Putting It All Together: Small Core M

Next to power constraints, the final element of Intel’s fanless challenge is the size of the SoC itself. Sub-10mm thickness doesn’t just put constraints on the heat capacity of the device but it also constrains just how large an SoC and its supporting circuitry can be. As a result Intel has focused on making Broadwell-Y the smallest Core processor yet, making the entire SoC under 500mm2 in size.

As was the case with power, reducing the size of Broadwell-Y is a multi-faceted effort. The 14nm process plays a big part here, allowing for one of the smallest Core CPU dice yet. At 82mm2 the Broadwell-Y CPU die is some 37% smaller than the Haswell-Y CPU die, none the less packing a dual-core CPU and a full GPU slice.

With such a small die Intel was in turn able to reduce the size of the entire SoC package through the combination of the reduced die area and further optimizations to the packaging itself. Haswell-Y’s already small ball pitch of .65mm was further reduced to just .5mm, producing a package with Intel’s smallest solder ball pads yet. Intel considers the reduction in the ball pitch to be the key change that allowed Broadwell-Y to be so small, as they were already pad-limited on Haswell-Y despite having ample excess packaging even after taking the CPU die’s larger size into account. As a result Broadwell-Y takes up almost 50% less surface area (XY) than Haswell-Y.

Intel has also made a number of changes for Broadwell-Y to reduce the Z-height of the Y SoCs, as even 1.5mm for the SoC starts to become a significant design constraint in a sub-10mm device. Again owing to the 14nm process, the Z-height of the Broadwell-Y die itself is down to 170um. Meanwhile the Z-height of the substrate has been cut in half from 400um to 200um, which accounts for nearly half of the total reduction in SoC Z-height.

The final element in reducing the SoC Z-height, and what’s likely the most unorthodox change for Broadwell-Y’s packaging, is Intel’s 3DL inductors. The 3DLs aren’t just to improve energy efficiency as we discussed before, but they are part of Intel’s efforts to reduce the SoC size. For Broadwell-Y the 3DLs are on their own PCB on the back of the SoC, extending well below the back of the package. To accommodate this, logic boards housing Broadwell-Y will have a hole in them where the 3DL PCB would be in order to allow the complete SoC to fit. Because there are no BGA connections here this change isn’t quite as radical as it first appears, but it’s a very good example of just what lengths Intel was willing to go to reduce the package Z-height.

All told then, the combination of these space optimizations has reduced Broadwell-Y’s Z-height by nearly 30%, from 1.5mm on Haswell-Y to 1.04mm on Broadwell-Y (3DL PCB not included). By bringing Broadwell-Y’s thickness under 1.1mm, the SoC is now no taller than the other common components on a logic board (e.g. RAM), meaning the SoC will no longer stick out above the other components, which is useful both for saving space and for allowing simpler (flatter) heatsinks.

Finally, the smaller size of the Broadwell-Y package will also have a knock-on effect on the size of the logic board, further feeding into Intel’s goals to get Broadwell-Y into smaller devices. Intel tells us that the size of a complete platform (logic) board for Broadwell-Y has been reduced by roughly 25% as compared to Haswell-Y, allowing Broadwell-Y to better fit into not just thinner devices but overall smaller devices too.

Putting It All Together: Low Power Core M Closing Thoughts
Comments Locked

158 Comments

View All Comments

  • Jaybus - Thursday, August 14, 2014 - link

    TDP is "Total Dissipated Power" and as used in describing processors refers to the maximum instantaneous power that the processor will dissipate. Watt, being a unit unit of power, is the only SI unit suitable for specifying TDP. Since nearly all of the energy expended by a processor is due to heat transfer, a Watt is in this case essentially a measure of the rate of heat transfer, and TDP is the rate of heat transfer required when operating the processor at or near its maximum operating temperature.
  • 473NG3R - Friday, August 15, 2014 - link

    @Jaybus Nope. Please don't correct people with false info. TDP is Thermal Design Power, and it is NOT "peak" power. It is used for marketing as mkozakewich said above as the definition of TDP measurement varies from company to company.
  • 473NG3R - Friday, August 15, 2014 - link

    Oops! I read mkozakewich comment backwards. It is a marketing term in a sense... It is used to design cooling systems but it gets thrown around by marketing groups all the time since TDP limits how small a device can be.
  • frostyfiredude - Monday, August 11, 2014 - link

    With all these reductions in power use on a per core basis and a stagnation of clock speeds we very well could see a quad core i7-5770k with a 65W TDP. I hope Intel plans on bumping up their mainstream high end SKUs to six core, the desktop market doesn't need maximum power use that low.
  • klmccaughey - Monday, August 11, 2014 - link

    Yes this worries me too. I think we will see yet another pointless desktop release, with hardly any improvements (once you consider the bad overclocking).

    I still don't see a desktop Broadwell replacing my 2500k, which runs low(ish) heat at 4.3GHz. What I would love to see is a 5.5GHz Broadwell monster, otherwise I will be skipping broadwell, just like Hazwell and IB before it.

    Surely there has to be some light on the horizon for us gamers?!
  • mapesdhs - Monday, August 11, 2014 - link


    And someone in your position also of course has the double irony of still having more
    performance headroom if they're willing to explore better cooling, etc. H80i would let
    your 2500K run at 5GHz no problem, assuming your particular sample isn't naturally
    limited in some way (actually, even an old TRUE and two typical fans would be fine,
    something I can confirm as I've done it with several different 2700Ks, never mind a
    2500K, so I know the heat isn't an issue). Thus, except for threaded workloads, an
    upgrade would have to be a lot better than what you already have in order to be
    worth bothering with, except if you wanted something new due to newer tech sych
    as PCI Express, M.2, etc.

    Anyway, this is why IB was so bad with its default metal cap setup, ie. Intel made SB
    too good which resulted in lots of 2500K owners just not seeing the point in upgrading.
    Thing is, they've still not boosted performance enough to make upgrading worthwhile
    for most SB users (the exception of course being non-K owners).

    My 2700K runs at 5.0. Takes me 3 mins to set this up on an ASUS M4E, ludicrously
    easy to do. HW at 4.5 is about the same or a bit quicker, but much more complex to
    do, given how many samples won't go that high, and for gaming it makes no difference
    at all. I suppose the 4790K can help on the oc front a bit, but the cost has put me off so
    far, used 2700Ks are so much cheaper. I want to build a 4K gaming system next, but
    that'll be X99 instead I think (don't see the point of mainstream chipsets for 4K gaming,
    surely with multiple GPUs the limited PCIe lanes is going to hold back performance
    eventually).

    Ian.
  • garadante - Tuesday, August 12, 2014 - link

    I wish I had your luck with SB. My 2500k will only do 4.5 GHz stable with 1.365 vcore (in reality, monitoring software reports 1.391~ vcore with my motherboard's way of handling vdroop). If I could go any higher, I'd have to dump a load of voltage into it that I'm just afraid of doing. Though this may be a motherboard problem as it's given me problems before, but that still wouldn't account for a 500 MHz variance. I guess I just got a bad SB. :/
  • pt2501 - Monday, August 11, 2014 - link

    You said it. My 2500k at 4.6 Ghz isn't going anywhere unless something revolutionary is released. Best processor I ever bought.
  • ZeDestructor - Monday, August 11, 2014 - link

    They want you to go up to LGA2011 for the solder-based TIM. And only because of die-cracking issues (I found the papers covering that) on smaller dies that force them to use boring old polymer TIM.
  • steve wilson - Tuesday, August 12, 2014 - link

    My thoughts exactly. My 2500k runs everything fine for me at 4.4ghz. I will be skipping Broadwell and praying Skylake will deliver.

Log in

Don't have an account? Sign up now