Homework: How Turbo Mode Works

AMD and Intel both figured out the practical maximum power consumption of a desktop CPU. Intel actually discovered it first, through trial and error, in the Prescott days. At the high end that's around 130W, for the upper mainstream market that's 95W. That's why all high end CPUs ship with 120 - 140W TDPs.

Regardless of whether you have one, two, four, six or eight cores - the entire chip has to fit within that power envelope. A single core 95W chip gets to have a one core eating up all of that power budget. This is where we get very high clock speed single core CPUs from. A 95W dual core processor means that individually the cores have to use less than the single 95W processor, so tradeoffs are made: each core runs at a lower clock speed. A 95W quad core processor requires that each core uses less power than both a single or dual core 95W processor, resulting in more tradeoffs. Each core runs at a lower clock speed than the 95W dual core processor.

The diagram below helps illustrate this:

  Single Core Dual Core Quad Core Hex Core
TDP
Tradeoff

 

The TDP is constant, you can't ramp power indefinitely - you eventually run into cooling and thermal density issues. The variables are core count and clock speed (at least today), if you increase one, you have to decrease the other.

Here's the problem: what happens if you're not using all four cores of the 95W quad core processor? You're only consuming a fraction of the 95W TDP because parts of the chip are idle, but your chip ends up being slower than a 95W dual core processor since its clocked lower. The consumer has to thus choose if they should buy a faster dual core or a slower quad core processor.

A smart processor would realize that its cores aren't frequency limited, just TDP limited. Furthermore, if half the chip is idle then the active cores could theoretically run faster.

That smart processor is Lynnfield.

Intel made a very important announcement when Nehalem launched last year. Everyone focused on cache sizes, performance or memory latency, but the most important part of Nehalem was far more subtle: the Power Gate Transistor.

Transistors are supposed to act as light switches - allowing current to flow when they're on, and stopping the flow when they're off. One side effect of constantly reducing transistor feature size and increasing performance is that current continues to flow even when the transistor is switched off. It's called leakage current, and when you've got a few hundred million transistors that are supposed to be off but are still using current, power efficiency suffers. You can reduce leakage current, but you also impact performance when doing so; the processes with the lowest leakage, can't scale as high in clock speed.

Using some clever materials engineering Intel developed a very low resistance, low leakage, transistor that can effectively drop any circuits behind it to near-zero power consumption; a true off switch. This is the Power Gate Transistor.

On a quad-core Phenom II, if two cores are idle, blocks of transistors are placed in the off-state but they still consume power thanks to leakage current. On any Nehalem processor, if two cores are idle, the Power Gate transistors that feed the cores their supply current are turned off and thus the two cores are almost completely turned off - with extremely low leakage current. This is why nothing can touch Nehalem's idle power:

Since Nehalem can effectively turn off idle cores, it can free up some of that precious TDP we were talking about above. The next step then makes perfect sense. After turning off idle cores, let's boost the speed of active cores until we hit our TDP limit.

On every single Nehalem (Lynnfield included) lies around 1 million transistors (about the complexity of a 486) whose sole task is managing power. It turns cores off, underclocks them and is generally charged with the task of making sure that power usage is kept to a minimum. Lynnfield's PCU (Power Control Unit) is largely the same as what was in Bloomfield. The architecture remains the same, although it has a higher sampling rate for monitoring the state of all of the cores and demands on them.

The PCU is responsible for turbo mode.

New Heatsinks and Motherboards Lynnfield's Turbo Mode: Up to 17% More Performance
POST A COMMENT

341 Comments

View All Comments

  • Supershanks - Thursday, September 10, 2009 - link

    Great Article Anand, I read it with great interest.
    However I found somthing that was strange and unexpected today
    My I7-860 was running at 30x133 3990 when running nucleus, that's according to cPU-z I have a link , but can't post http in this comment ?
    http://www.clunk.org.uk/forums/reviews/24295-asus-...">http://www.clunk.org.uk/forums/reviews/...-deluxe-...

    I'd appreciate your insight ?
    thanks
    Reply
  • Gary Key - Thursday, September 10, 2009 - link

    You need to change your version of CPU-Z. I have a screenshot somewhere around here of 48x133. ;) Reply
  • Scali - Thursday, September 10, 2009 - link

    On a slightly related note... With these new P55 boards I see that especially Asus uses a new line of onboard audio chips from VIA.
    I haven't really been able to find much info on these chips. I'd like to know how they compare to Realtek and other onboard offerings.
    Could you guys spend some time on reviewing the onboard audio next time you review one of these boards? Eg, what does the control panel for these chips look like, what features does it have (eg, can you have realtime encoding like DTS connect or DD Live?), what is the general driver quality like (proper support of 3d/eax effects etc)?

    I think that's what's been missing in general, the past few years. Onboard audio has gotten quite advanced, to the point where most people no longer use a separate soundcard (some boards actually come with some sort of X-Fi card). However, I rarely see onboard audio reviewed, only audio cards.
    Reply
  • agawtrip - Thursday, September 10, 2009 - link

    for me, i find this review is misleading if you are not a gamer - for i5-750 and PII x4 965.

    why?

    1. motherboard and video card - non-gamer dont buy sli/xfire board. onboard graphics is fine(780g/785g). for now, boards for i5 setup doesn't have onboard graphics. what will you do? you will be forced to buy a video card (maybe 4550/9400gt for $40).

    i5-750 - $195, GIGABYTE GA-P55M-UD4 - $150, nvidi 9400gt - $40
    --- TOTAL -------- $385

    PII x4 965 - $245, GIGABYTE GA-MA785GMT-UD2H - $90, no video card
    --- TOTAL -------- $335

    AMD setup is actually cheaper but slower. it's all up to you.

    2. power consumption - since you are forced to buy video card, it will consume additional power while AMD setup (780/785G) won't.

    well that's just my opinion.
    please inform me and the others if i given up wrong informatin
    Reply
  • chrnochime - Thursday, September 10, 2009 - link

    Or you can get a motherboard that's quite a bit cheaper:


    MSI P55-CD53

    http://www.newegg.com/Product/Product.aspx?Item=N8...">http://www.newegg.com/Product/Product.aspx?Item=N8...

    which is 120..

    or

    GIGABYTE GA-P55M-UD2

    http://www.newegg.com/Product/Product.aspx?Item=N8...">http://www.newegg.com/Product/Product.aspx?Item=N8...

    And yes I know these are NE prices, and that it might be more expensive/cheaper somewhere else
    Reply
  • Hrel - Wednesday, September 09, 2009 - link

    It DOES NOT have hyperthreading. Enabling it wouldn't increase the cost of the chip, it purely a political move. I HATE that! If a CPU maker has something that can increase the performance of my cpu, at no cost to them, then should enable it, at no cost to me. EVERY CPU should have unlocked multipliers. EVERY CPU should have hyperthreading. Reply
  • jnr0077 - Friday, July 27, 2012 - link

    hyperthreading steal ram from your pc i have a i5 750 + radeon hd 4850 it plays crysis 2 maxed the witcher 2 maxed skyrim maxed i think you need to read up on it before you comment :) Reply
  • PhilTaylor - Wednesday, September 09, 2009 - link

    FSX is well known as a CPU-limited game, it might be more interesting to test it then a GPU limited game like Crysis. For instance, the difference between 2 and 3 memory channels might have a greater impact on FSX, again due to its CPU-limited nature. Reply
  • cactusdog - Wednesday, September 09, 2009 - link

    A very good, detailed bunch of tests but there is a surprising lack of information regarding temps? Other review sites have done the same thing but there IS temp issues with these CPUs as with i7 1366 ones. That CPU at 4.0 Ghz on air will be around 90 degrees but little is said.

    Anandtech is more honest than most review sites (most of which are really just advertisements) but sometimes i get the impression that nobody wants to upset Intel.

    Reply
  • araczynski - Wednesday, September 09, 2009 - link

    ...bottom line to me, my E8500/crossfire setup still has plenty of gaming life left. I'll check back in a year. Reply

Log in

Don't have an account? Sign up now