Power Improvements

Although Haswell’s platform power is expected to drop considerably in mobile, particularly with Haswell U and Y SKUs (Ultrabooks and ultrathins/tablets), there are benefits to desktop Haswell parts as well.

There’s more fine grained power gating, lower chipset power and the CPU cores can transition between power states about 25% quicker than in Ivy Bridge - allowing the power control unit to be more aggressive in selecting lower power modes. We’ve also seen considerable improvements on lowering platform power consumption at the motherboard level as well. Using ASUS’ Z77 Deluxe and Z87 Deluxe motherboards for the Haswell, Ivy and Sandy Bridge CPUs, I measured significant improvements in idle power consumption:

Idle Power

These savings are beyond what I’d expect from Haswell alone. Intel isn’t the only one looking to make things as best as can be in the absence of any low hanging fruit. The motherboard makers are aggressively polishing their designs in order to grow their marketshare in a very difficult environment.

Under load, there’s no escaping the fact that Haswell can burn more power in pursuit of higher performance:

Load Power - x264 HD 5.0.1 Benchmark

Here I’m showing an 11.8% increase in power consumption, and in this particular test the Core i7-4770K is 13% faster than the i7-3770K. Power consumption goes up, but so does performance per watt.

The other big part of the Haswell power story is what Intel is calling FIVR: Haswell’s Fully Integrated Voltage Regulator. Through a combination of on-die and on-package circuitry (mostly inductors on-package), Haswell assumes responsibility of distributing voltages to individual blocks and controllers (e.g. PCIe controller, memory controller, processor graphics, etc...). With FIVR, it’s easy to implement tons of voltage rails - which is why Intel doubled the number of internal voltage rails. With more independent voltage rails, there’s more fine grained control over the power delivered to various blocks of Haswell.

Thanks to a relatively high input voltage (on the order of 1.8V), it’s possible to generate quite a bit of current on-package and efficiently distribute power to all areas of the chip. Voltage ramps are 5 - 10x quicker with FIVR than with a traditional on-board voltage regulator implementation.

In order to ensure broad compatibility with memory types, there’s a second input voltage for DRAM as well.

FIVR also comes with a reduction in board area and component cost. I don’t suppose this is going to be a huge deal for desktops (admittedly the space and cost savings are basically non-existent), but it’ll mean a lot for mobile.

No S0ix for Desktop

You’ll notice that I didn’t mention any of the aggressive platform power optimizations in my sections on Haswell power management, that’s because they pretty much don’t apply here. The new active idle (S0ix) states are not supported by any of the desktop SKUs. It’s only the forthcoming Y and U series parts that support S0ix.

Introduction Memory, Platform & Overclocking
POST A COMMENT

209 Comments

View All Comments

  • Nacho - Saturday, June 01, 2013 - link

    Maybe it's time to upgrade my C2D E4300? :P Reply
  • krumme - Saturday, June 01, 2013 - link

    Absolutely, go get a good ssd and this processor or IB, if its for desktop. Reply
  • Oscarcharliezulu - Monday, June 03, 2013 - link

    Do it. Reply
  • Boissez - Saturday, June 01, 2013 - link

    So my 2½ year old 2600K ($317) performs about the same as todays 4560K ($242). Color me underwhelmed.

    Meanwhile in mobileland we've went from 1 Ghz Tegra 2 to 2 Ghz Snapdragon S600 within the same timespan.
    Reply
  • tential - Saturday, June 01, 2013 - link

    Because for 99.9% of the population what's out there today is more than fast enough. Hell, the Core2Duo Conroe/Penryn processors are fast enough for most people today. I'm still using one in fact.

    On the mobile side however, we have tons of applications that could use more power. My galaxy S3 takes a little to load up some games, and while the data may have been downloaded to the phone through wifi, it still isn't on my screen yet.

    I think it's pretty obvious why you see mobile land having to progress so fast while desktop processors are focusing on power consumption as the AVERAGE consumer (not people who are techies) would prefer smaller PCs and pushing more power efficient processors into smaller and smaller things like the intel NUC is what the consumer desires.

    In short:
    Your desires are the 1%. The 99% are being catered to.
    Reply
  • klmccaughey - Monday, June 03, 2013 - link

    @tential: Your last statement, maybe it should read "OUR desires are the 1%"? ;) I bet we would all be clapping right now if the 4770k was a big upgrade. Well, most of us, I think? Reply
  • jeffkibuule - Saturday, June 01, 2013 - link

    Mobile is having the same performance renaissance that desktop chips had from 2004-2006 when we went from a hot, bloated Pentium 4 to a cool, efficient Core 2 Duo. And certainly we've had performance gains since then, but eventually the gains won't come so easily. You can start to see that a bit now with how the Exynos 5250 in the Nexus 10 is thermally throttled to 4W such that CPU and GPU can't be both running full tilt at the same time. Reply
  • Homeles - Saturday, June 01, 2013 - link

    You're disappointed because your understanding of physics and Moore's Law is poorly developed. The scenario you've provided is a blatant false equivalency.

    According to your desperate desires, the roughly 4GHz processors that launched with Sandy Bridge should be running at twice the clock speed today.

    When you understand that leakage power grows exponentially as transistor geometries shrink, and that power consumption raises exponentially as clock speed rises, you will realize that even the 10% gains that Haswell makes here are a big deal.
    Reply
  • Dal Makhani - Saturday, June 01, 2013 - link

    Homeles, I really appreciate your well said comment, im taking a business degree with an accounting major, but ive always loved building PC's as a hobby. When some of my computer science/engineer friends try to show me the stuff they are learning, i am baffled as its not my area of expertise. I can only imagine how challenging it is to combat the shrinking processes and make performance gains as you said. I have deep respect for Intel and AMD, always trying to utilize their research and engineers to try and make any gains for society. These forum people are just so ignorant sometimes and it baffles me. Reply
  • chizow - Saturday, June 01, 2013 - link

    Hey, similar path as me. :) Don't worry about lack of understanding now, stick to it, keep reading great technical sites like AT, keep an open mind, and you'll get a really good grip on the industry, especially if you are an actual user/enthusiast of the products. Reply

Log in

Don't have an account? Sign up now