The 5 GHz, Six-Core Project: Core i7-980X Gets Chilly

By Thomas Soderstrom
published

Tom’s Hardware has been ringing the 5 GHz bell for years, yet those efforts have never yielded a practical daily-use solution. Today we take a look at how far off-the-shelf parts can take us with an ambitious 5 GHz, six-core overclocking project.

Power And Efficiency

The Core i7-980X is a fairly effective power miser at stock settings, as our entire system consumed only 104W idle and 202W when running 12 Prime95 threads.

Power consumption begins to look ugly when we go beyond 1.35V core, more than doubling by the time we reach 5 GHz. That can’t be a good sign for an overclock that had a peak performance gain of around 50%.

Average performance gains were far smaller than peak gains, with our fastest configuration pushing a mere 20% over stock.

Intel’s no fool when it comes to picking the clock speeds and voltage levels of its high-end products. Articles as recent as our June System Builder Marathon have shown that efficiency can be improved when an overclocked processor’s performance increases more than its power consumption. But the above charts indicates that our smallest increase in voltage might have been a little bit overzealous for a daily-use machine.

Thomas Soderstrom
Thomas Soderstrom
Thomas Soderstrom is a Senior Staff Editor at Tom's Hardware US. He tests and reviews cases, cooling, memory and motherboards.
87 Comments Comment from the forums
  • Poisoner
    Pretty impressive. I wish Athlon II would over clock to 4ghz on air. :(
    Reply
  • zorky9
    Cooler Master’s reputable CMPSU-850HX

    Did you mean Corsair?
    Reply
  • tacoslave
    i thought it would perform much better in games
    Reply
  • Crashman
    tacoslavei thought it would perform much better in games Check this out:
    http://www.tomshardware.com/reviews/core-i3-gaming,2588.html
    For the CPU to become the choking point, you need the GPU to be extremely powerful. Tom's Hardware formerly used unrealistic tests like Half Life 2 at 640x480 just to prove the CPU performance difference in games, but the fact that nobody used those settings eventually lead to the discontinuation of that testing method.
    Reply
  • noob2222
    tacoslavei thought it would perform much better in gamesThere was a thread several months ago comparing the Intel Vs AMD platforms regarding the graphic card bottleneck. Suprising enough Intel cpus capped at a certain overclock where as AMD did not, eventually becoming faster FPS but required a much higher clock speed. It was determined that Intel has a limit on pcie bandwidth. Good luck finidng it, its probably over 6 months old.

    back to the arcitle, very interesting, and extremely expensive to even consider doing something like this.
    Finally there’s the expense. Our $900 cooler requires around 480W of power in addition to that consumed by the rest of the PC
    Add in the cost and time required to set this type of thing up as well as coating the MB ... lol, I don't even want to think about actually trying to go this extreme.

    Water is good for me, and if I want extreme, I will wait till winter and throw my radiator out the window while its freezing outside and pump antifreeze through it lol.
    Reply
  • liquidsnake718
    zorky9Cooler Master’s reputable CMPSU-850HXDid you mean Corsair?Nope coolmaster has a whole line of PSU's... they are decent and perhaps compareable to Corsair for albeit a slightly lower price point. But Oc'ing to 5.ghz and above is really crazy stuff here. I am actually happy with 3.0 and above already stock but damn, I dont think i would like to tax my system to 5.0 and above, regardless of cooling! It would cost more, but I do see the sport of it and commend those that take time to reach 5.0 and above figures with regular water cooled systems.
    Reply
  • vinehoyle
    Amusing article. However some correcting is needed. The next to last paragraph states the cooler consumed 480W and the cpu consumes according to ur graphic 220w? My math says 700w JUST for the cooler and cpu. This calls in to question ur power bill figures. It's clearly much greater. I won't even talk about once u add things, though it would have been a good idea to calculate that figure too. Simply to show the madness of this project to ur power bill lol!
    Reply
  • Crashman
    vinehoyleAmusing article. However some correcting is needed. The next to last paragraph states the cooler consumed 480W and the cpu consumes according to ur graphic 220w? My math says 700w JUST for the cooler and cpu. This calls in to question ur power bill figures. It's clearly much greater. I won't even talk about once u add things, though it would have been a good idea to calculate that figure too. Simply to show the madness of this project to ur power bill lol!The article specifically states that the "added expense" figures are based on the 480W it takes to run the cooler. It assumes you're already planning to use the rest of the system at whatever speed you can get WITHOUT the cooler, and tells you how much MORE it costs to use the cooler. I think its fairly well explained, but feel free to point out any specific spot I missed, thanks!
    liquidsnake718Nope coolmaster has a whole line of PSU's... He's right, CMPSU-850HX is a Cosair model. Cooler Master does have some decent 850W power supplies though, I have one sitting in my liquid cooling bench station.
    Reply
  • dark bishop
    id like to see them try it on the phenom II x6
    Reply
  • Moshu78
    I'm sorry but once again, Tom's is testing CPU's while bottlenecking the Video. Can't you guys just use a simple rule like: When you test the CPU, use highest available video power (like Xfire of 5970), and when testing the GPU, use highest available CPU (980X). It's THAT simple. The 3D benchmarks are meaningless, waste of time AND money AND information. I predicted those graphs in the moment I saw you used a 5850 for the tests.
    Reply