The Skylake CPU Architecture

As with any new Intel architecture, the devil is in the details. Previously at AnandTech we have been able to provide deep dives into what exactly is going on in the belly of the beast, although the launch of Skylake has posed a fair share of problems.

Nominally we rely on a certain amount of openness from the processor/SoC manufacturer in providing low level details that we can verify and/or explain. In the past, this information has typically been provided in advance of the launch by way of several meetings/consultations with discussions talking to the engineers. There are some things we can probe, but others are like a black box. The black box nature of some elements, such as Qualcomm’s Adreno graphics, means that it will remain a mystery until Pandora’s box is opened.

In the lead up to the launch of Intel’s Skylake platform, architecture details have been both thin on the ground and thin in the air, even when it comes down to fundamental details about the EU counts of the integrated graphics, or explanations regarding the change in processor naming scheme. In almost all circumstances, we’ve been told to wait until Intel’s Developer Forum in mid-August for the main reason that the launch today is not the full stack Skylake launch, which will take place later in the quarter. Both Ryan and I will be at IDF taking fastidious notes and asking questions for everyone, but at this point in time a good portion of our analysis comes from information provided by sources other than Intel, and while we trust it, we can't fully verify it as we normally would.

As a result, the details on the following few pages have been formed through investigation, discussion and collaboration outside the normal channels, and may be updated as more information is discovered or confirmed. Some of this information is mirrored in our other coverage in order to offer a complete picture in each article as well. After IDF we plan to put together a more detailed architecture piece as a fundamental block in analyzing our end results.

The CPU

As bad as it sounds, the best image of the underlying processor architecture is the block diagram:

From a CPU connectivity standpoint, we discussed the DDR3L/DDR4 dual memory controller design on the previous page so we won’t go over it again here. On the PCI-Express Graphics allocation side, the Skylake processors will have sixteen PCIe 3.0 lanes to use for directly attached devices to the processor, similar to Intel's previous generation processors. These can be split into a single PCIe 3.0 x16, x8/x8 or x8/x4/x4 with basic motherboard design. (Note that this is different to early reports of Skylake having 20 PCIe 3.0 lanes for GPUs. It does not.)

With this, SLI will work up to x8/x8. If a motherboard supports x8/x4/x4 and a PCIe card is placed into that bottom slot, SLI will not work because only one GPU will have eight lanes. NVIDIA requires a minimum of PCIe x8 in order to enable SLI. Crossfire has no such limitation, which makes the possible configurations interesting. Below we discuss that the chipset has 20 (!) PCIe 3.0 lanes to use in five sets of four lanes, and these could be used for graphics cards as well. That means a motherboard can support x8/x8 from the CPU and PCIe 3.0 x4 from the chipset and end up with either dual-SLI or tri-CFX enabled when all the slots are populated.

DMI 3.0

The processor is connected to the chipset by the four-lane DMI 3.0 interface. The DMI 3.0 protocol is an upgrade over the previous generation which used DMI 2.0 – this upgrade boosts the speed from 5.0 GT/s (2GB/sec) to 8.0 GT/s (~3.93GB/sec), essentially upgrading DMI from PCIe 2 to PCIe 3, but requires the motherboard traces between the CPU and chipset to be shorter (7 inches rather than 8 inches) in order to maintain signal speed and integrity. This also allows one of the biggest upgrades to the system, chipset connectivity, as shown below in the HSIO section.

CPU Power Arrangements

Moving on to power arrangements, with Skylake the situation changes as compared to Haswell. Prior to Haswell, voltage regulation was performed by the motherboard and the right voltages were then put into the processor. This was deemed inefficient for power consumption, and for the Haswell/Broadwell processors Intel decided to create a fully integrated voltage regulator (FIVR) in order to reduce motherboard cost and reduce power consumption. This had an unintended side-effect – while it was more efficient (good for mobile platforms), it also acted as a source of heat generation inside the CPU with high frequencies. As a result, overclocking was limited by temperatures and the quality of the FIVR led to a large variation in results. For Skylake on the desktop, the voltage regulation is moved back into the hands of the motherboard manufacturers. This should allow for cooler processors depending on how the silicon works, but it will result in slightly more expensive motherboards.

A slight indication of this will be that some motherboards will go back to having a large amount of multiplexed phases on the motherboard, and it will allow some manufacturers to use this as a differentiating point, although the usefulness of such a design is sometimes questionable.

Also Launching Today: Z170 Motherboards, Dual Channel DDR4 Kits Skylake's iGPU: Intel Gen9
Comments Locked

477 Comments

View All Comments

  • ES_Revenge - Friday, August 14, 2015 - link

    Umm what the heck happened to the power consumption? In particular the i7/6700K. It's not really shown thoroughly in this review but the Broadwell CPUs are more power-efficient it seems. While the 6700K has a half GHz faster clock speed, it also has a much lesser GPU. To begin with, both the i5 and i7 Skylake parts have higher TDPs than the Broadwell desktop parts, and then the 6700K can actually draw over 100W when loaded. This is above its TDP and also significantly more than its 6600K counterpart which runs only a few hundred MHz slower. Odd.

    I mean I think we were all waiting for a desktop CPU that didn't have the power constraints as the Broadwell CPUs did but I don't think this is exactly what anyone was expecting. It's like these Skylake CPUs don't just take more power but they do so...for no reason at all. Sure they're faster but not hugely so; and, again, their iGPUs are significantly slower than Broadwell's. So their slight speed advantage came at the price of markedly increased power consumption over the previous gen.

    That only leads me to the question--WTF? lol What happened here with the power consumption? And losing that IVR didn't seem to help anything, eh? Skylake is fast and all but TBH I was more impressed *overall* with Broadwell (and those CPUs you can't even find for sale anywhere, the last time I checked--a few weeks ago). Granted as we've seen in 2nd part of the Broadwell review it's not a stellar OCer but still, overall it seems better to me than Skylake.

    It's kind of funny because when Broadwell DT launched I was thinking of how "Intel is mainly focusing on power consumption these days", meaning I thought they weren't focused enough on performance of DT CPUs. But it seems they've just thrown that out the window but the performance isn't anything *spectacular* from these CPUs, so it just seems like a step backwards. It's like with Broadwell they were showing just how much performance they could do with both CPU and iGPU with a minimum of power consumption--and the result was impressive. Here it's like they just forgot about that and said "It's Skylake...it's new and better! Everyone buy it!" Not really that impressive.
  • janolsen - Friday, August 14, 2015 - link

    Stupid question:
    Can Skylake IGP easily play back 4K video. Thinking of a person just using a 4K screen for Youtube stuff, not gaming...
  • ES_Revenge - Saturday, August 15, 2015 - link

    Yeah it can. This one of the very few improvements over Broadwell/previous HD Graphics implementations. It has a "full" HEVC decode solution built in, unlike the "hybrid" solutions they had previously. If you look on the 4th page of the review it actually goes pretty in-depth about this (not sure how you missed that?).
  • alacard - Friday, August 14, 2015 - link

    It's clear you put a ton of work into this Ian, many thanks.
  • Flash13 - Monday, August 17, 2015 - link

    So, far Intel Core i7-6700K 8M Skylake Quad-Core 4.0GHz is just vapor to the public.
  • somatzu - Wednesday, August 19, 2015 - link

    "So where'd you get your degree?"

    "Anandtech comments section."
  • superjim - Friday, August 21, 2015 - link

    I'm still not convinced this is a worthwhile upgrade from Sandy Bridge. If I can get 4.8 from a 2700K and maybe 4.6 from a 6700k, factor in cost difference, is it really worth it? At the end of the day, cpu/mobo/ram would be near $700 for maybe a 15% speed bump overall.
  • watzupken - Friday, September 4, 2015 - link

    From a desktop standpoint, there is very little incentive for one to upgrade. The new gen mainly targets power savings, so likely to benefit mobile users, i.e. Ultra Books and tablets.

    As far as Intel is trying to target those people still on Sandy and Ivy Bridge to upgrade, they fail to account for the cost of upgrade for a paltry improvement in performance. To upgrade from SB, one has to upgrade the ram, motherboard and CPU, and on top of that, need to separately purchase a heatsink since they want to cut cost.
  • CynicalCyanide - Saturday, August 22, 2015 - link

    Question to the Authors: You've noted two DDR4 equipped mobos in the "Test Setup" section, but you've also tested DDR3 equipped Skylake. Which motherboard did you use for that?

    Furthermore, in a previous article it was mentioned that Z170 wouldn't be able to handle 'regular' 1.5V DDR3, but here apparently it wasn't an issue reusing old 1.5V RAM after a voltage adjustment. Was there any special method required aside from booting as per normal into the BIOS and adjusting the voltage?
  • TiberiuC - Saturday, August 22, 2015 - link

    Everything comes down eventualy to "Intel vs AMD". What Intel did with Core2Duo was the right path to go, what AMD did was so wrong and that sometimes happen when you inovate. AMD stopped with the last FX series and went back to the drawing board and that is a wise decision. What will ZEN do? i am expecting Ivi Bridge performance maybe touching haswell here and there. If this wont happen, it is bad for them and very bad for us. Intel is starting to milk the customers acting like there is a monopoly. I did buy my 2600k for 300$ (after rebates), i have to say that the price of the 6700k is well, meh...

Log in

Don't have an account? Sign up now