CPU Tests: SPEC Performance

SPEC2017 is a series of standardized tests used to probe the overall performance between different systems, different architectures, different microarchitectures, and setups. The code has to be compiled, and then the results can be submitted to an online database for comparison. It covers a range of integer and floating point workloads, and can be very optimized for each CPU, so it is important to check how the benchmarks are being compiled and run.

For compilers, we use LLVM both for C/C++ and Fortran tests, and for Fortran we’re using the Flang compiler. The rationale of using LLVM over GCC is better cross-platform comparisons to platforms that have only have LLVM support and future articles where we’ll investigate this aspect more. We’re not considering closed-sourced compilers such as MSVC or ICC.

clang version 10.0.0
clang version 7.0.1 (ssh://git@github.com/flang-compiler/flang-driver.git

-Ofast -fomit-frame-pointer
-mfma -mavx -mavx2

Our compiler flags are straightforward, with basic –Ofast and relevant ISA switches to allow for AVX2 instructions. We decided to build our SPEC binaries on AVX2, which puts a limit on Haswell as how old we can go before the testing will fall over. This also means we don’t have AVX512 binaries, primarily because in order to get the best performance, the AVX-512 intrinsic should be packed by a proper expert, as with our AVX-512 benchmark. All of the major vendors, AMD, Intel, and Arm, all support the way in which we are testing SPEC.

To note, the requirements for the SPEC licence state that any benchmark results from SPEC have to be labeled ‘estimated’ until they are verified on the SPEC website as a meaningful representation of the expected performance. This is most often done by the big companies and OEMs to showcase performance to customers, however is quite over the top for what we do as reviewers.

SPEC2017 Rate-1 Estimated Total

In the single threaded test, the jump over the regular Zen 3 Ryzen mobile variant (5980HS) at the same power is quite substantial: +9.6% on integer performance and +14.1% on floating point. The move from DDR4 to DDR5 is quite substantial in that regard, and it’s seen in a lot of our upcoming benchmarks.

We didn’t see any change from 35 W to 45 W to 65 W in our AMD testing as the power consumption of the chip in single threaded workloads did not exceed 24 W, however we did see performance difference in Intel’s Alder Lake going from 45 W to 65 W, showcasing how much power the core can consume.

But if we compared that to Intel’s latest Alder Lake offerings, there’s a deficit in both categories – even though our lowest data here is at 45 W, we can see that the 45 W testing of the previous generation Intel also beats the 6900HS at SPECint (but AMD wins in SPECfp). This is something that carries through to multi-threaded performance.

SPEC2017 Rate-N Estimated Total

For Multi-Threaded performance, we only saw the slightest improvement from AMD moving up to 65 W, perhaps showcasing that the hardware is limited in other ways than just power and the uplift from DDR4 to DDR5. In any event, at 35 W, AMD still surpasses what the previous generation Intel i9-11980HK can provide at 65 W.

But if we compare it to Intel’s latest Alder Lake processors, featuring 6 performance cores and 8 efficiency cores, we now have 20 threads up against AMD’s 16 threads. If we compare 45 W to 45 W, Intel has a +14.0% lead in integer and a +13.3% lead in floating point, despite the 20% increase in threads. With Intel introducing this dual tier performance with hybrid SoCs, multi-threaded performance is going to be a combination of fast+slow and it all comes down to how the system can divide up the work.

Performance Per Watt CPU Tests: Office and Science


View All Comments

  • web2dot0 - Wednesday, March 2, 2022 - link

    PC fanbois would pretend M1 isn't in the convo.


    They will just tell you that nobody cares about performance per watt... because they said so.
  • Qasar - Wednesday, March 2, 2022 - link

    and the apple fanboys just keep talking about the m1 like its the best things since sliced bread, whats your point ?
    bottom line is this, IF you are already using the over priced apple eco system, then the m1 makes sense, if not, then there is no point to it. i only know 3 people that have either an ipad, or a mac based comp, the rest wont touch apple cause of the price, too expensive for what they would need it for, windows based products, suit their needs just fine.
  • schujj07 - Wednesday, March 2, 2022 - link

    For sure Apple is way too expensive for what you get. Working in industry I hate it when I have to give support to someone using a Mac because the VPN we send them doesn't work. Of the hundreds of VPNs I have sent to people, less then 10 have responded to me saying they need the one for Mac. Mac just doesn't play as nicely with the things we use in a lot of IT. Reply
  • Obiwanbilly - Friday, March 4, 2022 - link

    Oh, the VPN doesn’t work? I have an M1 MacBook Air and I use WireGuard for my VPN. It works perfect!

    Oh 🤔, do you mean “Legacy” VPNs, that are based on IPSEC or OpenVPN? You know, the one’s where Wikipedia says, “… are often complex to set up, disconnect easily (in the absence of further configuration), take substantial time to negotiate reconnections, may use outdated ciphers, and have relatively massive code bases of over 400,000 and 600,000 lines of code, respectively, which hinders debugging.“

    That one? 👆😱

    Hey bro, some of us Mac users are “Pros” too. Maybe you should stop using legacy VPN software and switch to something better? Instead of blaming Macs! 🤦‍♂️


  • BushLin - Friday, March 4, 2022 - link

    Try connecting to your employer's VPN, chances are that it's not wireguard. Reply
  • Obiwanbilly - Friday, March 4, 2022 - link

    Yep, it’s legacy. It needs to be upgraded. I can help, if you want. 😬

    I don’t know why you would continue to use OpenVPN or IPSec. I first used that tech like 14 yrs ago. Move on! 🤓

    Go research the benefits of using WireGuard. It’s a waaaay better solution. Dropped connections on IPSec or Open VPN take FOREVER to reconnect. You don’t need licenses, you can support unlimited users. Your limitation is the hardware you choose to host your WireGuard Server endpoint. WireGuard supports Windows and Android endpoint client devices. Oh yeah, and Mac OS too. 🥳

  • Dug - Friday, March 11, 2022 - link

    Then you have an outdated or crappy vpn. Every industry standard has a mac client that works fine. Even Azure has a mac configuration for mac vpn. Reply
  • wolfesteinabhi - Wednesday, March 2, 2022 - link

    it will be in convo when Apple starts selling Windows "PC" with M1 in it. till then its definitely not in Convo. Reply
  • tyger11 - Thursday, March 3, 2022 - link

    OS X is enough to keep me away from that hardware, so it doesn't matter. Reply
  • GeoffreyA - Thursday, March 3, 2022 - link

    Frieza's stronger than Goku right now--- Reply

Log in

Don't have an account? Sign up now