Competing Against Itself: 3.9 GHz Ice Lake-U on 10nm vs 4.9 GHz Comet Lake-U on 14nm

At the same time that Intel is releasing Ice Lake, we have confirmed from multiple sources that the company intends to release another generation of mobile products based on 14nm as well. This line of hardware, also called Intel 10th Gen Core, will be under the internal codename ‘Comet Lake’, and go after a similar power distribution to what Ice Lake will. There are a few differences in the design worth noting, and a big one that Intel will have a hard time organizing its marketing materials for.

The differences between Ice Lake-U and Comet Lake-U are set to be quite confusing. Leaks from various OEMs about upcoming products give us the following:

Ice Lake: The Core i7-1065G7

Ice Lake-U hardware, based on 10nm, will be given a ‘G’ in the product name, such as i7-1065G7. This breaks down such that

  • i7 = Core i7
  • 1065 = from the 10th Gen Core
  • 1065 = position ‘65’ relative to the rest of the other Ice lake processors,
  • G7 = ‘Graphics Level 7’, which we believe to be the highest.

Intel has stated that the Ice Lake-U hardware will come in at 9W, 15W, and 28W, as described in the previous pages, offering a highest turbo clock of 4.1 GHz, 64 EUs of Gen11 graphics, suitable for up to 1.1 TF of FP64 calculations. We suspect that the 4.1 GHz turbo frequency will be given to the 28W model following previous Intel launches, which means that the 15W part is likely to turbo to a few hundred MHz lower. Based on the Ice Lake plans we know, it seems that Intel is only targeting up to quad-core designs, but Ice Lake does support LPDDR4. Due to using the 10nm process, and with additional power refinements, Ice Lake hardware is expected to have longer a battery life compared to Comet Lake, although we will see this in product reviews through the year.

Comet Lake: The Core i7-10510U

Contrast this to Comet Lake-U, which is another round of processors based on 14nm. OEMs have given some light onto these processors, which should offer up to six cores. The naming of the processors follows on from the 8th Gen and 9th Gen parts, but is now under 10th Gen. This means that the Core i7-10510U breaks down as:

  • i7 = Core i7
  • 10510 = from the 10th Gen Core family,
  • 10510 =  position ‘51’ relative to the rest of Comet Lake
  • U = U-series processor, 15-28W

OEM listings have shown Comet Lake-U to turbo up to 4.9 GHz on the best quad-core processor, while we have seen 9th gen hardware hit 5.0 GHz in the larger H-series designs.

For a full side-by-side comparison:

Ice Lake-U vs Comet Lake-U
Ice Lake-U* AnandTech Comet Lake-U*
10+ Lithography '14nm class'
i7-1065G7 Example CPU Name i7-10510U
9W
15W
28W
TDP Options 15W
28W?
Same as 9th Gen?
Up to 4C Core Counts Up to 6C (expected)
Sunny Cove CPU Core Skylake+++
Up to 64 EUs
Gen11
GPU GT2 Core Up to 24 EUs
Gen9.5
3.9G (15W)
4.1G (28W)
Highest Turbo 4.9G? (15W)
5.0G+ ?
DDR4-3200
LPDDR4-3733
DDR DDR4-2667
LPDDR3-2133
AVX-512 AVX AVX2
*All details are not yet confirmed by Intel, but shown on partner websites/trusted sources

Should Intel go ahead with the naming scheme, it is going to offer a cluster of mixed messages, even to end-users that understand the naming scheme. For those that don’t, there might not be an obvious way to tell a 10th Gen Ice Lake system and a 10th Gen Comet Lake system part from just reading the specification sheet, especially if the vendor lists it just as ‘10th Gen Core i7’.

Intel is trying to mitigate some of this with Project Athena, which is a specification for premium 10th Gen designs. In order to meet Athena specifications, you technically don’t need to have an Ice Lake processor, but it definitely does help with the graphics and battery life targets. We’re unsure at this point if Intel will add in distinct labeling to Athena approved devices or not, but this might be one way to discern between the two. The other is to look for the letter: G means Ice Lake, U means Comet Lake.

So the question is about what matters most to users?

If you want raw CPU frequency and cores, then Comet Lake still has benefits there, even if we add on Intel’s expected ‘+18%’ IPC claims. It would all come down to how the turbo plays out in each device, and Intel states that it is working closer than ever before with its OEM partners to optimize for performance.

Ice Lake systems on the other hand are going to offer better graphics, are initially all likely to be under the Project Athena heading, and provide good connectivity (Wi-Fi 6), good displays, and really nice battery life for the class of device. Ice Lake is going to play more in the premium space too, at least initially, which might indicate that Comet Lake could be angled down the price bracket.

To be honest, we should have been expecting this. When Dr. Murthy Renduchintala joined Intel a couple of years ago, he was quoted as saying that he wants to disaggregate the ‘generation’ from the lithography, and offer a range of products within each generation. The fruits of that campaign started with the last round of mobile platforms, and the fruits of that endeavor will ripen through the Ice Lake/Comet Lake kerfuffle*. It’s going to be agonizing to tell users the difference, and even more so if OEMs do not list exact CPU specifications in their online listings. Intel has been so forthright with two distinct brands, the ‘X’ Gen Core and the Core ‘i7/i5/i3’ naming, that now both are ultimately meaningless to differentiate between two different types of products.

What should be the solution here? On initial thoughts, I would have pushed Ice Lake as an 11th Gen Core. It’s a new and exciting product, with a updated microarchitecture, better graphics, and leading edge lithography, along with Project Athena, it needs to be categorically separated from any other processors it might be competing with. It’s either that, or come up with an alternative naming scheme for it all. At this point, Intel is heading to a sticky mess, where it’s competing against itself and the casual user who hasn’t done meticulous research might not end up with the optimum product.

*To be clear, in the past, Intel used to separate product line microarchitecture based on the nth Gen Core designation. This does not happen anymore – a single ‘nth Gen Core’ brand might have 3+ different microarchitectures depending on what product you are looking at. It is ultimately confusing for any end-customers that have a passing knowledge of Intel’s product lines, and highly annoying to anyone with technical prowess discussing Intel’s products. I hate it. I understand it, but I hate it.

Performance Claims: +18% IPC vs. SKL, +47% Perf vs. BDW Intel’s Ice Lake and Sunny Cove: A Welcome Update, with Questions on Execution
Comments Locked

107 Comments

View All Comments

  • The_Assimilator - Wednesday, July 31, 2019 - link

    Getting Thunderbolt on-die is huge for adoption. While I doubt many laptop manufacturers will enable more than a single TB port, desktop is an entirely different kettle of fish.
  • umano - Wednesday, July 31, 2019 - link

    I am afraid but I cannot consider 4 cores cpu as premium
  • Khenglish - Wednesday, July 31, 2019 - link

    This honestly is looking like the worst architecture refresh since Prescott. IPC increases are getting almost completely washed out by loss in frequency. I wonder if this would have happened if Ice Lake came out on 14nm. Is the clock loss from uArch changes, process change, or a mix of both?

    Performance of an individual transistor has been decreasing since 45nm, but overall circuit performance kept improving due to interconnect capacitance decreasing at a faster rate at every node change. It looks like at Intel 10nm, and TSMC 7nm that this is no longer true, with transistor performance dropping off a cliff faster than interconnect capacitance reduction. 5nm and 3nm should be possible, but will anyone want to use them?
  • Sivar - Wednesday, July 31, 2019 - link

    "...with a turbo frequency up to 4.1 GHz"
    This is the highest number I have come across for the new 10th generation processors, and according to SemiAccurate (which is accurate more often than not), this is likely not an error.

    If this value is close to desktop CPU limitations, the low clock speed all but erases the 18% IPC advantage -- an estimate likely based on a first-gen Skylake.
    Granted, the wattage values are low, so higher-wattage units should run at least a bit faster.
  • Farfolomew - Wednesday, July 31, 2019 - link

    I’m a bit confused by the naming scheme. Ian, you say: “The only way to distinguish between the two is that Ice Lake has a G in the SKU and Comet Lake has a U”

    But that’s not what’s posted in several places throughout the article. The ICL processors are named Core iX-nnnnGn where CML are Core iX-nnnnnU. Comet lake is using 5 digits and Ice Lake only 4 (1000 vs 10000 series).

    Is this a typo or will ICL be 1000-series Core chips?
  • name99 - Wednesday, July 31, 2019 - link

    Regarding AI on the desktop. The place where desktop AI will shine is NLP. NLP has lagged behind vision for a while, but has acquired new potency with The Transformer. It will take time for this to be productized, but we should ultimately see vastly superior translation (text and speech), spelling and grammar correction, decent sentiment analysis while typing, even better search.

    Of course this requires productization. Google’s agenda is to do this in the cloud. MS’ agenda I have no idea (they still have sub-optimal desktop search). So probably Apple will be first to turn this into mainstream products.

    Relevant to this article is that I don’t know the extent to which instructions and micro-architectures optimized for CNNs are still great for The Transformer (and the even newer and rather superior Transformer-XL published just a few months ago). This may all be a long time happening on the desktop if INTC optimized too much purely for vision, and it takes another of their 7 years to turnaround and update direction...
  • croc - Thursday, August 1, 2019 - link

    It seems that Ice Lake / Sunny Cove will have hardware fixes for Spectre and Meltdown. I would like to see some more information on this, such as how much speed gain, whether the patch is predictive (so as to block ALL such OOE / BP exploits) etc.
  • MDD1963 - Thursday, August 1, 2019 - link

    A month or so ago, we heard a few rumors that the CPUs were ahead ~18% in IPC (I see that number again in this article), but are down ~20+% in clock speed.... ; it would be nice to see at least one or two performance metrics/comparisons on a shipped product. :)
  • isthisavailable - Thursday, August 1, 2019 - link

    Unlike Ryzen mobile, intel’s “upto” 64 EUs part will probably only ship in like 2 laptops. Therefore amd has more designs in my book. I don’t understand people who buy expensive 4K laptops with intel integrated gfx which can’t even render windows 10 ui smoothly.
    Looking forward to Zen2 + navi based 7nm APU.
  • Bulat Ziganshin - Thursday, August 1, 2019 - link

    > it can be very effective: a dual core system with AVX-512 outscored a 16-core system running AVX2 (AVX-256).

    it's obviously wrong - since ice lake has only one avx-512 block but two avx2 blocks, it's not much faster in avx-512 mode compared to avx2 mode

    the only mention of HEDT cpus at the page linked is "At a score of 4519, it beats a full 18-core Core i9-7980XE processor running in non-AVX mode". Since AVX-512 can process 16 32-bit numbers in a single operation, no wonder that a single avx-512 core matches 16 scalar cores

Log in

Don't have an account? Sign up now