Intel Broadwell Architecture Preview: A Glimpse into Core M
by Ryan Smith on August 11, 2014 12:01 PM ESTPutting 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.
Facebook
Twitter
RSS
158 Comments
View All Comments
crispbp04 - Monday, August 11, 2014 - link
Intel has something impressive in the works with Broadwell (at least on paper). I can't wait to get a Broadwell based Surface Pro. Assuming that Microsoft improves an already impressive hardware design from the sp3, the Broadwell iteration will likely be my next computer purchase.frostyfiredude - Monday, August 11, 2014 - link
I have a feeling SP4 will be fundamentally the same design as SP3 save for minor tweaks and improvements. SP3 was clearly designed for a processor with the kind of power profile Broadwell is set to deliver rather than the current Haswell profile. It will be interesting to see which set of SKUs Microsoft will put in the SP4, Core M or Broadwell ULT. Core M has a number of obvious benefits for power and area efficiency, but will it be powerful enough for their market with some features reduced from Haswell and Boradwell ULT.MonkeyPaw - Monday, August 11, 2014 - link
Pure speculation, but I think Intel might already be giving MS premium bins of Haswell for SP3, because SP3 is the only device to date to actually show off the ability to run premium Intel CPUs in a tablet format. Sure, MBA looks great, but SP3 took it to the next level.That said, I doubt that MS will use Core M in SP4, for the same reason we don't have Haswell-Y in SP3 (at least at the high end). It will probably be a step back in processing power to use one.
frostyfiredude - Tuesday, August 12, 2014 - link
Do we know what wattage the Broadwell ULT and Core M chips will be targeting? 15W TDP is clearly too high for the SP3 to handle so moving all SP4 chips to 11.5W like the current Haswell Y looks quite plausible at the moment, it just seems to be a matter of which version of Broadwell will have the 11.5W TDP.Samus - Wednesday, August 13, 2014 - link
15W is only a problem in SP3 to people who use it like a high performance computer (24x7 full load applications) but for general purpose use it barely warms up. We have people running Lightroom 8 hours a day on these things and like the Surface 2's (which I still have) they never got "hot" or "loud".That said, someone in the office infected their SP3 with some malware a few weeks ago (they literally owned the tablet not even 24 hours) and when they handed it to me, it was VERY hot with the fans whirling. Some 800kb task was using 100% of their CPU doing who knows what...at first I thought it was Cryptolocker but it turned out to be retrying a network connection. This was an i5 model, however, and it didn't seem to be throttling. The i3 will presumable run cooler, even at the same TDP.
What people need to keep in mind is these are mobile devices.
IntelUser2000 - Wednesday, August 13, 2014 - link
Broadwell ULT: 15WCore M(previously Broadwell-Y): 4.5W
vlad0 - Friday, August 15, 2014 - link
Isn't the core i3 version of the sp3 based on a Y series chip ?bebimbap - Monday, August 11, 2014 - link
agreed, Broadwell, and skylake will be vast improvements to PCs in general. Intel's Broadwell-Y announcement is all about "small, cool, efficient" while the recent FX-9590 seems more about "big, hot, gluttony" similar to the David vs Goliath story, the interesting part was the small one besting the big one. Ironically Intel is the bigger company. Hopefully AMD's new A1100 pans out as I don't want another Comcast, Microsoft, De Beers or Luxottica.wurizen - Monday, August 11, 2014 - link
well, if amd was as agressive as intel in shrinking dies or whathaveyou, then an AMD FX chip will probably be toe-to-toe to an intel i7-4930k or whatever the 6-core enthusiast intel chip is labeled. and not even die shrinks, but, also aggressive in producing a $500 cpu. imagine that. and you'd probably see an a10-7850k performance in a laptop by now. but, AMD seems content is sitting back and letting the other company do all the work, creating a path. as long as AMD doesn't completely die out, it's fine. we just need an alternative and AMD is the only one. so, go AMD. don't worry about broadwell. build it and we will come. be a niche. convert future x99 users to a future AMD product. and start from there.StevoLincolnite - Monday, August 11, 2014 - link
Except AMD can't be aggressive at shrinking dies.For one, die-shrinks costs money... For fab contracts, man-hours, research and possibly buying technology from other companies such as IBM.
AMD can't aggressively shrink dies anyway, they are at the mercy of fabrication companies like TSMC and Global Foundries, so what they can produce is limited to what they provide.
Intel has always been ahead of the industry in fabrication, the only way AMD can beat Intel is through something ground breaking (Like moving away from silicon?) or if Intel drops the ball, like they did with Netburst.
Or, AMD buys a fab company who is far ahead of Intel, which simply isn't going to happen.
Otherwise they can only compete on price and using an older more mature fabrication process allows them to do just that as the chips are much cheaper to produce, they just need to provide "Good enough" performance to mostly stay relevant, which the FX doesn't really do.