SPEC2006 & 2017: Industry Standard - ST Performance

One big talking point around the new Ryzen 3000 series is the new augmented single-threaded performance of the new Zen 2 core. In order to investigate the topic in a more controlled manner with better documented workloads, we’ve fallen back to the industry standard SPEC benchmark suite.

We’ll be investigating the previous generation SPEC CPU2006 test suite giving us some better context to past platforms, as well as introducing the new SPEC CPU2017 suite. We have to note that SPEC2006 has been deprecated in favour of 2017, and we must also mention that the scores posted today are noted as estimates as they’re not officially submitted to the SPEC organisation.

For SPEC2006, we’re still using the same setup as on our mobile suite, meaning all the C/C++ benchmarks, while for SPEC2017 I’ve also went ahead and prepared all the Fortran tests for a near complete suite for desktop systems. I say near complete as due to time constraints we’re running the suite via WSL on Windows. I’ve checked that there are no noticeable performance differences to native Linux (we’re also compiling statically), however one bug on WSL is that it has a fixed stack size so we’ll be missing 521.wrf_r from the SPECfp2017 collection.

In terms of compilers, I’ve opted to use LLVM both for C/C++ and Fortran tests. 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 8.0.0-svn350067-1~exp1+0~20181226174230.701~1.gbp6019f2 (trunk)
clang version 7.0.1 (ssh://git@github.com/flang-compiler/flang-driver.git 
  24bd54da5c41af04838bbe7b68f830840d47fc03)

-Ofast -fomit-frame-pointer
-march=x86-64
-mtune=core-avx2 
-mfma -mavx -mavx2

Our compiler flags are straightforward, with basic –Ofast and relevant ISA switches to allow for AVX2 instructions.

The Ryzen 3900X system was run in the same way as the rest of our article with DDR4-3200CL16, same as with the i9-9900K, whilst the Ryzen 2700X had DDR-2933 with similar CL16 16-16-16-38 timings.

SPECint2006 Speed Estimated Scores

In terms of the int2006 benchmarks, the improvements of the new Zen2 based Ryzen 3900X is quite even across the board when compared to the Zen+ based Ryzen 2700X. We do note however somewhat larger performance increases in 403.gcc and 483.xalancbmk – it’s not immediately clear as to why as the benchmarks don’t have one particular characteristic that would fit Zen2’s design improvements, however I suspect it’s linked to the larger L3 cache.

445.gobmk in particular is a branch-heavy workload, and the 35% increase in performance here would be better explained by Zen2’s new additional TAGE branch predictor which is able to reduce overall branch misses.

It’s also interesting that although Ryzen3900X posted worse memory latency results than the 2700X, it’s still able to outperform the latter in memory sensitive workloads such as 429.mcf, although the increases for 471.omnetpp is amongst the smallest in the suite.

However we still see that AMD has an overall larger disadvantage to Intel in these memory sensitive tests, as the 9900K has large advantages in 429.mcf, and posting a large lead in the very memory bandwidth intensive 462.libquantum, the two tests that put the most pressure on the caches and memory subsystem.

SPECfp2006(C/C++) Speed Estimated Scores

In the fp2006 benchmarks, we gain see some larger jumps on the part of the Ryzen 3900X, particularly in 482.sphinx3. These two tests along with 450.soplex are characterized by higher data cache misses, so Zen2’s 16MB L3 cache should definitely be part of the reason we see such larger jumps.

I found it interesting that we’re not seeing much improvements in 470.lbm even though this is a test that is data store heavy, so I would have expected Zen2’s additional store AGU to greatly benefit this workload. There must be some higher level memory limitations which is bottlenecking the test.

453.povray isn’t data heavy nor branch heavy, as it’s one of the more simple workloads in the suite. Here it’s mostly up to the execution backend throughput and the ability of the front-end to feed it fast enough that are the bottlenecks. So while the Ryzen 3900X provides a big boost over the 2700X, it’s still largely lagging behind the 9900K, a characteristic we’re also seeing in the similar execution bottlenecked 456.hmmer of the integer suite.

SPEC2006 Speed Estimated Total

Overall, the 3900X is 25% faster in the integer and floating point tests of the SPEC2006 suite, which corresponds to an 17% IPC increase, above AMD's officially published figures for IPC increases.

Moving on to the 2017 suite, we have to clarify that we’re using the Rate benchmark variations. The 2017 suite’s speed and rate benchmarks differ from each other in terms of workloads. The speed tests were designed for single-threaded testing and have large memory demands of up to 11GB, while the rate tests were meant for multi-process tests. We’re using the rate variations of the benchmarks because we don’t see any large differentiation between the two variations in terms of their characterisation and thus the performance scaling between the both should be extremely similar. On top of that, the rate benchmarks take up to 5x less time (+1 hour vs +6 hours), and we're able run them on more memory limited platforms (which we plan on to do in the future).

SPECint2017 Rate-1 Estimated Scores

In the int2017 suite, we’re seeing similar performance differences and improvements, although this time around there’s a few workloads that are a bit more limited in terms of their performance boosts on the new Ryzen 3900X.

Unfortunately I’m not quite as familiar with the exact characteristics of these tests as I am with the 2006 suite, so a more detailed analysis should follow in the next few months as we delve deeper into microarchitectural counters.

SPECfp2017 Rate-1 Estimated Scores

In the fp2017 suite, things are also quite even. Interesting enough here in particular AMD is able to leapfrog Intel’s 9900K in a lot more workloads, sometimes winning in terms of absolute performance and sometimes losing.

SPEC2017 Rate-1 Estimated Total

As for the overall performance scores, the new Ryzen 3900X improves by 23% over the 2700X. Although closing the gap greatly and completely, it’s just a hair's width shy of actually beating the 9900K’s absolute single-threaded performance.

SPEC2017 Rate-1 Estimated Performance Per GHz

Normalising the scores for frequency, we see that AMD has achieved something that the company hasn’t been able to claim in over 15 years: It has beat Intel in terms of overall IPC. Overall here, the IPC improvements over Zen+ are 15%, which is a bit lower than the 17% figure for SPEC2006.

We already know about Intel’s new upcoming Sunny Cove microarchitecture which should undoubtedly be able to regain the IPC crown with relative ease, but the question for Intel is if they’ll be able to still maintain the single-thread absolute performance crown and continue to see 5GHz or similar clock speeds with the new core design.

Test Bed and Setup Benchmarking Performance: Web Tests
POST A COMMENT

451 Comments

View All Comments

  • Irata - Monday, July 8, 2019 - link

    Thanks for your reply Ryan. I did not intend to be rude when saying "lazy" but rather show that I do not think this is something that was done by intent.

    Like I said - mention these things and it helps clear up misunderstandings.

    It is definitely very positive that you test the Ryzen CPU with the latest builds though.
    I also like that you mention if prices include an HSF or not, but it would have been nice to mention the price of HSF used for Intel systems (when not boxed), as e.g. the Thermalright True Copper is a rather expensive CPU cooler.

    I think you already addressed not using a faster nVME drive (a PCIe 4 version would have been ideal if available - this would also have given an indication of potentially increased system power use for the Ryzen with PCIe 4 drives) on Twitter.

    Those are little nitpicks, so not intended to be a criticism of the overall article. It is just that people tend to be rather sensitive when it comes to Intel vs. AMD CPU comparisons, given Intel's history of things they are willing to do to keep mind- and marketshare.
    Reply
  • Daeros - Monday, July 15, 2019 - link

    Whether or not it is intentional, AT has had an increasing Intel bias over the last several years. Watch to see how long it takes for an AMD article to get pushed down by rumors or vaporware from Intel or Nvidia. Reply
  • rarson - Monday, July 8, 2019 - link

    I think Ryan brings up several salient points, and whether or not you think that they did or did not have the time to do what you wanted (they were also a man down without Dr. Cuttress), the fact of the matter is that AMD dropped a bunch of CPUs and GPUs all at once and literally everyone was scrambling to do what they could in order to cover these launches.

    I don't think it's coincidence that even in the tech Youtube space, if you watch 10 different reviews you'll largely see 10 different testing methodologies and 10 (somewhat) different results. Every single reviewer I've talked to said that this was one of, if not the most, difficult launch windows they've ever dealt with. Additionally, launching on a weekend with all of its associated complications (not just on reviewers' ends, but partners as well) is a bitch, with everyone scrambling at the last minute on their days off getting in last-minute updates and whatnot.

    When AMD tells you at the last minute, "Oh, the brand new Windows 10 update includes something new" you don't necessarily have time to go back and redo all the benchmarks you had already done on the Intel platform.

    TL;DR while there may have been flaws in some of the testing, take the details with a grain of salt and compare them to the myriad of other reviews out there for a better overall picture if necessary.
    Reply
  • Irata - Tuesday, July 9, 2019 - link

    You are making a good point and unfortunately this was an - unfortunately - typical AMD CPU launch with things still being beta. I would assume testers are none too happy about having to re-do their tests.

    What I don't get from AMD (even if (and that's a capital IF) it's not their fault, it's their responsibility) is how they cannot see how this makes their products appear in a less favorable light. Let's say the buggy bios cost them 5%, the conclusion with a 5% better performance would have been even more in Ryzen 3000's favor.

    It's a bit like showing up to a job interview wearing the clothes you wore for the previous day's physical activity.
    Reply
  • Daeros - Monday, July 15, 2019 - link

    Lazy isn't in it. Intentionally misleading is more like it. On one page, where AMD wins more than it looses in the charts, out of 21 paragraphs, 2 had something positive to say about AMD or Ryzen 3k without following up with something along the lines of "but we know Intel's got new tech coming, too" Reply
  • Ryan Smith - Monday, July 8, 2019 - link

    To be sure, they're still valid. The patches for Fallout and ZombieLoad are not out yet (I only mention them because the vulnerabilities have already been announced). Reply
  • RSAUser - Monday, July 8, 2019 - link

    They've been out since 14 May, what are you talking about? Reply
  • djayjp - Monday, July 8, 2019 - link

    Don't forget RIDL Reply
  • Meteor2 - Sunday, July 14, 2019 - link

    RIDL and Zombieload are the same thing.

    Yes, the Intel CPUs should have been re-benchmarked on 1903, updated after 14 May when the OS-side fixes for the new MDS-class flaws were released. That's only fair and it's quite reasonable to expect that users will apply security updates, not leave their systems unpatched and vulnerable for perhaps a percent or two of performance.
    Reply
  • FireSnake - Monday, July 8, 2019 - link

    Ryan: how is this not explained in the article? I am reading this site for more then a decade and I trust you most. and I trust you will provide such information. I would expect, you update the article with this info. Reply

Log in

Don't have an account? Sign up now