Sign in with
Sign up | Sign in

Conclusion

Clarkdale Efficiency: The Perfect Clock Rate For Intel's Core i5-661
By

We don’t want to spend too much time talking about the Clarkdale-based Core i5/i3 dual-core processors themselves. We’ve done that in great detail, and it’s obvious that these CPUs deliver great performance and impressive efficiency, albeit at a significant price premium compared to AMD’s offerings. Instead, this article looks at the performance gains and the impact on power efficiency when overclocking a Core i5-661.

Our testing revealed that the processor shows fairly steady efficiency up to about 3.9 GHz. This means that every voltage and clock speed increase causes an increase in power consumption that is equivalent to the performance impact. However, if you exceed 4 GHz nominal clock speed, you will have to turn up the processor voltage (Vcore), and this clearly decreases performance per watt. Our recommendation is to stay in the range of 3.7 to 3.9 GHz nominal speed. Intel’s Turbo Boost technology will still be there to add another one or two clock speed increments.

Two conclusions can be drawn: first of all, the difference in power consumption between the stock settings and our highest overclock—4.0 GHz nominal and 4.32 GHz with Turbo Boost max'ed out—is relatively small. Our measurements say that the delta is around 5W at idle and roughly 30W at peak load. Even though efficiency decreases above 4 GHz, it might still make sense to go higher. The chips have decent clock speed margins, and the additional power draw isn’t a huge deal. All you’ll need is a better cooling solution. Last but not least, there's room for improvement on the power supply side, as the 750W PC Power and Cooling device we’ve been using for various reviews is by no means an ideal partner for this test system. Go for a more conservative PSU to match power requirements so you can operate within power supply’s highest efficiency range, and thus save some power.

Our second conclusion refers to Turbo Boost: we want more of it! Obviously, there's headroom. Intel could take better advantage of the existing architecture, so why not work on more sophisticated power schemes? Turbo Boost and SpeedStep are conceptually similar, but they originate from different requirements. The PCU, Intel’s on-die power control unit, is already there to take care of dynamic clock speed adjustments. We would love to see processors differentiated mainly by their thermal specifications, meaning that a 65W desktop CPU would feature large power saving potential, but limited (reasonable) “overclocks” to stay within the power envelope. At the same time, a 130W high-end CPU could offer smaller power saving options but much higher clock speed boosts at peak loads, given the available thermal headroom, of course. Finally, the Extreme Edition flagship could allow users to modify thermal thresholds at their own risk. This is only a vision, but it would make more sense than keeping the processor world revolving around clock speeds.

Ask a Category Expert

Create a new thread in the Reviews comments forum about this subject

Example: Notebook, Android, SSD hard drive

Display all 17 comments.
This thread is closed for comments
  • 0 Hide
    JohnnyLucky , February 17, 2010 5:12 AM
    Thanks for an informative overclocking article that helps put things in perspective.
  • 0 Hide
    shubham1401 , February 17, 2010 6:15 AM
    Nice article!!

    These i5 are good overclock-able chips :) 
  • 0 Hide
    racermx187 , February 17, 2010 6:30 AM
    Very nice article and I totally agree with the conclusion paragraph and changing the way CPUs are looked at for use.
  • 0 Hide
    anamaniac , February 17, 2010 7:43 AM
    Cool.

    But after all that, I'm just wondering what kind of efficiency I could get out of a high binned dual core that's left on stock clocks but heavily undervolted with 2x2GB 1.5v DDR3, a 80GB x18-m and a 5670 and a board/PSU designed for low powert usage. Hell, my current CPU would use as much power as that entire system. =)
    But, we didn't buy i7's for the performance-to-watt ratio now, did we?
  • 0 Hide
    JeanLuc , February 17, 2010 8:06 AM
    Patrick Schmid and Achim Roos - Can I ask why are your articles are always co-authored?

    Thank you.
  • 1 Hide
    coldmast , February 17, 2010 1:51 PM
    JeanLucPatrick Schmid and Achim Roos - Can I ask why are your articles are always co-authored? Thank you.

    The co-authoring probably has to do with one being the writer of the original article and the other being the translator, or one is the SME and the other is the technical writer.
  • 0 Hide
    envolva , February 17, 2010 3:07 PM
    We could use a i5 750 at stock speed to see how it compare with an overclocked i5 661. Maybe throw an i3 in there too to keep things in perspective.

    It would be nice to see Turbo Boost disabled (power savings still up) and see how the overal efficiency behaves. Whitout Turbo Boost, you can keep the multiplier under control maybe reducing the core voltage needed.
  • 0 Hide
    burnley14 , February 17, 2010 4:24 PM
    Pretty interesting that there was not much of an efficiency increase even with small overclocks. Maybe this is because of Turbo Boost? If so, I think envolva's got the right idea.
  • 0 Hide
    Anonymous , February 17, 2010 5:54 PM
    My question is what role the 45 nano graphics core plays in this?

    Might have been more interesting to use a i5-670 part, where the base core speed is 133 mhz higher and graphics core starts at 733 mhz versus 900.

    The power/clock speed graph looks a little odd, I wonder how much of it is the cpu core staying low power while the graphics core uses more and more energy?
  • -1 Hide
    ta152h , February 17, 2010 8:10 PM
    I'd be curious if overclocking the graphics part increases pure processor performance at all. The knee-jerk reaction is that this is an idiotic question, but consider the memory controller is in the video controller, and there's a possibility that even if you don't use the IGP, you might be well to overclock it to improve the memory performance slightly.
  • -1 Hide
    sandypants , February 17, 2010 8:41 PM
    What were the temperatures like?
  • -1 Hide
    Anonymous , February 18, 2010 9:27 AM
    I must say that in my mind your testing is faulty from the start: if you just start increasing base clock with stock voltages, the outcome is obvious even without testing (that best efficiency is around the point you can go without rising voltages). If you want to get us some real and even remotely useful data on microarchitecture/manufacturing process efficiency you must UNDERVOLT every test frequency you use down to minimum stable voltage! So any chance we will see that in future?
  • -2 Hide
    fernandogmd , February 18, 2010 2:38 PM
    It would´ve been interesting to see some game benchmarks as well. To see if there´s any benefit to efficiently overclock the cpu for games.
  • 0 Hide
    teknic111 , February 18, 2010 5:48 PM
    I want Gulftown!!!
  • 0 Hide
    Catalina588 , February 19, 2010 12:44 AM
    Is it just me, or did I miss the Voltage Table on the Voltage Table and Settings page? What voltages were used at each BCLK?
  • 0 Hide
    Anonymous , February 20, 2010 9:27 PM
    It's not just you.
  • 0 Hide
    tiktianc , February 22, 2010 2:59 AM
    i think there should be a effeciency article done for an i3, i mean they're really easy to overclock with the one in my mITX system doing 3.8GHz easy on stock cooler