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

  • t.s - Tuesday, January 30, 2024 - link

    well, I care.
  • goatfajitas - Tuesday, January 30, 2024 - link

    I do see your point. You can put in some high end cooling and take advantage of the speed. That still doesnt fix the fact that it runs extremely hot and power hungry compared to its competition.
  • ricebunny - Tuesday, January 30, 2024 - link

    Highly dependent on application. In games they pull around 125W, roughly the same as high end Zen 4 CPUs.
  • goatfajitas - Tuesday, January 30, 2024 - link

    "Highly dependent on application. In games they pull around 125W, roughly the same as high end Zen 4 CPUs."

    Agreed on some setups. I have seen some equivalently equipped SFF/Tiny desktop setups (with obviously limited thermals) and Intel drops off alot earlier. The point being it runs too hot.
  • WaffleTech - Sunday, February 4, 2024 - link

    "roughly the same"

    ComputerBase measured this over more than a dozen games and on average it's 149W for the 14900K and 72W for the 7950X3D, which is less than half. Even the Non-X3D 7950X with 105W uses almost 30% less in games.
  • Thunder 57 - Tuesday, January 30, 2024 - link

    It's the enormous heat output that is the problem, not the electric bill.
  • yankeeDDL - Thursday, February 1, 2024 - link

    "no issue"? Are you joking?
    Cost aside, to dissipate 400+W you need massive cooling, and the MoBo needs to be way over-designed.
    Intel peaks at 4x the power of Ryzen to, maybe, marginally beat it in few benchmarks. That's not irrelevant at all: this kind of delta in efficiency applies also on mobile (albeit not so extreme): that translates in battery life (and slower performance, as on laptops you can't peak at those levels).
  • is4u2p - Wednesday, January 31, 2024 - link

    Was this in the 12th gen mobile CPUs?
  • lakedude - Friday, February 9, 2024 - link

    "Peak power is an irrelevant metric." maybe to you.

    Going AMD this time allowed me to save money on the cooler and the power supply. Gonna save money on electricity as well.

    I hate the way Intel plays games with their TDP these days. Don't say something is 125w TDP if it can pull over 300w. Call it a 300w chip that can be underclocked or power capped to 125w. Or have it only pull 125w out of the box rather than leave it up to the user to fiddle around with turning it down.
  • wwenze - Monday, February 5, 2024 - link

    Well, Intel definitely has the advantage when it comes to "not dying when pushed hard" or even "not dying when running at the actual stated specs" but outdate efficiency/IPC, while AMD is the opposite on both fronts.

    So Intel put the pieces together and said "what if we clock it until 250W turbo TDP while the competition only has 100W at that price point"

Log in

Don't have an account? Sign up now