Experiment: Does Intel’s Turbo Boost Trump Overclocking?

I still remember the PC I owned back in 1998. It was based on a Pentium II 233 with Intel’s Deschutes core, dropped into an Asus P2B motherboard. That system was fast, but I was a bored engineering student and wanted to do more with it. I started with aftermarket air cooling. And although I don’t remember how much overclocking headroom I was able to realize, I do remember that it wasn’t enough. At one point, I pried the plastic cartridge away from the slot-mounted processor and started experimenting with Peltier coolers for better cooling performance. When the proverbial smoke cleared, I was running at a stable 400 MHz—as fast as the most expensive model available at the time, for significantly less money.

Of course, the overclocks today are a lot more significant than 166 MHz. But the principle remains the same: take a processor running at its default settings and squeeze additional value out of it by trying to match the performance of higher-end and more expensive models. With a little effort, it’s actually quite easy to get a sub-$300 Core i7-920 beyond the performance levels of a $1,000 Core i7-975 Extreme without obliterating its reliability.

What About “Auto-Overclocking?”

Overclocking, in general, has always been a bit of a sore subject with AMD and Intel, which officially have to discourage the practice with threats of voided warranties should your CPU show signs of manipulation. Publically, however, both vendors try to appear enthusiast-friendly by giving away overclocking software, facilitating aggressive BIOSes, and selling CPUs with unlocked clock multipliers. Despite those off-the-record endorsements, though, power users simply accept that there’s no such thing as a free lunch, and killing a CPU with too much voltage is sometimes just part of the game.

But with the introduction of Turbo Boost technology in Intel’s LGA 1366-based Core i7 and the subsequent debut of an even more aggressive implementation in the LGA 1156-based Core i5 and Core i7 processors, Intel took it upon itself to implement a form of intelligent overclocking based on a handful of different factors: voltage, amperage, temperature, and operating system P-state requests directly related to CPU utilization.

In monitoring each of those parameters, Intel’s onboard power control unit is able to augment performance by increasing clock rate in situations where the processor’s maximum TDP isn’t being reached. By essentially shutting down unused cores, thereby dropping power consumption, more headroom is freed up in single-threaded workloads, a little less when two threads are active, still less with three cores utilized, and so on. Thus, Intel’s “automatic overclocking” exists as an elegant, more granular way to increase performance without taking power consumption over the maximum TDP rating of any given CPU (130W in the case of Intel’s Bloomfields and 95W in the case of the Lynnfields).

Can You Do Better?

The question we asked ourselves—especially after seeing that the Core i7-860 and -870 would accelerate a fantastic 667 MHz in single-threaded apps—was whether it was still worth it for the power user to go all-out with processor overclocking and risk nuking a perfectly good CPU, or simply let Intel’s version of the technology handle business? I hoped that I wasn’t just getting lazy in my old(er) age, and that there’d still be palpable gains to taking the enthusiast’s path to better performance. But I also wasn’t ready to dismiss the efforts Intel’s engineers made in optimizing Nehalem for balanced performance in single- and multi-threaded software.

Turbo Boost Versus Overclocking

We decided to run a little experiment: take a Core i5-750 and Core i7-860, overclock each, and compare the results to what the two processors can do with Turbo Boost enabled, and then disabled. Of course, we have samples from Intel here in the lab, but we can’t really believe that those are representative of retail models. So we bought both chips off of Newegg, just to be sure. While we debated sticking with Intel’s retail cooling solution, at the end of the day it was decided that we’d never seen 4 GHz if we didn’t augment the reference heatsink. Thus, all of our testing is done with Thermalright’s MUX-120, too.

This thread is closed for comments
    Your comment
  • phantomtrooper
    No one needs a Core i7 for gaming. I'm still using a e8400 with a gtx275 and I run everything fine at 1080p, even Crysis.
  • apache_lives
    i play GTAIV online alot - your e8400 gets left in the dust there sorry PhantomTrooper, and theres no adverse effects having spare cores for future use with newer games etc
  • dtemple
    I second that, PhantomTrooper. I'm on a slightly lower end spec PC than you're using (Athlon 7750 with Radeon HD4830) and with it hooked up to my 1366x768 TV through VGA, everything I play maxes without lag. Mind you, I don't play any titles that are extremely demanding, but I'm playing 2008 and 2009 titles maxed out, on a $60 CPU and an $80 video card.
  • curnel_D
    Great article. On another note, the useless Mass Effect 2 ad blaring it's stupid music in my ears at every new page is really starting to piss me off.
  • cangelini
    Yeah, you guys are going to get a kick out of the upcoming Clarksdale story. It's amazing how badly a Core 2 Duo E8500 gets killed by a Phenom II X4 or Core 2 Quad in some of these more optimized titles.

    Curnel--sorry about the ad. I also find it pretty annoying to play automatically every time I open a page for proofing. I'll ask about it.
  • jj463rd
    Nice article.I agree about the annoying video ads that Curnel_D mentioned earlier.
  • HansVonOhain
    Curnel_DGreat article. On another note, the useless Mass Effect 2 ad blaring it's stupid music in my ears at every new page is really starting to piss me off.

    Thanks Chris for your courtesy.
  • descendency
    Curnel_DGreat article. On another note, the useless Mass Effect 2 ad blaring it's stupid music in my ears at every new page is really starting to piss me off.

    I know it's not a viable solution for all, but I never have my sound on... so I didn't even notice the advert (other than seeing it.)
  • gilbertfh
    I have been building computers for myself family and friends for years and I remember some of the different ways utilized to speed up your computer (including the turbo button on your computer). This method seems like a viable way to speed up computers of those of us that don't really want to overclock.

    On a side note: Woot!!! I just saw the Mass Effect 2 ad has been removed. It did have the option to mute it but it was removed fast enough I didn't have a chance to check to ensure it saved to all pages. Thanks Tom's.
  • gilbertfh
    Nope I was wrong I guess it was random and just didn't come up for a while and the mute does not save :(.
  • apache_lives
    AbBlock :)
  • gti88
    Indeed, Adblock all the way. It's a necessity these days.
  • wuzy
    It was quite obviously already the only useful purpose for Turbo Boost is if the chip ran @stock. With it enabled it hinders scaling potential greatly. What I found most useful in this article is it provided more data on which programs benefited from the 'new & improved' Hyper-Threading.
  • cannon_foddermn
    Interesting article. What I don't understand is every time a pc website, magazine article, or review refers to the difference in system cost for value consumers vs. high end or performance consumers.

    There is a considerable price overlap for the low to high range for a 1156 i5/i7 system and the low to high range for a 1366 i7 system.

    The system assembled for the testing of the value oriented consumer system in this article with out OS is $1700+ at Newegg! Of course, as Chris Angelini points out in the article, the non chipset specific components (GPU, PSU, SSD and heat sync) were chosen to remove potential bottlenecks to facilitate a reliable review.

    If you add up the current Newegg cost Intel Core i7-860, Intel DP55KG, and CORSAIR DOMINATOR 4GB DDR31600 (PC3 12800) alone it is $653.97 while the Intel Core i7-920, "Tom’s Hardware 2009 Recommended Buy" ASRock X58 Extreme LGA 1366, and CORSAIR DOMINATOR 6GB DDR3 1600 (PC3 12800) total $683.97. The performance gains, better feature set (triple channel memory/3 x PCI Express 2.0 x16 slots) and future upgrade options (SLI and Crossfire support) of the X58/1336 based system are well worth the extra $30! These prices are not even shopping around for deals on the components, since Microcenter has had the i7 920 for $199.00 since launch and you can save $88.00 and add the saving back into a better GPU or mobo or save the extra money for a real value.

    Don't believe the hype from Intel about the "value" and "enthusiast" product lines. The only difference is Intel's greater profit margin on the stripped down Lynnfield and P55 chips.
  • dfusco
    I would like to see how a minor OC with Turbo enabled would stack up.
  • Gandalf
    I've got the i7-975 quad core on a DFI LP UT X58-T3eH8 motherboard.
    I'm finding the Turbo feature great. If the CPU workload decreases, the clock rate decreases allowing the system to cool down. There is no messing with RAM voltages. Only turning on the feature in the BIOS. Even I could do that.

    Also, this is a designed feature. You're not trying to run the CPU faster than it was designed to run as is often the case with overclockers and there is no trial and error with possibly burning out your CPU. I love the Turbo feature.
  • nicklasd87
    dfuscoI would like to see how a minor OC with Turbo enabled would stack up.

    I agree, I am currently gunning my i7 920 at 3.2 ghz on stock cooling, and with turbo boost enabled, the clock increases to 3.3ghz when the multiplier increases from 20 to 21. Without turbo boost, I am not able to reach a multiplier of 21 because of the limitations on the processor, and correct me if I'm wrong, but disabling turbo boost also disables speed step, so my processor would always run with a multiplier of 20 instead of scaling back to a multiplier of 12 when it is idle.
  • grimjester
    dfuscoI would like to see how a minor OC with Turbo enabled would stack up.

    Same here. It seemed to work well for the $1300 SBM.
  • Could we see an apples-vs.-apples comparison of load power usage as well please? Such as Prime95 running for 10 minutes on all (virtual) cores, for example.

    I've read a fair few reviews on the socket 1156 processors and they have all pointed to significant overclocking/overvolting absolutely wrecking power statistics with a draw some three _times_ the base model. Sometimes more.
  • cyberkuberiah
    this is actually nice , i dont have to overclock my i5 for any game out there ... with powerful single gpu's here , looks like i would be happy without 1366 , sli or CF ... with an easy 3.6 oc when i need it , i think this would last me at least 3 years ... same for any body who has phenom 2 x4 , q9xxx (even q6600) , they're all powerful and can oc well .. the only sad thing's that have happened to gaming this year is radeon's higher than launch prices and delayed fermi .
  • Catalina588
    This article was useful as far as it went. Unfortunately, I think Tom's and the PC testing industry need to look at benchmarks that better simulate real-world users, and most specifically, multi-tasking. The fact is, most of the time users are today running a mix of jobs simultaneously. This article treats applications like they ran under MS-DOS 20 years ago: one at a time. How about throwing a few mixes together and reporting the results? Examples from this article include: iTunes + productivity; gaming + zipping + encoding.

    I find that in a typical day's computing, the Core i7 and Core i5 with modest overclock on (about) stock voltages, hyperthreading and TurboBoost on do great, and use electricity modestly. But I can't measure that good feeling. Tom's can do that.
  • hanafyaa
    I've left my turboboost on and overclocked to 164 bus speed and am just short of 4.2 Ghz when running prime95. I am more than happy with that and see good use of the turboboost when required.
  • juliom
    We are presented with yet another Core i5/7 overclock article... The bias from Tom's is becoming to great to bear. I miss Tom's from the old days...
  • eaclou
    Good article, but I would have liked to see a more in-depth analysis of the power-usage and using that, some efficiency statistics.

    Of course the 4Ghz will be faster, but where does it stand as far as efficiency goes? That's really what you should be comparing here.

    It might be nice to throw in some heat numbers, as well.