Leaked Lunar Lake's Ultra 7 chip debuts in Geekbench database — impressive performance from low-power chips

Intel Lunar Lake
(Image credit: Tom's Hardware)

Two new Intel Lunar Lake results have cropped up on the Geekbench browser (as spotted by @BenchLeaks on X), featuring the upcoming Core Ultra 7 268V with four non-Hyper-Threading P-cores and four E-cores. If we take these results at face value, the new CPU's performance in Geekbench is a mixed bag. It boasts superior single-core performance but not enough multi-threaded juice to defeat Intel's Meteor Lake Core Ultra 7 parts. That's a solid result, though, as Lunar Lake is designed for lower-TDP thin-and-lights and compact notebooks than the higher-tier Meteor Lake processors.

Two listings featuring the Core Ultra 7 268V have appeared on Geekbench; the first reveals a score of 2,713 points and 10,036 points for the Lunar Lake chip in the single and multi-core benchmarks. The other shows a slightly better single core but a slightly inferior multi-threaded score of 2,739 and 9,907 points, respectively. 

For simplicity's sake, we will use the average of both scores to better compare the chip to other chips further down the article. Our simple calculation results in scores of 2,726 and 9,972 for the Lunar Lake part, in single and multi-threaded workloads.

Swipe to scroll horizontally
Lunar Lake vs Meteor Lake - Geekbench
CPU:Single-CoreMulti-Core
Lunar Lake Ultra 7 268V (Average)2,7269,972
Meteor Lake Core Ultra 7 155H2,35611,926

Compared to the Ultra 7 268V's outgoing Meteor Lake-based predecessors, its performance in Geekbench is mixed. The chip is 15% faster in single-core performance compared to the Core Ultra 7 155H, but it is 19% slower in Geekbench's multi-core benchmark. Our Ultra 7 155H result was taken from an HP Spectre laptop sporting a single-core score of 2,356 and an 11,926 multi-core score. There are certainly inferior scores in the Geekbench browser, but we found this one in particular to be a good balance between the fastest and slowest Core Ultra 7 155H results in the online results.

It will be interesting to see if this behavior exists in subsequent benchmarks and with production-ready devices packing Intel's Lunar Lake chips. One of the biggest changes Intel has made is removing its Hyper-Threading technology from its Lunar Lake chips in an effort to boost power efficiency. In addition, the Ultra 7 268V only has four E-cores, while the Core Ultra 7 155H has eight. 

Intel will rely entirely on its substantially faster E-cores in Lunar Lake to combat the architecture's E-core deficit compared to Meteor Lake and even Raptor Lake / Alder Lake. Intel's upcoming Lunar Lake chips come with upgraded E-cores based on the Skymont architecture, sporting up to a 68% IPC gain compared to Meteor Lake's Crestmont LPE cores.

Again, it'll be very interesting to see if Intel's new E-core architecture can rectify Lunar Lake's disadvantage in core count compared to Meteor Lake. However, we must note that a major focus of Intel's Lunar Lake is reduced power consumption for thin-and-lights, so it still impresses in the multi-threaded tests if you take into account the comparative thread counts and target TDPs (reported default of 17W for most models, with 30W peak). In contrast, Meteor Lake processors are designed to scale to higher TDPs. Intel has yet to introduce its successor for the higher-powered 45W/115W Meteor Lake chips.

In conclusion, Geekbench is only one benchmark (and this is a leak), so we'll have to see how Lunar Lake performs in other benchmarks/apps before we get a clearer picture.

Aaron Klotz
Contributing Writer

Aaron Klotz is a contributing writer for Tom’s Hardware, covering news related to computer hardware such as CPUs, and graphics cards.

  • abufrejoval
    Why do you judge them impressive?

    They seem to be really along the same lines as the competition on single, less impressive on multi, but that's because they are designed for the ultra low power classs.
    Reply
  • KnightShadey
    abufrejoval said:
    Why do you judge them impressive?

    Agreed, even the Zen5 results, were just.... "yep seems about right" .. and the march goes on.
    Reply
  • thestryker
    abufrejoval said:
    Why do you judge them impressive?

    They seem to be really along the same lines as the competition on single, less impressive on multi, but that's because they are designed for the ultra low power classs.
    Disagree completely on the multithreaded not being impressive as it's much more impressive than single threaded. We're talking 4P/4E/8T, E-cores that don't have L3 access, P-cores that don't have HT about matching the performance of a 13400 (ADL specs even in the RPL version) at 6P/4E/16T. This configuration matching a 65W SKU in a package designed to not go past 30W with a steady of 17W (assuming the leaks are accurate) is extremely impressive.
    Reply
  • JRStern
    I totally welcome good performance on a low-power chips.
    Not even thin-and-light, I want a fanless workstation.
    Real performance now I worry about only in the cloud on much bigger machines than I'm likely to run IRL.
    Reply
  • JamesJones44
    abufrejoval said:
    Why do you judge them impressive?
    Just playing devils advocate here, but if we are talking about an 8 core part (268V is supposed to be 4 P/4 E) vs a 16 core part of similar product level coming in semi close in multi score, that's at least notable. Not Earth shattering, but interesting.
    Reply
  • JTWrenn
    abufrejoval said:
    Why do you judge them impressive?

    They seem to be really along the same lines as the competition on single, less impressive on multi, but that's because they are designed for the ultra low power classs.
    The article explains why they call them impressive in the last paragraph.

    "so it still impresses in the multi-threaded tests if you take into account the comparative thread counts and target TDPs (reported default of 17W for most models, with 30W peak). In contrast, Meteor Lake processors are designed to scale to higher TDPs. Intel has yet to introduce its successor for the higher-powered 45W/115W Meteor Lake chips."
    Reply
  • KnightShadey
    OK, so I guess potato / tomato regarding 'impressive'.

    But even on multi-threaded performance, is it "impressive" or "not" to y'all?
    Article says one thing, and most here are arguing the opposite.

    Quick look on NBC multi- core performance of 9,972 is way below an equally low TDP QualcommX1P-64 which scores 13,278 running 10 cores @ 3.4Ghz max, vs the LNL 4/4 @ 3.9/4.7/49Ghz. And Single & multi-core performance is right near the similar TDP 15W Ryzen 8840U's score for something with similar TDP but far more threads. Both below the Z1 extreme in MC.


    It's definitely 'good' for efficiency & power from 4+4 cores, but is it 'impressive' , or.... what you would expect from a node update, freq/memory bump, and cutting down extra cores/multi-threaded performance to focus on TDP and single threaded performance ala Apple/Qc?

    All seems pretty inline to me, and it's only impressive versus particularly bad previous generation efficiency. However, whether TDP or cores, there's existing product all around it, above and below and beside. 🤔

    By impressive I was expecting Apple intel -> M1 change in single core - efficiency. Especially after all the wild early claims from Qualcomm, AMD, and intel. These seem like more down to earth improvements.

    BTW, I wonder about that 155H result in the article, seems like a throttled version, 155U or 165 U would've been a better choice, especially if TDP is the selling point of 'impressive' in this case. 🤷🏻‍♂️
    Reply
  • Alvar "Miles" Udell
    Beats the heck out my AMD 4700U (8C/8T) Lenovo laptop

    1500 - Single-Core Score
    5051 - Multi-Core Score


    https://browser.geekbench.com/v6/cpu/6685914
    Reply
  • abufrejoval
    JRStern said:
    I totally welcome good performance on a low-power chips.
    Not even thin-and-light, I want a fanless workstation.
    Real performance now I worry about only in the cloud on much bigger machines than I'm likely to run IRL.
    I understand the desire, but I've learned that some of the things I want can't be had.
    Among them is workstation performance on a phone power budget, which would make fanless a snap.

    And in the mean-time I've settled for unoticeable noise.

    That's worked out pretty well with clunky and noisy HDDs transitioning to silent NVMes, which allowed for big heat sinks and large slow moving fans to fill the tower chassis I don't keep next to my ear.

    Of course an 800 Watt GPU can't really keep silent, nor can CPU eating 300 Watts. But systems designed to manage such peak loads, can be rather unnoticable when running below maximum.

    Which is why I try to avoid running at peak, when the results aren't actually all that critical or the extra performance doesn't deliver real benefits (e.g. frame rates above monitor capabilities).

    That's often not as trivial to control as it should be, perhaps some type of decibel regulator should be integrated into PCs on top of variable TDP. But when the annoyance level is high enough, I usually find a solution.
    Reply
  • abufrejoval
    KnightShadey said:
    By impressive I was expecting Apple intel -> M1 change in single core - efficiency. Especially after all the wild early claims from Qualcomm, AMD, and intel. These seem like more down to earth improvements.
    We tend to overlook that that Intel to M1 gain came at a somewhat singular moment, when x86 evolution had basically stopped going into new silicon because Intel had fab troubles and AMD was still going through the Zen redesign.

    There simply isn't as much all-around magic in the x86 to ARM transition as people infer from that one point in time.

    And there won't be, if the entire focus remains on top Geekbench CPU scores, because there is no way to raise those except with either lots of transistors or Wattage or both, while efficiency suffers.

    All of these vendors tried to sidestep the limits of CPU performance scaling by telling us how much faster these devices would be running AIs on NPUs intead of CPUs or GPUs. But that requires that actually being thing and is now under the active threat of paradigm shifts in LLM hardware implementations (e.g. reduction of MatMul effort in transformers).
    Reply