CPU Benchmark Performance: AI and Inferencing

As technology progresses at a breakneck pace, so too do the demands of modern applications and workloads. With artificial intelligence (AI) and machine learning (ML) becoming increasingly intertwined with our daily computational tasks, it's paramount that our reviews evolve in tandem. Recognizing this, we have AI and inferencing benchmarks in our CPU test suite for 2024. 

Traditionally, CPU benchmarks have focused on various tasks, from arithmetic calculations to multimedia processing. However, with AI algorithms now driving features within some applications, from voice recognition to real-time data analysis, it's crucial to understand how modern processors handle these specific workloads. This is where our newly incorporated benchmarks come into play.

As chip makers such as AMD with Ryzen AI and Intel with their Meteor Lake mobile platform feature AI-driven hardware within the silicon, it seems in 2024, and we're going to see many applications using AI-based technologies coming to market.

We are using DDR5-5200 memory as per the JEDEC specifications on the Ryzen 7 8700G and Ryzen 5 8600G, as well as DDR4-3200 on the Ryzen 7 5700G and Ryzen 5 5600G. The same methodology is also used for the AMD Ryzen 7000 series and Intel's 14th, 13th, and 12th Gen processors. Below are the settings we have used for each platform:

  • DDR5-5200 CL44 - Ryzen 8000G
  • DDR4-3200 CL22 - Ryzen 5000G
  • DDR5-5600B CL46 - Intel 14th & 13th Gen
  • DDR5-5200 CL44 - Ryzen 7000
  • DDR5-4800 (B) CL40 - Intel 12th Gen

(6-2) DeepSpeech 0.6: Acceleration CPU

(6-3) TensorFlow 2.12: VGG-16, Batch Size 16 (CPU)

(6-3b) TensorFlow 2.12: VGG-16, Batch Size 64 (CPU)

(6-3d) TensorFlow 2.12: GoogLeNet, Batch Size 16 (CPU)

(6-3e) TensorFlow 2.12: GoogLeNet, Batch Size 64 (CPU)

(6-3f) TensorFlow 2.12: GoogLeNet, Batch Size 256 (CPU)

(6-4) UL Procyon Windows AI Inference: MobileNet V3 (float32)

(6-4b) UL Procyon Windows AI Inference: ResNet 50 (float32)

(6-4c) UL Procyon Windows AI Inference: Inception V4 (float32)

(6-1) ONNX Runtime 1.14: CaffeNet 12-int8 (CPU Only)

(6-1b) ONNX Runtime 1.14: CaffeNet 12-int8 (CPU Only)

A major focal point of AMD's Ryzen 8000G series is the inclusion of the Xilinx-based Ryzen AI NPU. While AI benchmarks and those measuring capabilities using large language models (LLMs) are thin off the ground, none of our benchmarks utilize the NPU itself. Much of the Ryzen AI NPU is based and, as such, is focused on enabling software features such as those generative AI capabilities within Microsoft Studio Effects and software such as Adobe and Davinci.

In ONNX Runtime using the utilized INT8 model, we can see that the Ryzen 7 8700G and Ryzen 5 8600G don't offer world-beating AI performance, but we intend to investigate this more deeply.

Using the latest firmware, which removes the STAPM limitations, we can see that the Ryzen 5 8600G shows the most gains, especially in DeepSpeech 0.6, where we saw a 12% bump in performance. The Ryzen 7 8700G also posted some very impressive gains in the UL Procyon Windows AI Inferencing benchmark, with a 34% jump in performance in our charts, but this could be a case where it underperformed in the MobileNet V3 test in the first place.

CPU Benchmark Performance: Science And Simulation iGPU Gaming Performance: 720p And Lower
Comments Locked

111 Comments

View All Comments

  • James5mith - Monday, January 29, 2024 - link

    "These limitations primarily come in highly intensive multi-threaded workloads such as rendering or encoding, where the performance of processors such as the Ryzen 7000 desktop series, but the key point is that these APUs aren't inherently designed for these tasks in mind, and users looking for more CPU grunt are almost certainly likely to opt for a higher grade processor with faster cores, more cores, and more threads. "

    That is a) a massive run-on sentence, and b) doesn't make much actual sense. For example:

    "where the performance of processors such as the Ryzen 7000 desktop series"

    Where the performance does what? Or is what? There is no coherent thought in that comma delimited side note.
  • GeoffreyA - Tuesday, January 30, 2024 - link

    The performance of the 7000 series excels at multithreading.
  • yankeeDDL - Monday, January 29, 2024 - link

    I wish there were some more remarks vs Intel's offering.
    CPU-wise Ryzen is more efficient, generally speaking. The performance seems same or slightly lower when compared with CPU that burn 400W to reach crazy boost rate. Still, that's my view.
    Comparing it only against Ryzen 5*** seems a bit limited, no?
  • meacupla - Monday, January 29, 2024 - link

    Ryzen 5000G is the primary competitor for Ryzen 8000G.
    It's next closest competition comes from the mobile segment in the form of mini-PCs.
    Intel hasn't offered an APU for its desktop socket in ages.
  • yankeeDDL - Thursday, February 1, 2024 - link

    Not correct.
    Ryzen 8000G is the only Zen4 desktop CPU on TSMC 4nm process (Ryzen 7000 is also Zen4, but on 5nm).
    So - iGPU aside - I expect the 8000G to be more efficient than the 7000. Hence, I am curious as to how it would perform against Intel and also against the 7000.
  • Grapple - Monday, January 29, 2024 - link

    Comparing against Intel’s 65W T-series processors would have made this article much more interesting.
  • FWhitTrampoline - Monday, January 29, 2024 - link

    T series is 35W and mostly for thin client Mini Desktop Business PCs but at least the T series are Socket Packaged! And Intel Made a Big mistake not releasing at least a 65W Socket Packaged Meteor Lake SKU. And Intel is very well capable of offering a socket packaged Mobile Processor because I'm still using my HP ProBook 4540s laptop with a Socket Packaged Ivy Bridge generation 3632QM processor and that laptop can get a processor update of MB replacement and reuse the same processor!
  • meacupla - Monday, January 29, 2024 - link

    I disagree. But only on processor choice.
    Ryzen 7840HS, Intel i7-13700H and i7-1370P would be my choice.
  • TheinsanegamerN - Tuesday, January 30, 2024 - link

    Given that this is a desktop chip, it'd make more sense to compare it to other desktop chips.

    Otherwise, throw an M3 and a nvidia t100 car processor while youre at it.
  • meacupla - Tuesday, January 30, 2024 - link

    Hahaha, nope.
    8000G are mobile chips using a desktop socket.

Log in

Don't have an account? Sign up now