked their tech support about sleep mode with my memory set at 1.64v. They said there should be no problem. I put it to sleep for 5 minutes and it mimicked the first system and scared me stiff! I have contacted them 4 times and can't get anything of use from them.
I am now running my memory at 1.5v, CPU on auto 1.325v, BLCK at 150 which gives m3 1501mhz on te memory an sets the CPU at 3.75 ghz. It is going in/out of sleep perfectly and feels like it's running the best I've had.
Thanks AMD for another great processor :)
Submitted by ghot on Fri, 02/06/2009 - 9:28pm
You all know the old story about two disparate people from differnet walks of life ...siwthing jobs or lives for a day etc? I'd really like to see what AMD could do with Intels fabs, for a day, and visa versa. I still can't figure out why Intel just didn't name the processor K11 lol. The "I" in I7 makes it almost sounds like Intel came up with the architecture. Yes, that WAS sarcasm.
Take efficiency, and edit out all the intelligence and what you have left is a post-XP Microsoft operating system :)
Are you mental?
Submitted by Silentblaze513 on Wed, 02/11/2009 - 3:50am
This is about the Intel i7, not the "budget" amd. The only thing AMD has to stand on is its so called "6GHz" headroom; of course you won't be hitting anything near that unless you have liquid-nitrogen and liquid-Helium. Even if you did hit 6GHz on an amd, an intel at 4 to 4.5GHz (on fucking water-cooling) would kick its ass.
Another Rant from and AMD fanboy
Submitted by pppriot on Sat, 02/07/2009 - 7:19am
LOL. i wonder why AMD can't make something better than the i7 and actually market it. rookie.
Whoa - no sure we need all this...
Submitted by canbbb on Thu, 02/05/2009 - 8:38am
Sorry MPC, but I followed HardOCP's method this time for overclocking my 920. And it's MUCH simpler:
[1] I simply overclocked my BCLK to 180, for a 3.6 Ghz machine; and
[2] I selected 1523 Mhz for the RAM (I know, weird choice)
Everything else I left at "AUTO". That's it ! Used Prime95 for 2 hours. Stable as a rock. Did not even need to change the voltages.
I'm not one to shy away from the complexities if that would be the only way to overclock the beast, but it's simply not necessary in this case, so why do it so complicated ??
The Rig:
Core i7 920, overclocked to 3.6 Ghz
6Gb Corsair 1600 Mhz DDR3
Swiftech H20 water cooling kit (I know, I could push my CPU even further with this)
Asus P6T Deluxe mobo
and lots of other stuff :-)
Oh good grief!
Submitted by Cache on Thu, 02/05/2009 - 9:41am
Ugh--that is sloppy and utterly irresponsible. First of all, 'auto' almost always means extra volts--unnecessary volts--and that means heat that was generated for no reason. It's potentially dangerous if you don't have the faintest clue how much power is being used by the components of your system.
I won't disagree that it may be 'easier', but it's far from a good, reliable solution. As a 'first pass' if you want to see if something is possible, maybe--but if you really want to earn any OC chops you have to know how to actually tweak things so you get the best results for a given chip/motherboard/RAM combo. I can look at my Q6600 and recite by memory what all the settings are and I know--I KNOW--that everything is in place as good as I can get with the settings I've set.
Using the 'auto' means that all you did was take a guess and hope for the best, without any real idea what you could have done with your system.
Okay...
Submitted by canbbb on Thu, 02/05/2009 - 1:16pm
Cache is right. :-)
Really, understanding the components, how it all works, and making sure you're doing the right thing is all important. I have to admit my first post sounded like anyone could do just this.
In my defense, the P6T does a very good job of balancing things out - better than other mobos I've had before. And it's only after reviewing at length the videos on HardOCP's site that I went that route for my last overclock. 'Sloppy' and 'irresponsible' are certainly not the right terms to be used here (but go ahead... flame away if you feel like it).
For the record, my voltages currently are: CPU: 1.275, RAM: 1.5 and QPI: 1.2125... well below what this article proposes, and giving me a very stable 3.6 Ghz machine. Yep, the P6T does deserve that good overclocking reputation it has...
Also, I want to apologize
Submitted by Cache on Thu, 02/05/2009 - 6:31pm
Also, I want to apologize for sounding snappish earlier--it was not my intent. That said, I'm running an Asus p5q pro and have had other Asus boards in the past--they really do make OC'ing fairly effortless for people who may not know everything they 'should' do. As it stands now, I've undervolted my Q6600 because I'm really not going too far with it, and I see no reason to throw extra electrons around in there.
And yeah, I left a couple things on Auto when I got around to balancing everything out, it made it very easy to work on one thing at a time and know that if the OC froze up exactly which voltage did it. I think Auto has it's place--it's just not meant to be used as a crutch, if that makes any sense. Also--nice work there!
"Cache is right." Could
Submitted by Cache on Thu, 02/05/2009 - 4:00pm
"Cache is right."
Could you relay that to my wife? She's under the mistaken impression that I'm rarely right... ;)
Great analysis of the BIOS issues....
Submitted by Marcus_Soperus on Thu, 02/05/2009 - 7:46am
...this should be required reading for any overclockers and bench techs dealing with the Core i7 procs. The BIOS changes are the biggest I've ever seen. Thanks for posting this!
--------------------------------------------------------------------------------------------------------------------
It's amazing how illogical a business built on binary logic can be.
Feature
Review
Feature
Feature
Feature









