Test Bed and Setup

As per our processor testing policy, we take a premium category motherboard suitable for the socket, and equip the system with a suitable amount of memory running at the manufacturer's maximum supported frequency. This is also typically run at JEDEC subtimings where possible. It is noted that some users are not keen on this policy, stating that sometimes the maximum supported frequency is quite low, or faster memory is available at a similar price, or that the JEDEC speeds can be prohibitive for performance.

While these comments make sense, ultimately very few users apply memory profiles (either XMP or other) as they require interaction with the BIOS, and most users will fall back on JEDEC-supported speeds - this includes home users as well as industry who might want to shave off a cent or two from the cost or stay within the margins set by the manufacturer. Where possible, we will extend out testing to include faster memory modules either at the same time as the review or a later date.

The Current CPU Test Suite

For our AMD Ryzen 9 7950X and Ryzen 5 7600X testing, we are using the following test system:

AMD Ryzen 7000 Series System (DDR5)
CPU Ryzen 9 7950X ($699)
16 Cores, 32 Threads
170 W TDP

Ryzen 5 7600X ($299)
6 Cores, 12 Threads
105 W TDP
Motherboard GIGABYTE X670E Aorus Master (BIOS 813b)
Memory G.Skill Trident Z5 Neo
2x16 GB
DDR5-5200 CL44
Cooling EK-AIO Elite 360 D-RGB 360 mm AIO
Storage Crucial MX300 1TB
Power Supply Corsair HX850
GPUs NVIDIA GeForce RTX 2080 Ti, Driver 496.49
Operating Systems Windows 11 21H1

As we are in a transitional period between our current CPU 2021 suite and data, and optimizing our CPU 2023 suite with different data comparisons required, we have included a varied selection of benchmarks for this review. This ranges from our traditional un-updatable Google Octane 2.0 web test, through a variety of rendering benchmarks such as CineBench R23 and Blender, to encoding, and all the way to our more scientific-related tests.

With our processor reviews, especially on a new generational product such as AMD's Ryzen 9 7950X, we also include SPEC2017 data to account for any increases (or decreases) to generational single-threaded and multi-threaded performance. It should be noted that due to the terms of the SPEC license because our benchmark results are not vetted directly by the SPEC consortium, we have to label them as ‘estimated’. The benchmark is still run and we get results out, but those results have to have the ‘estimated’ label.

Moving Foward to Our CPU 2023 Suite: What to Expect

Looking ahead to our updated CPU 2023 suite, we've updated some of our existing benchmarks to the latest and current versions (as of Sept 22) such as Blender 3.3. In terms of benchmarks from our CPU 2021 suite, we've included benchmarks such as Dwarf Fortress, Factorio, and Dr. Ian Cutress's 3DPMv2 and Crysis CPU benchmarks. 

We've also added some completely new benchmarks and workloads to our suite, including an update to Blender (v3.3), C-Ray 1.1 rendering, as well as more scientific-based workloads such as SciMark 2.0 and Primesieve 1.9.0. We have also decided to add UL's latest Procyon suite which measures overall system performance when doing tasks such as office-based tasks, as well as video, and photo editing.

As it stands, we have also updated our pool of games going forward into 2023 and beyond, including the latest F1 2022 racing game, the CPU-intensive Total War Warhammer 3 real-time strategy, and the popular Hitman 3 assassin-based title.

Our aim is to provide varying levels of data points across a variety of different workloads, instruction sets, and tasks. Going forward, we will keep our CPU 2023 suite updated as frequently as possible, and when we have a consistent and suitable number of data points, it will feature on our Bench database as we continue testing new and older CPUs for varying data points.

Some of these new benchmarks will make an appearance in this review, while others won't. Our aim is to assess and subjugate our way through whatever CPUs we have on hand to add vital data points. However, some parts of our CPU 2023 suite are still under testing and it should make a full debut in our next CPU review.

Zen 4 Execution Pipeline: Familiar Pipes With More Caching Core-to-Core Latency
POST A COMMENT

205 Comments

View All Comments

  • Iketh - Thursday, October 6, 2022 - link

    why are you giving so much credit to ddr5? moving to new memory has always given very small gains (if any) in the beginning

    tjunction is an arbitrary number set by AMD, so using that as an argument is irrational
    Reply
  • xol - Tuesday, September 27, 2022 - link

    ..but my main criticism was of the article - eg phrases like " increase the overall TDP ... without too much penalty" doesn't really make any sense - increase TDP is the penalty

    But much of the article is written as if letting TDP go *much* higher is some sort of gift from AMD -eg the examples I gave

    The article is full of nothin-burgers like this statement :
    " We feel that the higher all-core frequencies under maximum load, 95°C is a sufficient level of heat for what is on offer when it comes to overall performance"
    Reply
  • kwrzesien - Monday, September 26, 2022 - link

    Whomever was the last to edit the front page needs to disable the trackpad and clean their mouse ball! 🤣 Reply
  • Threska - Monday, September 26, 2022 - link

    "But now with AMD’s modern RDNA 2 graphics architecture and TSMC fabrication process, AMD has finally seen the (ray traced) light, and is building a small GPU block into the IOD to offer integrated graphics throughout the Ryzen 7000 product stack."

    I see things like SAM and HSA being a future trend.

    https://www.electronicdesign.com/technologies/micr...
    Reply
  • erotomania - Wednesday, September 28, 2022 - link

    Yes, AMD thought so too, in 2012...

    https://www.tomshardware.com/reviews/fusion-hsa-op...

    and in 2014 here at AT...

    https://www.anandtech.com/show/7677/amd-kaveri-rev...

    Hopefully this time!
    Reply
  • nandnandnand - Monday, September 26, 2022 - link

    It seems that going up by 1 GHz didn't help it that much in gaming benchmarks.

    Meanwhile, the 65W results show that any Zen 4 and later APUs are going to be absurdly powerful. Especially Dragon Range.
    Reply
  • Josh128 - Monday, September 26, 2022 - link

    Any way you guys can add the single core ECO mode results to the conclusion page or to the R23 results on its respective page? Reply
  • donquixote42 - Monday, September 26, 2022 - link

    Single threaded workload would not use more than 65W anyway. So performance should be the same in ECO and non-ECO mode. Reply
  • Josh128 - Monday, September 26, 2022 - link

    Still using a 2080Ti for the games testing is not good. Most certainly many of these results are GPU bound. Reply
  • snowdrop - Monday, September 26, 2022 - link

    No power consumption numbers? Will the article be updated with these when they're ready? Reply

Log in

Don't have an account? Sign up now