Silicon and Process Nodes: 14++

Despite being somewhat reserved in our pre-briefing, and initially blanket labeling the process node for these chips as ‘14nm’, we can confirm that Intel’s newest ‘14++’ manufacturing process is being used for these 8th Generation processors. This becomes Intel’s third crack at a 14nm process, following on from Broadwell though Skylake (14), Kaby Lake (14+), and now Coffee Lake (14++).

With the 8th Generation of processors, Intel is moving away from having the generation correlate to both the process node and microarchitecture. As Intel’s plans to shrink its process nodes have become elongated, Intel has decided that it will use multiple process nodes and microarchitectures across a single generation of products to ensure that every update cycle has a process node and microarchitecture that Intel feels best suits that market. A lot of this is down to product maturity, yields, and progress on the manufacturing side.

Intel's Core Architecture Cadence (8/20)
Core Generation Microarchitecture Process Node Release Year
2nd Sandy Bridge 32nm 2011
3rd Ivy Bridge 22nm 2012
4th Haswell 22nm 2013
5th Broadwell 14nm 2014
6th Skylake 14nm 2015
7th Kaby Lake 14nm+ 2016
8th Kaby Lake Refresh
Coffee Lake
Cannon Lake
9th Ice Lake?
10nm+ 2018?
Unknown Cascade Lake (Server) ? ?

Kaby Lake was advertised as using a 14+ node with slightly relaxed manufacturing parameters and a new FinFET profile. This was to allow for higher frequencies and better overclocking, although nothing was fundamentally changed in the core manufacturing parameters. With Coffee Lake at least, the minimum gate pitch has increased from 70nm for 84nm, with all other features being equal.

Increased gate pitch moves transistors further apart, forcing a lower current density. This allows for higher leakage transistors, meaning higher peak power and higher frequency at the expense of die area and idle power.

Normally Intel aims to improve their process every generation, however this seems like a step ‘back’ in some of the metrics in order to gain performance. The truth of the matter is that back in 2015, we were expecting Intel to be selling 10nm processors en-masse by now. As delays have crept into that timeline, the 14++ note is holding over until 10nm is on track. Intel has already stated that 10+ is likely to be the first node on the desktop, which given the track record on 14+ and 14++ might be a relaxed version of 10 in order to hit performance/power/yield targets, with some minor updates. Conceptually, Intel seems to be drifting towards seperate low-power and high-performance process nodes, with the former coming first.

Of course, changing the fin pitch is expected to increase the die area. With thanks to HEKPC (via Videocardz), we can already see a six-core i7-8700K silicon die compared to a quad-core i7-7700K.

The die area of the Coffee Lake 6+2 design (six cores and GT2 graphics) sits at ~151 mm2, compared to the ~125 mm2 for Kaby Lake 4+2 processor: a 26mm2 increase. This increase is mainly due to the two cores, however there is a minor adjustment in the integrated grpahics as well to support HDCP 2.2, not to mention any unpublished changes Intel has made to their designs between Kaby Lake and Coffee Lake.

The following calculations are built on assumptions and contain a margin of error

With the silicon floor plan, we can calculate that the CPU cores (plus cache) account for 47.3% of the die, or 71.35 mm2. Divided by six gives a value of 11.9 mm2 per core, which means that it takes 23.8 mm2 of die area for two cores. Out of the 26mm2 increase then, 91.5% of it is for the CPU area, and the rest is likely accounting for the change in the gate pitch across the whole processor. 

The Coffee Lake 4+2 die would then be expected to be around ~127 mm2, making a 2mm2 increase over the equivalent Kaby Lake 4+2, although this is well within the margin of error for measuring these processors. We are expecting to see some overclockers delid the quad-core processors soon after launch.

In previous Intel silicon designs, when Intel was ramping up its integrated graphics, we were surpassing 50% of the die area being dedicated to graphics. In this 6+2 design, the GPU area accounts for only 30.2% of the floor plan as provided, which is 45.6 mm2 of the full die.

Memory Support on Coffee Lake

With a new processor generation comes an update to memory support. There is always a small amount of confusion here about what Intel calls ‘official memory support’ and what the processors can actually run. Intel’s official memory support is typically a guarantee, saying that in all circumstances, with all processors, this memory speed should work. However motherboard manufacturers might offer speeds over 50% higher in their specification sheets, which Intel technically counts as an overclock.

This is usually seen as Intel processors having a lot of headroom to be conservative, avoid RMAs, and maintain stability. In most cases this is usually a good thing: there are only a few niche scenarios where super high-speed memory can equate to tangible performance gains* but they do exist.

*Based on previous experience, but pending a memory scaling review

For our testing at least, our philosophy is that we test at the CPU manufacturers’ recommended setting. If there is a performance gain to be had from slightly faster memory, then it pays dividends to set that as the limit for official memory support. This way, there is no argument on what the rated performance of the processor is.

For the new generation, Intel is supporting DDR4-2666 for the six-core parts and DDR4-2400 for the quad-core parts, in both 1DPC (one DIMM per channel) and 2DPC modes. This should make it relatively simple, compared to AMD’s memory support differing on DPC and type of memory.

It gets simple until we talk about AIO designs using the processors, which typically require SODIMM memory. For these parts, for both quad-core and hex-core, Intel is supporting DDR4-2400 at 1DPC and DDR4-2133 at 2DPC. LPDDR3 support is dropped entirely. The reason for supporting a reduced memory frequency in an AIO environment with SODIMMs is because these motherboards typically run their traces as chained between the memory slots, rather than a T-Topology which helps with timing synchronization. Intel has made the T-Topology part of the specification for desktop motherboards, but not for AIO or integrated ones, which explains the difference in DRAM speed support.

These supported frequencies follow JEDEC official sub-timings. Familiar system builders will be used to DDR4-2133 at a CAS Latency of 15, but as we increase the speed of the modules, the latency increases to compensate:

Intel’s official sub-timing support at DDR4-2666 is 19-19-19. Outside of enterprise modules, that memory does not really exist, because memory manufacturers can seem to mint DDR4-2666 16-17-17 modules fairly easily, and these processors are typically fine with those sub-timings. CPU manufacturers typically only state ‘supported frequency at JEDEC sub-timings’ and do not go into sub-timing discussions, because most users care more about the memory frequency. If time permits, it would be interesting to see just how much of a performance deficit the official JEDEC sub-timings provide compared to what memory is actually on sale.

The Intel Coffee Lake Early Review Physical Design, Integrated Graphics, and the Z370 Chipset: Differences


View All Comments

  • mkaibear - Wednesday, October 11, 2017 - link

    Sheesh. Where to start?

    1) Yes, you're right, if the iGPU isn't being used then it will be disabled, and therefore you don't need to cool it. Conversely, if you have additional cores then your OS *will* use them, and therefore you *do* need to cool them.

    iGPU doesn't draw very much power at all. HD2000 drew 3W. The iGPU in the 7700K apparently draws 6W so I assume the 8700K with a virtually identical iGPU draws just as much (figures available via your friendly neighbourhood google). Claiming the iGPU has a higher power budget than the CPU cores is frankly ridiculous. (in fact it also draws less than .2W when it's shut down which means that having it in there is far outweighed by the additional thermal sink available, but anyway)

    2) Large companies with big IT organisations don't actually forego the Core line altogether and go with Xeons. They could if they wanted to, but in general they still use off-the shelf Dells and HPs for everything except extremely bespoke setups - because, as I previously mentioned, "hardware is cheap, people are expensive" - getting an IT department to build and maintain bespoke computers is hilariously expensive. No-one is arguing that for an enthusiast building their own computer that the option of the extra cores would be nice, but my point all along has been that Intel isn't going to risk sacrificing their huge market share in the biggest market to gain a slice of a much smaller market. That would be extremely bad business.

    3) The market isn't "clamoring for a more rational and effective alternative" because if it was then Ryzen would have flown off the shelves much faster than it did.

    Bottom line: business IT wants simple solutions, the fewer parts the better. iGPUs on everything fulfil far more needs than dGPUs for some and iGPUs for others. iGPUs make designing systems easier, they make swapouts easier, they make maintenance easier, they reduce TCO, they reduce RMAs and they just make IT staff's lives easier. I've run IT for a university, a school and a manufacturing company, and for each of them the number of computers which needed a fast CPU outweighed the number of computers which needed a dGPU by a factor of at least 10:1 - and the university I worked for had a world-leading art/media/design dept and a computer game design course which all had dGPUs. The average big business has even less use for dGPUs than the places I've worked.

    If you want to keep trying to argue this then can you please answer one simple question: why do you think it makes sense for Intel to prioritise a very small area in which they don't have much market share over a very large area in which they do? That seems the opposite of what a successful business should do.
  • watzupken - Saturday, October 7, 2017 - link

    There are pros and cons of having integrated graphics. It sure takes up a lot of die space, but it is something that allows Intel to sell a lot of chips. Amongst enthusiasts, this is unnecessary, but this group may only represent a small percentage vs corporates that need only decent CPU and no need for fancy graphics. To be honest, Intel could likely have created a 8 core processor easily since the die size is still fairly small for Coffee Lake, but they chose not to. I don't think it is a matter of the graphic that is holding them back. Reply
  • James5mith - Friday, October 6, 2017 - link

    Now to wait for the generation of Intel CPU's with native Thunderbolt3 on-die like Intel announced earlier this year. Reply
  • Zingam - Saturday, October 7, 2017 - link

    Why is that a good thing? Reply
  • ReeZun - Friday, October 6, 2017 - link

    "The difference between the Ryzen 5 1500X and the Core i3-8350K would be interesting, given the extreme thread deficit (12 threads vs 4) between the two."

    The 1500X houses 8 threads (not 12).
  • watzupken - Saturday, October 7, 2017 - link

    The difference between the R5 1500X and i3 8350K goes beyond just the number of threads. The cache is also 2x more on the Ryzen chip. However, the i3 chip have the advantage of being able to reach higher clockspeed. I do agree that this will be an interesting comparison. Reply
  • sweeper765 - Friday, October 6, 2017 - link

    I'm not up to date with current bios versions.
    Is multi-core enhancement still present in z370 motherboards? That would get rid of all those differences in turbo speeds. I know it is technically overclocking but i bet it's a pretty safe procedure without increasing the voltages.

    Also, what's the deal with the 8700? Is it just as good as 8700k (minus 100mhz) if one decides not to overclock? Just trying to gather as many practical facts as i can before formulating an upgrade plan (sandy bridge user hehe )

    This cpu family looks good on specs and benches (maybe the first worthy successor to sandy bridge) but it's not perfect. I hate that Intel decided not to solder, i expect temperatures to soar in the high 80's. Also the current motherboards are somewhat lacking in ports (usb, lan, sata).

    I love my sandy bridge setup though.
    6 1/2 years old and still going strong. Overclocked, cool, stable, silent. With current cpu's you don't get all these points.
    Even if i upgrade i'm not going to touch it.
  • Ian Cutress - Saturday, October 7, 2017 - link

    Is multi-core enhancement still present in z370 motherboards?

    As an option, yes.
    As default? Will vary board to board. You can disable it.
    However we had trouble with one of our boards: disabling MCT/MCE and then enabling XMP caused the CPU to sit at 4.3 GHz all day. Related to a BIOS bug which the vendor updated in a hurry.
  • Jodiuh - Friday, October 6, 2017 - link

    What’s up with those rise of Tomb Raider benchmarks? Am I too seriously believ the i5 7400 is more capable than the 8700K...did I miss the overclocking part?

    Tech reports review much better with results that make sense.
  • peevee - Friday, October 6, 2017 - link

    "Core i5-8600K and the Core i7-8700. These two parts are $50 apart, however the Core i7-8700 has double the threads, +10% raw frequency"

    +10%? Count again.

Log in

Don't have an account? Sign up now