When overclocking doesn’t work out

This blog entry I write because I feel it’s necessary to raise a point about overclocking processors and GPUs. This is hot off another late night attempt to coax just a little more love from my Core 2 Quad Q6600, which ended in tears, misery, and reinforcement of the status quo. I’m not overclocking on an ECS board either, here, so we’re clear — I’m using an X38-based Gigabyte motherboard. Simply put, my G0-stepping Q6600 isn’t happy going over 3GHz. I shoot for round numbers and at one point was able to get it up to 3.2GHz, but now we hang out in the magical land of 3GHz, using a multiplier of nine and a 1,333MHz front-side bus. This seems to be the best compromise, since I don’t have to ramp up my CPU fan just to keep temperatures down and thus defeat the purpose of having a silent machine.

My Visiontek Radeon HD 4870 512MB fares no better. It’s rock stable at stock speeds, but a journey into overclocking land is always fraught with disappointment. The GPU can’t handle any more than the stock core speed, and overclocking the memory on the RV770 chip is, in my experience at least, somewhat pointless without a good core clock to match. When I tested my 4870 at a core clock of 800MHz, it felt measurably smoother than at stock speed, which was great…if I only intended to game for maybe five or ten minutes. And this is under Unreal Tournament 3; imagine if I’d thrown Crysis at it.

I bring these things to your attention because if you Google overclocks on either of these chips, you’ll see far better results than mine. People on forums will talk about getting their G0-stepping Q6600’s all the way up to 3.6GHz and better, at 1.25 vcore or lower, and it’s Prime95 stable, we swear! The fact is that overclocking is a massive crapshoot. It’s the kind of thing that we warn you about in our CPU reviews, in every overclocking section, and other sites will warn you about, as well. Overclocking is incredibly unpredictable. When the Phenom II was being promoted, AMD was shouting about how the chip could probably hit 4GHz, but our samples only hit the neighborhood of 3.5GHz—still a good overclock, but not the kind of angels-singing-from-the-rafters overclock for which one might hope.

If it seems like I’m beating a dead horse, it’s only in an attempt to pound into steel what some consumers still don’t fully understand. You read about these crazy overclocks, or even a legit review will get a pretty good one, so you’ll go and buy the processor.  Lo and behold, you don’t get anywhere near what you heard about. That’s the nature of the beast, so you’d better buy a processor that’s going to be good enough for you at or close to its stock clock speed.

Likewise, overclocking isn’t an exact science. We have an excellent guide on overclocking here, written by Geoff, that can at least get you started, but there’s a lot of tweaking that you may have to do, and you also have to decide just how you define "stable." For me, for example, "stable" means rock stable. I do heavy high-definition video editing on my desktop, which has been very heavily designed and tuned specifically for that task. An overclock with "good enough" stability isn’t going to be good enough for me when Adobe Media Encoder slams all four cores and redlines them for hours on end. Your standards may be a bit looser than mine, but an overclock could easily involve tweaking timings and all kinds of esoteric settings in the BIOS to reach the kind of crazy heights some of the chips on the web hit.

Overclocking video cards becomes even less exact. The "old standby" for video card stability testing these days seems to be Furmark, with ATITool running a close second, but I’ve had ATITool green-light overclocks that have been tested for over an hour and promptly cause crashes when the card enters an actual game. The overclock testers in software like ATI’s Catalyst Control Center and Nvidia’s System Tools, formerly known as nTune, have seemed even less reliable. Catalyst Control Center is all too happy to let me overclock to 800MHz on the core, "stable," and watch the card choke in actual gameplay. Nvidia’s System Tools had the same issue in my laptop (Asus X83Vm-X2), where my immensely overclockable GeForce 9600M GS nonetheless had higher-clocked confirmed stability tests than bore themselves out in Left 4 Dead. For what it’s worth, though, the 9600M GS does purr along quite happily at near-9700M GT speeds.

I’m not sure exactly how this is going to go over, how successful the warning will be. We’re certainly past the era where the Barton-core AMD Athlon XP 2500+’s quick tweak to 3200+ speeds was considered a major overclock, especially if I’m disappointed in only scoring a 25% overclock off of my Q6600. The vast majority of chips on the market these days do have some pretty impressive overclocking headroom in their designs, but these overclocks aren’t guaranteed. They can be facilitated, as AMD does with its Fusion Overdrive utility, but they can’t be guaranteed by any stretch of the imagination. You may wind up being the unlucky sucker with the Phenom II X4 that only goes up to 3.2GHz or the Core i7 that only hits 3.4GHz at most. I do say "unlucky" with tongue planted firmly in cheek, given how crazy some overclocks can be these days, but we’ve also gotten to the point where the increased headroom has brought increased expectations with it.

All the forum posts in the world aren’t going to change the fact that you’re gambling when you bet on overclocking your next processor. The ATI Radeon HD 4890 may have been designed to hit 1GHz, and some partners are even releasing cards running at 1GHz from the factory, but that doesn’t mean you can buy a cheaper one and just magically expect to hit 1GHz.

Buyer beware: you may get exactly what you paid for.

Comments closed
    • potatochobit
    • 10 years ago

    I am running my X3 720 at 3.2ghz why I have to be called unlucky :0
    to me overclocking is a ‘bonus’ it would be nice if it was uniform though
    but the trend where gamers expect an overclock is kind of asking too much
    squeezing out just a little more with no effort and no rigorous testing is what I like and I think the current chips are a great value for this

    I do not think people should expect to overclock their GPU cards unless companies start releasing 4850 ‘black’ editions
    sure it is nice to try, but its like buying a processor with a locked multiplier, there is no guarantee it will overclock well and you should not expect it even though there is software out there to make the adjustments user friendly

    • sotti
    • 10 years ago

    I’ve been o/c’n for quite a while

    P1 200mmx @ 250 (25%)
    Celeron 300a @ 450 (50%)
    Dual Celeron 366 @ 540 (47%)
    Duron 700 @ 1ghz (42%)
    Athlon 1ghz @ 1.4ghz (40%)
    AthlonXP 2100+(1.73)@2.3ghz (32%)
    Opteron 165 (1.8ghz)@2.5ghz (38%)

    • Kougar
    • 10 years ago

    I completely agree with the post, good to see the subject mentioned again. I’ve never had great sucess overclocking graphics cards, I presume because manufacturers like EVGA are always binning their chips religiously in order to support having 4+ SKU bins for every single individual model. FurMark and OCCT’s new test do seem to be the best tools.

    I recently bought a D0 Core i7 920, XtremeSystems forums were ablaze with seemingly magical overclocks anywhere from 4.2-4.8GHz, even the occasional 5GHz CPUz screenshot. Low and behold my D0 920 is only 100% stable up to 4.2GHz, a few hertz beyond that is akin to falling off a cliff into a deep bluescreen sea of death. Thankfully I would of been happy with just a “mere” 4GHz overclock, but from the comments over at XS most of them would’ve tossed the CPU and bought another. I’m quite happy with my 4.2GHz overclock and find the temptation to try another chip easy to ignore.

      • MadManOriginal
      • 10 years ago

      It’s called xtremesytems for a reason :p I rarely visit that forum and don’t go by what people say there. Too many of the posted overclocks are simple Windows booting or SuperPi :rolleyes: and not approaching 24/7 stability, although with an i7 cooling does play a big role once you get in to the 4GHz+ range.

    • Chrispy_
    • 10 years ago

    Celeron 300A was my overclocking champion, but even though it would post at 600MHz, I wasn’t able to get applications stable above 504MHz

    I’ve overclocked everything since and there was a long era of disappointing overclocks that ended when my 2.13GHz was able to run on a 400MHz FSB instead of the original 266MHz

    Even the supposedly excellent Athlon XP Barton 2500+ only clocked from 1.83 to 2.4GHz, and that was considered a decent overclock for air-cooling. (3GHz required phase-change)

    • Philldoe
    • 10 years ago

    No kidding. Overclocking is defentatly hit and miss. I’ve had CPU’s get some mad over clocks.

    S754 A64 2800+ 1.8Ghz hitting 2.8 on air cooling
    S754 A64 3700+ 2.4Ghz hitting 3.0 on air cooling
    LGA775 C2Q Q6600 hit 3.2 eazy as cake on air
    And after my introduction to lightning…
    My Current E8400 3.0Ghz hits 4Ghz on air like it was meant to do it by default

    But I’ve also had some CPU’s that would not budge, like..
    AthlonXP 2000+ that would only go 100Mhz over
    S754 A64 3700+ that would only go to 2.6 (differant one than the one stated above)

    GPU’s have been less kind to me. my old GeforeceFX 5200 went some ungodly amount over that I can’t remember and my old 9600XT went a good 200Mhz over stock.
    But all of my other GPU’s have HATED any amount of overclocking.

    I gave up overclocking my GPU’s with my last 3 cards simply because they were fast enough for anything I threw at them.

    I can say I’m damn lucky with the fact I’ve never killing a piece of ware with overclocking, but that’s probably because I’m /very/ reuctant to jump voltages on anything. My hardware nowadays is jsut too expensive to be risking for only a few FPS differance in game I’m already getting great speed out of.

    • cocobongo_tm
    • 10 years ago

    Hi guys! Back in 1996, I managed to overclock AMD’s 5×86 chip from 133Mhz to 166Mhz!!!! It was sweeping the floor with a Pentium 100Mhz. But it had stability issues.

    You guys remember MicroProse’s Grand Prix 2 game? (Simultator of Formula one cars and circuits). Awesome game. I ran it on the overclocked chip (160Mhz 5×86) and after 5 mins, all cars laterally banked upward in the air at 45 degrees! That was the cpu starting to give up. All cars ran on their left front and rear wheels, while having the right side wheels up in the air. This went on for about 1 min more and then – reboot.

    Aaaaaaa….glory days! ๐Ÿ™‚

      • _Sigma
      • 10 years ago

      Such a good game, I haven’t played that for years!

    • MadManOriginal
    • 10 years ago

    Let’s look at this post and oc’ing from the perspective of what he got for some time spent tweaking even if it isn’t an especially high oc. This could be a reply to ‘budget overclocking’ ‘free performance’ ‘just buy higher components’ or other posts.

    At the time of its introdcution in January 2007 the Q6600 was $850, the only other quad core CPU was the QX6700 at $1000 (these are Intel 1k unit tray prices so retail box street prices were higher but whatever.) When G0 stepping came out in April 2007 the price was dropped to $266 and the $1200 QX6800 was introduced along with the $530 Q6700 – certainly an excellent time to buy a q6600. Later the QX6850 was the highest speed Kentsfield CPU at 3GHz and a $1000 price, replacing the QX6800. It was a year until the 45nm Yorkfields were released. So what did he get for his oc, even if it was modest? He got the value of a $1k CPU for $250-300 which clearly destroys the ‘buy higher components’ argument. Paying significantly more for small speed bumps is silly, it’s very rare that a CPU can’t go up at least one or two speed grades.

    Now if you were to buy a $30-40-50 aftermarket cooler you could have a CPU that’s ‘not available’ because it’s faster than anything that can be bought and the additional cost would still be less than going up multiple stock speed grades. I’d have to argue that the conclusion of this article is off. It’s certainly a gamble to think you can match i[

    • Habu
    • 10 years ago

    Very timely article. Read this OC report over at PCstats yesterday on E8500 OC’ing. And they “easily” got it to 4.2GHz. Had mine running at 3.8GHz from start and got it to run stable at 4.1GHz (Intel burn of course. Orthos is fine but torture, that is Intel) but didn’t like the temps for 24/7. Not sure I’d say it was an easy one though.

    Anyways after reading it, I figured I’d go back and OC my one a little more. Just one suggestion for anyone OCing an NVDA MB with nTune. DON’T! Your BIOS settings getting raped once you hit Windows. Only way I could get my system back to old was a safe boot, and disable ntune with Autoruns.

    Well once I figured that was the issue (took a while, first 10 trials I figured it was just a bad OC till I noticed it died just after getting into Win ever time). I don’t use ntune actively but it does have a good fan control for us likers of silent PCs which I had forgotten about.

    So I figured since it is there, lets try OC with it. Well after fairly quickly reaching my old 4.1GHz I had to change mem timings and reboot to get them active. YUPP AWESOME SYSTEM FAIL! Never actually had this happen before with an nVidia MB, so first time running a clear CMOS and unistalling my four sticks to get it back to operate. And then complete reconfig of BIOS. Very funny… For a few secs I thought my MB or mem sticks got killed… Still don’t know what it did to my settings, was no crazy speeds for sticks or anything. Just CL4 to CL5.

    Happy camper with my 3.8 for now… and don’t trust nVidia software. If their HW is crummy and prone to die (I am a big fan of their mem controllers, not. Everyday it has your four sticks running you pray for another day), their software is like baby poop coming at you while changing diper!

    • alex666
    • 10 years ago

    Overclock efforts over the years:

    k6III-450+ – excellent, up to 600MHz
    t-bred b 2100 – excellent, up to 2.2GHz
    first mobile barton 2600: excellent, up to 2.6GHz
    second mobile barton 2600, mediocre, only 2.2GHz
    venice 3200, mediocre, can’t remember specifics
    venice 3500, mediocre, up to 2.5GHz stable
    x2 4200, so-so, up to 2.6 stable
    c2d e6600,not bad, up to 3.3 now at 3.2GHz and runs cool
    c2d e6750, awesome, stable and cool at 3.6 GHz

    all in all, a mixed bag, a crap shoot. My experience is that I know pretty quickly how well a cpu will oc, and I always look for that sweet spot of performance X power X temperature and never try to push beyond that.

    • uknowit90
    • 10 years ago

    Just as a question, what PSU are you sporting? I had a BFG 650 watt and my 3850’s (mildly overclocked) would crash every time I opened a game. However, upon replacing the BFG with a much quieter Corsair 750tx, I’ve had 0 problems with the cards crashing, even at a higher overclock.

    Maybe your PSU isn’t supplying clean enough power…

    • PerfectCr
    • 10 years ago

    I never felt the need to overclock a CPU, ever. I’ve done my fair share of GPU overclocking. When I build my next PC I won’t even bother doing that. ๐Ÿ™‚

      • heinsj24
      • 10 years ago

      To be really successful at overclocking you need to have quality components.

      One side of me just doesn’t see financial sense in spending top-dollar for high-end coolers, power supplies, and memory – then spending bottom dollar on a cpu to overclock to a tier where it may be beneficial to purchase a higher specced processor. But then…

      It is so much fun to see what your rig can do. I don’t run everyday with any sort of OC, but I will OC to see what combination of parts can do.

    • puppetworx
    • 10 years ago

    Unlucky man, the Q6600 is well heralded for it’s overclocking ability and frankly that’s a good and bad thing. The smart thing is to buy a chip with high rates of overclock and a good reputation, like the Q6600, the trouble comes when your individual chip doesn’t match with the limits of what others can achieve with their individual but equally named chips.

    I’m running my Q6600 G0 on an Asus P5K-Pro (P35) and it has a Thermalright IFX-14 strapped to the top of it. I have two BIOS profiles one overclocked and one standard and I switch between them when I need.
    The overclock profile has the Q6600 at 3.6GHz and 1.45v. The standard profile has the chip at around 1.28v.
    I’ve found that for gaming (mostly L4D) 2.4GHz is just fine, so I only switch to the overclocked mode for video conversion. I actually bought this setup 1 year ago with the aim of keeping it for 3-4 years and overclocking as needed, so far it’s gone too plan but I am continually tempted to buy new hardware despite the minor real world benefits.

      • MadManOriginal
      • 10 years ago

      Q6600s run HOT with an oc and overvolt and require good cooling. I don’t know what HSF he’s using but if it’s not one of the top tier tower-style coolers his oc doesn’t surprise me. I used the stock cooler on a Q6600 G0 for a bit and could do 3GHz at stock volts as well but with temps getting near the 65C ‘max’ most forums note. A better cooler gave a lot more temp and voltage headroom. Funny that he never mentioned temps or which cooler he’s using…

        • _Sigma
        • 10 years ago

        I have a TRUE with dual Noctua 120s (my case isn’t the most ventilated) and have my Q6600 G0 at 3.2 ghz ~1.45v.

        I do scientific computing on it, so my definition of stable is red-lined cores for days on end. It can pass prime, it can pass IntelBurn, it runs for months on end with no problems. I call that stable.

        puppet, why don’t you use the speedstepping (or whatever it is called) feature? It lowers the multiplier until you hammer the cores and then it’ll bring the multiplier back up to 9x (from 6x) on the fly.

        For me, for what I do, the boost from 2.4ghz to 3.2ghz has been worth it. I have found though that stable RAM is needed for a good OC. I’m running 8gb of RAM, and even though it is 1066, I have to run it at 888 for it to be stable. What I’m trying to get it is that I’ve frequently found an unstable CPU OC to actually be unstable RAM.

        • zqw
        • 10 years ago

        Yup, you need the $50 monster cooler. 3.6ghz here, but I rarely boot into that – just final render of a 3D animation. In many ways, it’s enough or too much CPU already. There’s plenty of ways to work proxy – low res, less lighting detail, etc. And I’ve never thought of 90min video conversion as an on-the-fly activity.

    • eitje
    • 10 years ago

    To be contrary, I’m a big fan of undervolting. ๐Ÿ˜›

      • OneArmedScissor
      • 10 years ago

      Somewhat on that subject, I find it kind of strange that desktop CPUs are always higher voltage than they are in laptops. It’s not as if the vast majority of desktops need that little bump in speed, so they’re pretty much just wasting power.

      You’d think that, to the average person, a note on the ad that says, “Saves money on your electricity bill!” would mean a lot more than “LOLWTFBBQFASTCORE!”

      I also find it curious that desktop CPUs don’t dynamically lower their voltage levels, even though laptop CPUs have been doing that for I don’t know how many years. My friend’s 5 year old Athlon 64 goes from 1.5v to 1.0v at idle, which is most of the time, and I doubt it was the first, though I don’t pay attention to that stuff. Modern desktop CPUs still tend to be around 1.1v to 1.2v, so I’m sure they could stand to shave off a little.

    • Bauxite
    • 10 years ago

    Dirty internet truth #1234:

    A large amount of joe blow overclocking claims are complete BS, if not made up numbers then laughable stability.

    I blame the e-peen nature of the anonymous internet.

    It was nice during “the good old days” when a 50% overclock with stock coolers was considered standard with some famous cpus (and old binning) but now? /care off

    • Krogoth
    • 10 years ago

    Overclocking’s #1 rule has always been “Your millage may vary”.

    In my personal experience, I have successfully overclock a number of components. The most tricky part of lot was a Venice 3200+. It needed a nice dose of voltage in order to obtain any decent stability at 2.4Ghz.

    I had an E6300 that was tricky to overclock. It was mainly because it had to work with a 975X board. The board needed a couple BIOS tweaks to get 333Mhz+ FSB. By comparison, my current EP45-DS4P is a breeze to overclock along with its Q6600 B3. I am certain my Q6600 B3 can handle 3.4Ghz with a few more volts and sufficient cooling. The only problem is that it is already toasty at 3.0Ghz with a tiny voltage boost.

    • continuum
    • 10 years ago

    No surprises. I personally overclocked a half-dozen different Q6600’s all from different steppings and production dates. One B3, five G0’s by the time all was said and done.

    One wouldn’t OC even a whisker. One had to be jacked decently high on Vcore to hit 3.0ghz stable. The other four hit 3.2ghz, but needed varying degrees of voltage applied. Two probably would have been the magical 3.4-3.6-3.8ghz chips that a lot of people had, but I didn’t feel like pushing them that hard once I got 3.2ghz at stock or stock + 0.05625v or something modest– with another 0.1v a lot of chips screamed to 3.6ghz+. I had a few more Q6600’s I never did bother to OC either.

    I am looking forward to Core i5 or Core i7 in my next upgrade to OC one of those, although I’ll probably opt for a more conservative overclock near stock voltages.

    • Hance
    • 10 years ago

    For my personal systems the CPU is almost always overclocked. Video cards I have had terrible luck overclocking so I just leave them alone most of the time. I am currently running an E6400 at 3.2ghz which is a 50% overclock. Its been running that way since day one. I figured if the overclock does shorten the life of my cpu I am not out much anyway. I can’t remember the exact number but I read something that stated intel plans on a 7 year lifespan for its cpu’s. Even if I cut the life span in half which isn’t likely I will have moved onto something newer and better long before the 3.5 years is up.

    For my wife and kids they get hand me down systems that I am done with. Their systems always run at stock speeds to be sure I don’t have to futz around with them.

    • Kent_dieGo
    • 10 years ago

    Prime95 is not a good indicator of stability as it can take many hours to find a problem and often misses marginal problems. Use IntelBurn and you will know if your system is stable in 10 minutes. A small overclock is OK but anything more is asking for trouble. A cool stable system is more important. When something crashes or your hard drive gets corrupted you will be sorry.

      • crazybus
      • 10 years ago

      q[< A small overclock is OK but anything more is asking for trouble. <]q That's completely relative. A chip at the lower end of the the frequency scale for its particular range will often overclock well. q[< A cool stable system is more important. <]q The two aren't mutually exclusive. Proper stability testing is good practice whether or not you overclock.

    • valrandir
    • 10 years ago

    Huge overclocks aren’t new, especially on low-end processors — my 11 year old Celeron 300A overclocked from 300MHz to 450MHz. Back in the 386/486 days, you could overclock many by 100% from 12 to 24MHz or 33 Mhz to 66 MHz (like the old style turbo mode).

    Some low-end Core 2 Duos these days seem to also overclock by 50% or more regularly, eg. 2.4GHz -> 3.6 GHz.

    See this old [H] review on the 300A: ยง[< http://www.hardocp.com/article.html?art=NTU=<]ยง halp me, I feel old for remembering 386s.

      • Palek
      • 10 years ago

      Heh, back in the day I managed to overclock my AMD 486 DX2-80 from 80 MHz to 120 MHz, but eventually settled on 100 MHz. That was still quite a boost. I’m pretty sure all I did was simply change the FSB clock and/or the multiplier. And that was using jumpers, too, not BIOS options. Jumpers are where it’s at! ๐Ÿ™‚

    • Ruiner
    • 10 years ago

    Overclocking is often a hobby, same as woodworking, hotrodding, collecting baseball cards or gardening. Extra performance in certain apps is gravy.

    ‘Back in the day’ you had to choose between OCing and spending a lot to to run games at playable framerates. That’s hardly necessary anymore unless you’re running jumbo monitors.

    • tfp
    • 10 years ago

    My last overclock the biggest issues has been my old northbridge however the new CPU is so fast vs what I had it’s not really a big deal I’m not getting a huge overclock.

    • Kurkotain
    • 10 years ago

    dustin, i noticed in the article you have a x83vm (nice budget gaming laptop BTW…) and i’ve posted before on the forums looking for advice on how to overclock the heck out of the 9600 gs; maybe you can iluminate me on the topic now ๐Ÿ™‚

    • RickyTick
    • 10 years ago

    I always thought that overclocking was for your 2+ yr old computer that seemed a little sluggish on the newer software you just bought. So a slight OC brings it back to where you expect it to be. So at some point (when you’ve reached the limits of the OC) your pc can’t keep up with current software demands, therefore it’s upgrade time, or new pc time.

    However, it seems that a lot of guys are overclocking brand new components just for the fun of it. Well that’s cool and fun, but if you get that cpu that won’t oc at all, the fun is over. My Q6600 will only go to 3.0 just like Dustin’s, so oh well. I’m done overclocking. I think for now on, I’ll just buy the fastest hardware I can afford and just live with it.

    I also get the impression that many of the overclockers are gamers that think it will give them a slight advantage in deathmatch. OC = pwnage, or something like that. ๐Ÿ™‚

      • Meadows
      • 10 years ago

      That means you overclock components before their useful life ends.
      I overclock them as soon as they enter the computer.

      Pretty similar in effect, but I get to enjoy the benefits right from the start.

        • RickyTick
        • 10 years ago

        But don’t you jeopardize the longevity of the components, as well as the warranty?

        • Arag0n
        • 10 years ago

        But you get stressing your computer all this time making the lifetime of the computer shorter. I won’t overclock if it’s not needed. And i won’t use as a buying time option for the system if can’t be massive overclocking.

          • Jigar
          • 10 years ago

          CPU’s normal life may be more than 20 years, you will never use the same CPU that long. Overclocking may bring that down to 5 years and you still may not use it that long. So what would you prefer ?

            • RickyTick
            • 10 years ago

            I guess a cpu might actually “function” for 20 years, but it’s far from useful.

    • marvelous
    • 10 years ago

    Perhaps you are doing it wrong? Testing for stability is key. No overclocking is the same as you said it yourself.

    I usually run my computer 24/7 and not a single hick up whether I run it stock or overclocked.

    • SnowboardingTobi
    • 10 years ago

    I overclock my own computers and never overclock computers I build for others. With all the photo editing I do and some minor video editing, then I find it beneficial to squeeze out some extra speed since it’s relatively free. Great thing is that my systems are very quiet. What I do not do is overclock the GPU since my ‘hardcore’ gaming days are essentially over.

    • TurtlePerson2
    • 10 years ago

    The most overhyped overclock is the unlocking of the fourth core in the X3 chips. I remember reading a forum post where the guy was building his computer with the idea that it was going to be an X4 by using an X3.

    • indeego
    • 10 years ago

    Yeah stopped OC’ing because the first crash due to stability was frankly the last I wanted due to OC. sta-/relia- bility > * to meg{<.<}g

    • Bombadil
    • 10 years ago

    Overclocking has never been about pushing high end products (e.g., 4870, Q6600) faster. Sure some people waste time in such a sport.

    I am still happily overclocking. Nothing substantial has changed about it since the beginning of ICs. My latest chip, a Pentium Dual Core E5200, has been boringly reliable at 3.75GHz on a $55 motherboard.

    Now most “factory” overclocked products like my MSI Geforce 9600GT are crap–unstable at “factory” speeds, but at least I actually paid less for this particular crap. My cheaper eVGA 9600GSO overclocks a bunch–probably since it was designed to be more expensive hardware.

    • Hattig
    • 10 years ago

    With multi-core you’re not just overclocking to the limits of the processor, far more often you’ll hit a TDP / cooling barrier first. We’re at a situation where a 6-core 2.6GHz Istanbul runs in ~90W – that’s 15W per core. Kick it up to 2.8GHz and that’s suddenly 20W a core and 120W overall. 3.0GHz and it’s 25W a core (fits in with Athlon II 250 at 65W TDP for two cores) and 150W overall. A voltage boost for 3.2GHz and it’s 35W a core, or 210W overall. Remember power consumption can often go up far more than the clock speed increase would suggest as well, as you get to high speeds.

    Yet a Phenom II core does overclock to around 4GHz with a little bit of a boost to voltage (see Athlon II 250 reviews). It isn’t clock limited by design. But at 4GHz a six core would be wanting to consume around 400W!

    (I’ve not really researched the figures in this post, but I think the point comes across).

    • herothezero
    • 10 years ago

    I don’t really see the point in overclocking anymore, especially for CPUs. I don’t do any video editing that would benefit from a faster CPU, and all my games are pretty GPU limited. You can get really good performance these days at ridiculously low prices, right out of the box.

    It’s just not 1998 anymore.

      • Sargent Duck
      • 10 years ago

      Last summer I bought a E7200 (Core 2 Duo 2.5ghz). I haven’t even bothered to raise it 1mhz, simply because it’s fast enough for everything I do (games and watching tv). I have no need to overclock. Now, my 4850, I’ve overclocked that to help for the graphics.

        • Waco
        • 10 years ago

        You’re likely CPU limited in many many games if you’ve got a 4850. Crank that poor little dual core!

        If you can keep it stable it’s literally free performance given you put the time in to make it stable. That’s why many people get turned off of it…they follow some guide on the internet and when it doesn’t work they just throw their hands up in the air, scream “screw it”, and clock back down to stock. ๐Ÿ˜›

          • derFunkenstein
          • 10 years ago

          Games that are old enough that at high resolutions/settings can’t push teh 4850 don’t need the OC to deliver smooth framerates, and newer games should be pushing the 4850 to higher resolutions and settings. In other words, the video card (at least nowadays) along with your display should be the two limiting factors. I disagree that he needs to OC, with the caveat that Supreme Commander could always use a little more CPU.

            • Meadows
            • 10 years ago

            Even in the worst case, a little more CPU juice (or RAM speed) always helps the graphics driver as well. That’s not so much about instruction throughput but instruction “latency”, even if you don’t fully max out your processor, it’s nice to do the work faster just to be on the safe side.

            • ImSpartacus
            • 10 years ago

            No, with a 2.5GHz CPU on a 4850 he is CPU limited in most cases. If he raised that E7200 to just 3.0GHz he would be fine.

      • Pax-UX
      • 10 years ago

      I edit quite a lot of 1080p video and it makes a big difference for me; from Quad 2.66 to 3.2GHz (that’s almost an extra core) it made a noticeable difference to a system that has 8GB RAM, 150GB & 74GB Raptors.

      • OneArmedScissor
      • 10 years ago

      You guys are right about games not really needing overclocking anymore.

      Even cheap CPUs now have stock 3 GHz-ish speeds, which seems to be the magic number. The Phenom II dual-core is even 3.1 GHz. It was what everyone overclocked previous generation CPUs to, and what the very most expensive ones had at stock not long ago, but finally, we have come to a point where it’s common place on both sides.

      However, games don’t seem to scale well with a further increase in CPU power, but they don’t seem to need it, either.

      For people who actually do CPU intensive things, I’m not sure there will ever be “enough,” as some things will scale a huge amount, and you can always find something to do with it. I do a lot of audio production, and it seems I can never get enough processing power for that. It’s not that it’s so technically demanding or inefficient, but the more I have, the crazier I get playing with it. Eventually, you get used to working with it, and it becomes a basic requirement to have an enormous amount of processing power.

      I was fine with a single-core Athlon 64 not even 2 years ago. At that point, dual-cores started to get cheap, and it was becoming common place for the things I use to support multi-core CPUs fully. So then I got an X2. Then I built a new computer with a Pentium dual-core E2140, which I overclocked to its limit of 3 GHz. I then outgrew that, and replaced it with an E5200 I sometimes had at 3.6 GHz. Not long ago, I reached a point where that was no longer enough, either, and it seemed to be giving me stability issues, as well.

      That’s the problem with any sort of significant overclock when you NEED the CPU power. That means you are actually pushing the CPU, and you are much more susceptible to any stability issues you have introduced. That doesn’t mix well with the fact that you aren’t playing video games, and it’s probably a bit more important than losing your progress on a level, should the computer crash.

      So now I’ve been working with a Q9400, which, despite the fact that what I am doing scales linearly as you add more cores, I still found ways to push too far, almost immediately. Presently, I am “somewhat” content with “overclocking” it to 3.2 GHz. That would have cost me $1,500 for a Core 2 Quad that runs that stock!

      Unfortunately, I can’t really push it further than that, even though I know for a fact it can hit at least 3.6 GHz. Since I’m running the CPU near a full load, sometimes for hours on end, I wouldn’t even attempt high voltages, so I have to shoot more for the range of “free” overclocks. At that voltage level, it seems to be too easy to get errors out of a Linpack/Intel Burn Test that runs a good length, once you cross the highest stock speed in a line of CPUs, so 3.2 GHz it is, as it handles that fine on “auto” voltage.

      And thus, I’m glad that AMD seems to be continually pushing for higher clock speeds on their quad-cores, without jacking the prices. Hopefully, it won’t be long before there are 3.5 GHz quad-core Phenom IIs at the $250 mark. If they can really hit 4 GHz at 32nm, I will be overjoyed.

      Fidgeting with overclocking is no fun when you’ve got work to do, and I would gladly avoid it. The problem is that Intel pretty much doesn’t let you, unless you’re filthy rich.

    • paulWTAMU
    • 10 years ago

    And that’s why I never plan on overclocking. I can and have OC’d CPUs but you don’t buy something inadequate and then hope you can OC it enough to make it decent. That’s asking for a kick in the nuts.

    • ssidbroadcast
    • 10 years ago

    Whoa. That CPU is sending psychic shock waves at me!

      • Vasilyfav
      • 10 years ago

      That is not the CPU you are looking for.

        • DrDillyBar
        • 10 years ago

        That is not the CPU I’m looking for…

      • ludi
      • 10 years ago

      b~b~buuuyyyyyy m~meeeeee…./[

    • DrDillyBar
    • 10 years ago

    I once looked into overclocking with my ancient 975x mobo, and did fine a few reviews where people were getting the FSB up to around 400MHz, but I could never reliably get it there. But at the time my 1066FSB CPU happily ran along at 1333FSB, so I counted my blessings and just left it there. I’d need a new motherboard before I ever considered OCing again.

      • Ryhadar
      • 10 years ago

      Yeah, like Dustin said it’s an absolute crapshoot. Take for example my 680i motherboard from Asus and my E6600. I’ve read that it’s just a crap overclocking board. However, on stock voltage I got my E6600 to 3.0GHz with a x9 multi and a 333MHz FSB. I don’t run it at this setting because I don’t really need it but I know the head room is there.

    • danny e.
    • 10 years ago

    what is amusing to me is reading processor reviews on NewEgg where the dorkus gives a bad review becuase “it barely overclocked”.

      • superjawes
      • 10 years ago

      And the reviewers also give themselves “high” tech expertise, and their ownership is “1 day to 1 week”…

Pin It on Pinterest

Share This