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

  • Ian Cutress - Saturday, October 7, 2017 - link

    We finally got in contact with the Civ6 dev team to integrate the AI benchmark into our suite better. You should see it moving forward. Reply
  • Koenig168 - Friday, October 6, 2017 - link

    Hmm ... rather disappointing that Anandtech did not include Ryzen 1600/X until called out by astute readers. Reply
  • mkaibear - Friday, October 6, 2017 - link

    ...apart from including all the data in their benchmark tool, which they make freely available, you mean? They put in the CPUs they felt that were most relevant. The readership disagreed, so they changed it from their benchmark database. That level of service is almost unheard of in the industry and all you can do is complain. Bravo. Reply
  • Koenig168 - Friday, October 6, 2017 - link

    Irrelevant. While I agree with most of what you said, that does not change the fact that Anandtech did not include Ryzen 1600/X until called out by astute readers. To make things a little clearer for you, the i7-8700 is a 6C/12T processor. The Ryzen 1600 is a 6C/12T processor. Therefore, a comparison with the Ryzen 1600 is relevant.

    You should have addressed the point I made. Instead all you can do is complain about my post. Bravo. (In case this goes over your head again, that last bit is added just to illustrate how pointless such comments are.)
  • mkaibear - Saturday, October 7, 2017 - link

    So your point is, in essence, "they didn't do what I wanted them to do so they're damned for all time".

    They put up the comparison they felt was relevant, then someone asked them to include something different - so they did it. They listened to their readers and made changes to an article to fix it.

    Should they have put the R5 in the original comparison? Possibly. I can see arguments either way but if pushed I'd have said they should have done - but since even the 1600X gets beaten by the 8400 in virtually every benchmark on their list (as per they would then have been accused by the lurking AMD fanboys of having picked comparisons to make AMD look bad (like on every other article where AMD gets beaten in performance).

    So what are you actually upset about? That they made an editorial decision you disagree with? You can't accuse them of hiding data since they make it publicly accessible. You can't accuse them of not listening to the readers because they made the change when asked to. Where's the issue here?
  • mkaibear - Saturday, October 7, 2017 - link

    OK on further reading it's not "virtually every" benchmark on the list, just more than half. It's 50% i5 win, 37% R5 win, 12% tied. So not exactly a resounding triumph for the Ryzen but not as bad as I made it out to be.

    In the UK the price differential is about £12 in favour of the i5, although the motherboard is about £30 more expensive (though of course Z370 is a lot more fully featured than B650) so I think pricing wise it's probably a wash - but if you want gaming performance on anything except Civ VI then you'd be better off getting the i5.

    ...oh and if you don't want gaming performance then you'll need to buy a discrete graphics card with the R5 which probably means the platform costs are skewed in favour of Intel a bit (£25 for a GF210, £32 for a R5 230...)
  • watzupken - Saturday, October 7, 2017 - link

    As mentioned when I first called out this omission, I would think comparing a 6 vs 4 core irrelevant. This is what AnandTech recommended to lookout for on page 4 "Core Wars": Core i5-8400 vs Ryzen 5 1500X.
    You be the judge if this makes sense when there is a far better competition/ comparison between the i5 8400 vs R5 1600. Only when you go reading around and you realized that hey, the i5 8400 seems to be losing in some areas to the 1600. I give AnandTech the benefit of the doubt, so I am done debating what is relevant or not.
  • KAlmquist - Friday, October 6, 2017 - link

    The Anandtech benchmark tool confirms what Ryan indicated in the introduction: the i7-8700k wins against the 1600X across the board, due faster clocks and better IPC. The comparison to the i5-8400 is more interesting. It either beats the 1600X by a hair, or loses rather badly. I think the issue is the lack of hyperthreading on the i5-8400 makes the 1600X the better all-around performer. But if you mostly run software that can't take advantage of more than 6 threads, then the i5-8400 looks very good.

    Personally, I wouldn't buy i5-8400 just because of the socket issue. Coffee Lake is basically just a port of Skylake to a new process, but Intel still came out with a new socket for it. Since I don't want to dump my motherboard in a landfill every time I upgrade my CPU, Intel needs a significantly superior processor (like they had when they were competing against AMD's bulldozer derivatives) to convince me to buy from them.
  • GreenMeters - Friday, October 6, 2017 - link

    So Intel still isn't getting their head out of their rear and offering the option of a CPU that trades all the integrated GPU space for additional cores? Moronic. Reply
  • mkaibear - Friday, October 6, 2017 - link

    Integrated graphics make up more than 70% of the desktop market. It's even greater than that for laptops. Why would they sacrifice their huge share of that 70% in order to gain a small share of the 30%? *that* would be moronic.

    In the meantime you can know that if you buy a desktop CPU from Intel it will have an integrated GPU which works even with no discrete graphics card, and if you need one without the integrated graphics you can go HEDT.

    Besides, the limit for Intel isn't remotely "additional space", they've got more than enough space for 8/10/12 CPU cores - it's thermal. Having an integrated GPU which is unused doesn't affect that at all - or arguably it gives more of a thermal sink but I suspect in truth that's a wash.

Log in

Don't have an account? Sign up now