2025-02-25

AMD Ryzen 9000 vs. Intel Core Ultra Arrow Lake On Linux For Q1-2025 In ~400 Benchmarks

For those wondering how the latest AMD Ryzen 9000 “Zen 5” series and Intel Core Ultra Series 2 “Arrow Lake” desktop processors are battling it out on Linux, here are some fresh benchmarks on Ubuntu Linux with the latest software updates as well as the newest system BIOS updates for a fresh, all-new look at these Intel Core Ultra and AMD Ryzen desktop CPUs on Linux.

Ahead of upcoming additional desktop CPU launches, I’ve been re-testing all of the AMD Ryzen 9000 series and Intel Core Ultra Series 2 desktop CPUs I have available. This incorporates all of the recent Linux software updates since the AMD Zen 5 and Intel Arrow Lake launches a few months ago as well as the new system BIOS and some new/updated workloads/benchmarks. An up-to-date look at the Intel and AMD desktop performance on Linux for mid-Q1 2025.

The current generation processors tested for this article based on the CPUs I had available included:

- Core Ultra 5 245K
- Core Ultra 9 285K
- Ryzen 5 9600X
- Ryzen 7 9700X
- Ryzen 7 9800X3D
- Ryzen 9 9900X
- Ryzen 9 9950X

Of course, soon it will be interesting to look at the upcoming Ryzen X3D parts and more on Linux.

Intel Core Ultra, AMD Ryzen 9000 - Linux 2025

All of these benchmarks were done on Ubuntu 24.10 compared to my original launch day testing on Ubuntu 24.04 LTS. With Ubuntu 24.10 comes the GCC 14.2 compiler with Znver5 support and other improvements. The Linux 6.13 stable kernel was used as the newest stable kernel series at the time of testing. Corsair MP700 PRO PCIe 5.0 NVMe 2TB storage and 2 x 16GB DDR5-6000 memory was used for testing along with an AMD Radeon RX 7900 XTX graphics card and liquid AIO cooling for the CPUs. Around 400 benchmarks were used ranging from common desktop applications like Google Chrome and Mozilla Firefox web browsers to image editing, audio/video encoding to various Linux server workloads and then also some graphics/gaming and a variety of other tested use-cases.

Selenium benchmark with settings of Benchmark: Speedometer 3.0, Browser: Firefox. Ryzen 7 9800X3D was the fastest.

Selenium benchmark with settings of Benchmark: Speedometer 3.0, Browser: Google Chrome. Ryzen 9 9950X was the fastest.

With the Speedometer 3.0 benchmark the AMD Zen 5 CPUs all delivered much better performance than the tested Intel Arrow Lake CPUs on Firefox 135 and Chrome 133. But when it came to the performance-per-Watt with Speedometer 3.0 is where Arrow Lake came out ahead of these Zen 5 desktop CPUs.

Selenium benchmark with settings of Benchmark: Jetstream 2, Browser: Firefox. Ryzen 9 9950X was the fastest.

Selenium benchmark with settings of Benchmark: Jetstream 2, Browser: Google Chrome. Ryzen 9 9950X was the fastest.

With the Jetstream 2 browser benchmark in Firefox and Chrome on Linux, the Arrow Lake and Zen 5 performance was much tighter while on a performance-per-Watt basis the Intel Core Ultra Series 2 was the leader here.

Selenium benchmark with settings of Benchmark: PSPDFKit WASM, Browser: Firefox. Ryzen 9 9950X was the fastest.

Selenium benchmark with settings of Benchmark: PSPDFKit WASM, Browser: Google Chrome. Ryzen 9 9950X was the fastest.

Selenium benchmark with settings of Benchmark: WASM collisionDetection, Browser: Firefox. Core Ultra 9 285K was the fastest.

Selenium benchmark with settings of Benchmark: WASM collisionDetection, Browser: Google Chrome. Core Ultra 9 285K was the fastest.

Selenium benchmark with settings of Benchmark: WASM imageConvolute, Browser: Firefox. Core Ultra 9 285K was the fastest.

Selenium benchmark with settings of Benchmark: WASM imageConvolute, Browser: Google Chrome. Core Ultra 9 285K was the fastest.

Selenium benchmark with settings of Benchmark: Kraken, Browser: Firefox. Core Ultra 9 285K was the fastest.

Selenium benchmark with settings of Benchmark: Kraken, Browser: Google Chrome. Ryzen 9 9950X was the fastest.

With some of the WebAssembly (WASM) browser benchmarks the Intel Arrow Lake CPUs were able to come out in front but AMD Zen 5 remained fiercely competitive.

Timed Linux Kernel Compilation benchmark with settings of Build: defconfig. Ryzen 9 9950X was the fastest.

Moving to multi-threaded code compilation is where the Ryzen 9 9950X and other Zen 5 CPUs tend to shine. The 16-core / 32-thread Ryzen 9 9950X continues outperforming the 24-core Core Ultra 9 285K in code compilation as well as consuming less power than that flagship Arrow Lake desktop processor.

Timed Linux Kernel Compilation benchmark with settings of Build: allmodconfig. Ryzen 9 9950X was the fastest.

Timed LLVM Compilation benchmark with settings of Build System: Ninja. Ryzen 9 9950X was the fastest.

Timed Gem5 Compilation benchmark with settings of Time To Compile. Ryzen 9 9950X was the fastest.

Timed Node.js Compilation benchmark with settings of Time To Compile. Ryzen 9 9950X was the fastest.

Timed Eigen Compilation benchmark with settings of Time To Compile. Ryzen 9 9950X was the fastest.

Timed Godot Game Engine Compilation benchmark with settings of Time To Compile. Ryzen 9 9950X was the fastest.

The AMD Ryzen 9 9950X remains a great processor for code compilation workloads while in a few weeks will be interesting to see how well the Ryzen 9 9950X3D does here. The Ryzen 9 9950X was consistently consuming less power than the Core Ultra 9 285K during the code compilation benchmarks.

7-Zip Compression benchmark with settings of Test: Decompression Rating. Ryzen 9 9950X was the fastest.

For many of these benchmarks the results in relative terms aren’t too different from the prior Arrow Lake and Zen 5 benchmarks.

Zstd Compression benchmark with settings of Compression Level: 19, Compression Speed. Ryzen 7 9800X3D was the fastest.

Stargate Digital Audio Workstation benchmark with settings of Sample Rate: 192000, Buffer Size: 512. Ryzen 9 9950X was the fastest.

Stargate Digital Audio Workstation benchmark with settings of Sample Rate: 192000, Buffer Size: 1024. Ryzen 9 9950X was the fastest.

In some cases the performance of all the tested CPUs are better than before thanks to the newer compiler, the latest Linux 6.13 kernel, and other software upgrades in recent months as well as new BIOS updates for both the AMD and Intel desktops.

Blender benchmark with settings of Blend File: BMW27, Compute: CPU-Only. Ryzen 9 9950X was the fastest.

Blender benchmark with settings of Blend File: Classroom, Compute: CPU-Only. Ryzen 9 9950X was the fastest.

Blender benchmark with settings of Blend File: Pabellon Barcelona, Compute: CPU-Only. Ryzen 9 9950X was the fastest.

Blender benchmark with settings of Blend File: Barbershop, Compute: CPU-Only. Ryzen 9 9950X was the fastest.

The Ryzen 9 9950X continued outpacing the Core Ultra 9 285K with the Blender 4.3 modeling software.

C-Ray benchmark with settings of Resolution: 4K, Rays Per Pixel: 16. Core Ultra 9 285K was the fastest.

POV-Ray benchmark with settings of Trace Time. Core Ultra 9 285K was the fastest.

All of the tested CPUs were running stable with Ubuntu 24.10 + Linux 6.13.

LuxCoreRender benchmark with settings of Scene: DLSC, Acceleration: CPU. Core Ultra 9 285K was the fastest.

LuxCoreRender benchmark with settings of Scene: Rainbow Colors and Prism, Acceleration: CPU. Ryzen 9 9950X was the fastest.

LuxCoreRender benchmark with settings of Scene: LuxCore Benchmark, Acceleration: CPU. Ryzen 9 9950X was the fastest.

LuxCoreRender benchmark with settings of Scene: Orange Juice, Acceleration: CPU. Ryzen 9 9950X was the fastest.

Appleseed benchmark with settings of Scene: Emily. Ryzen 9 9950X was the fastest.

Appleseed benchmark with settings of Scene: Disney Material. Ryzen 9 9950X was the fastest.

Appleseed benchmark with settings of Scene: Material Tester. Ryzen 9 9950X was the fastest.

Chaos Group V-RAY benchmark with settings of Mode: CPU. Ryzen 9 9950X was the fastest.

IndigoBench benchmark with settings of Acceleration: CPU, Scene: Supercar. Ryzen 9 9950X was the fastest.

IndigoBench benchmark with settings of Acceleration: CPU, Scene: Bedroom. Ryzen 9 9950X was the fastest.

The AMD Ryzen 9 9950X continued with a very strong showing across a range of open-source and proprietary rendering applications.

OSPRay benchmark with settings of Benchmark: gravity_spheres_volume/dim_512/ao/real_time. Ryzen 9 9950X was the fastest.

OSPRay benchmark with settings of Benchmark: gravity_spheres_volume/dim_512/scivis/real_time. Ryzen 9 9950X was the fastest.

OSPRay benchmark with settings of Benchmark: particle_volume/ao/real_time. Ryzen 9 9950X was the fastest.

OSPRay benchmark with settings of Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time. Ryzen 9 9950X was the fastest.

While OSPRay is an Intel oneAPI software project, it absolutely dominates with AMD Zen 5 processors thanks to the AVX-512 support compared to AVX2-only with Arrow Lake.

OSPRay Studio benchmark with settings of Camera: 1, Resolution: 1080p, Samples Per Pixel: 1, Renderer: Path Tracer, Acceleration: CPU. Ryzen 9 9950X was the fastest.

OSPRay Studio benchmark with settings of Camera: 3, Resolution: 1080p, Samples Per Pixel: 1, Renderer: Path Tracer, Acceleration: CPU. Ryzen 9 9950X was the fastest.

Embree benchmark with settings of Binary: Pathtracer ISPC, Model: Asian Dragon. Ryzen 9 9950X was the fastest.

Embree benchmark with settings of Binary: Pathtracer ISPC, Model: Crown. Ryzen 9 9950X was the fastest.

Intel Open Image Denoise benchmark with settings of Run: RT.ldr_alb_nrm.3840x2160, Device: CPU-Only. Ryzen 9 9950X was the fastest.

The AMD Ryzen 9000 series processors dominated across the range of Intel oneAPI software packages thanks to its AVX-512 support.

Coremark benchmark with settings of CoreMark Size 666, Iterations Per Second. Ryzen 9 9950X was the fastest.

OpenFOAM benchmark with settings of Input: drivaerFastback, Small Mesh Size, Execution Time. Core Ultra 9 285K was the fastest.

The Core Ultra 9 285K does well with OpenFOAM thanks to its 24 physical cores.

OpenRadioss benchmark with settings of Model: Bird Strike on Windshield. Ryzen 9 9950X was the fastest.

OpenRadioss benchmark with settings of Model: Rubber O-Ring Seal Installation. Ryzen 9 9950X was the fastest.

OpenRadioss benchmark with settings of Model: Cell Phone Drop Test. Core Ultra 9 285K was the fastest.

OpenRadioss benchmark with settings of Model: INIVOL and Fluid Structure Interaction Drop Container. Core Ultra 9 285K was the fastest.

The Arrow Lake CPU also stands up well for Altair’s OpenRadioss software.

GPAW benchmark with settings of Input: Carbon Nanotube. Core Ultra 9 285K was the fastest.

QMCPACK benchmark with settings of Input: H4_ae. Core Ultra 9 285K was the fastest.

High Performance Conjugate Gradient benchmark with settings of X Y Z: 104 104 104, RT: 60. Core Ultra 9 285K was the fastest.

Xcompact3d Incompact3d benchmark with settings of Input: input.i3d 193 Cells Per Direction. Core Ultra 9 285K was the fastest.

Pennant benchmark with settings of Test: sedovbig. Ryzen 9 9950X was the fastest.

In some of the other HPC applications the Core Ultra 9 285K with 24 physical cores does well in scenarios where SMP isn’t leveraged.

GROMACS benchmark with settings of Implementation: MPI CPU, Input: water_GMX50_bare. Ryzen 9 9950X was the fastest.

GROMACS benchmark with settings of Input: water_GMX50_bare. Core Ultra 9 285K was the fastest.

SPECFEM3D benchmark with settings of Model: Layered Halfspace. Core Ultra 9 285K was the fastest.

SPECFEM3D benchmark with settings of Model: Water-layered Halfspace. Core Ultra 9 285K was the fastest.

miniBUDE benchmark with settings of Implementation: OpenMP, Input Deck: BM1. Ryzen 9 9950X was the fastest.

ACES DGEMM benchmark with settings of Sustained Floating-Point Rate. Ryzen 9 9950X was the fastest.

ASKAP benchmark with settings of Test: tConvolve MPI, Degridding. Ryzen 9 9950X was the fastest.

ASKAP benchmark with settings of Test: tConvolve MPI, Gridding. Ryzen 9 9900X was the fastest.

CloverLeaf benchmark with settings of Input: clover_bm. Ryzen 7 9800X3D was the fastest.

In cases where AVX-512 can be utilized, the Ryzen 9000 series is the definitive winner over the Intel Core Ultra Series 2 desktop processors.

ASKAP benchmark with settings of Test: tConvolve OpenMP, Gridding. Ryzen 7 9800X3D was the fastest.

ASKAP benchmark with settings of Test: Hogbom Clean OpenMP. Ryzen 7 9800X3D was the fastest.

The AMD 3D V-Cache also helps out a lot in some of the HPC benchmarks.

nginx benchmark with settings of Connections: 500. Core Ultra 9 285K was the fastest.

nginx benchmark with settings of Connections: 1000. Core Ultra 9 285K was the fastest.

The Core Ultra 9 285K did take the lead in raw performance for the Nginx HTTPS web server.

Apache Cassandra benchmark with settings of Test: Writes. Ryzen 9 9950X was the fastest.

For Apache Cassandra 5.0, the Ryzen 9 9950X returned to the top spot.

Apache IoTDB benchmark with settings of Device Count: 800, Batch Size Per Write: 100, Sensor Count: 800, Client Number: 400. Core Ultra 9 285K was the fastest.

PostgreSQL benchmark with settings of Scaling Factor: 100, Clients: 500, Mode: Read Only. Ryzen 9 9950X was the fastest.

PostgreSQL benchmark with settings of Scaling Factor: 100, Clients: 500, Mode: Read Only, Average Latency. Ryzen 9 9950X was the fastest.

ClickHouse benchmark with settings of 100M Rows Hits Dataset, First Run / Cold Cache. Ryzen 7 9800X3D was the fastest.

ClickHouse benchmark with settings of 100M Rows Hits Dataset, Third Run. Ryzen 7 9800X3D was the fastest.

clickhouse

The Core Ultra 9 285K stands up well in some of the server workloads tested on these desktop CPUs but in other cases the Ryzen 9 9900 series dominates or even some of the database workloads like the Ryzen 7 9800X3D with its 3D V-Cache.

CockroachDB benchmark with settings of Workload: KV, 95% Reads, Concurrency: 128. Core Ultra 9 285K was the fastest.

Memcached benchmark with settings of Set To Get Ratio: 1:100. Ryzen 9 9950X was the fastest.

Memcached benchmark with settings of Set To Get Ratio: 1:10. Ryzen 9 9950X was the fastest.

Memcached benchmark with settings of Set To Get Ratio: 1:5. Ryzen 9 9950X was the fastest.

RocksDB benchmark with settings of Test: Random Read. Core Ultra 9 285K was the fastest.

RocksDB benchmark with settings of Test: Read While Writing. Ryzen 9 9950X was the fastest.

RocksDB benchmark with settings of Test: Read Random Write Random. Ryzen 9 9950X was the fastest.

Speedb benchmark with settings of Test: Random Read. Core Ultra 9 285K was the fastest.

Speedb benchmark with settings of Test: Read While Writing. Ryzen 9 9950X was the fastest.

These benchmarks are interesting for those thinking of these desktop CPUs for budget/SOHO server purposes.

Node.js V8 Web Tooling Benchmark benchmark with settings of . Ryzen 7 9800X3D was the fastest.

PyBench benchmark with settings of Total For Average Test Times. Core Ultra 9 285K was the fastest.

PyPerformance benchmark with settings of Benchmark: async_tree_io. Ryzen 7 9800X3D was the fastest.

PyPerformance benchmark with settings of Benchmark: asyncio_tcp_ssl. Ryzen 7 9800X3D was the fastest.

PyPerformance benchmark with settings of Benchmark: asyncio_websockets. Core Ultra 9 285K was the fastest.

PyPerformance benchmark with settings of Benchmark: float. Core Ultra 9 285K was the fastest.

PyPerformance benchmark with settings of Benchmark: json_loads. Core Ultra 9 285K was the fastest.

PyPerformance benchmark with settings of Benchmark: xml_etree. Core Ultra 9 285K was the fastest.

The single-threaded Python scripting benchmarks were mixed.

simdjson benchmark with settings of Throughput Test: PartialTweets. Ryzen 9 9950X was the fastest.

simdjson benchmark with settings of Throughput Test: LargeRandom. Ryzen 9 9950X was the fastest.

simdjson benchmark with settings of Throughput Test: DistinctUserID. Ryzen 9 9950X was the fastest.

With simdjson for speedy JSON parsing that can leverage AVX-512, the AMD Ryzen 9000 series perform much better than the Intel Core Ultra Series 2.

PHPBench benchmark with settings of PHP Benchmark Suite. Core Ultra 9 285K was the fastest.

For single-threaded PHP scripting performance, Arrow Lake was coming out ahead of the Zen 5 CPUs.

DaCapo Benchmark benchmark with settings of Java Test: Eclipse. Ryzen 9 9950X was the fastest.

DaCapo Benchmark benchmark with settings of Java Test: Batik SVG Toolkit. Core Ultra 9 285K was the fastest.

DaCapo Benchmark benchmark with settings of Java Test: Apache Tomcat. Ryzen 9 9950X was the fastest.

DaCapo Benchmark benchmark with settings of Java Test: H2 Database Engine. Ryzen 7 9800X3D was the fastest.

Renaissance benchmark with settings of Test: Apache Spark Bayes. Core Ultra 9 285K was the fastest.

Renaissance benchmark with settings of Test: Apache Spark PageRank. Ryzen 7 9800X3D was the fastest.

Java JMH benchmark with settings of Throughput. Core Ultra 9 285K was the fastest.

The Ryzen 9000 series processors were also picking up the most wins across a range of OpenJDK Java benchmarks on Ubuntu 24.10.

AOM AV1 benchmark with settings of Encoder Mode: Speed 11 Realtime, Input: Bosphorus 4K. Ryzen 7 9800X3D was the fastest.

AOM AV1 benchmark with settings of Encoder Mode: Speed 10 Realtime, Input: Bosphorus 4K. Ryzen 7 9800X3D was the fastest.

AOM AV1 benchmark with settings of Encoder Mode: Speed 8 Realtime, Input: Bosphorus 4K. Ryzen 7 9800X3D was the fastest.

With the AOM-AV1 video encoder that is less scalable than the likes of SVT-AV1, the Ryzen 7 9800X3D with 3D V-Cache was the sweet spot of the tested CPUs.

SVT-AV1 benchmark with settings of Encoder Mode: Preset 13, Input: Bosphorus 4K. Core Ultra 9 285K was the fastest.

SVT-AV1 benchmark with settings of Encoder Mode: Preset 8, Input: Bosphorus 4K. Ryzen 9 9950X was the fastest.

SVT-AV1 benchmark with settings of Encoder Mode: Preset 5, Input: Bosphorus 4K. Ryzen 9 9950X was the fastest.

But with SVT-AV1, the 16-core and 24-core parts were obviously battling it out for the top spot.

Kvazaar benchmark with settings of Video Input: Bosphorus 4K, Video Preset: Slow. Ryzen 9 9950X was the fastest.

Kvazaar benchmark with settings of Video Input: Bosphorus 4K, Video Preset: Very Fast. Ryzen 9 9950X was the fastest.

Kvazaar benchmark with settings of Video Input: Bosphorus 4K, Video Preset: Ultra Fast. Ryzen 9 9950X was the fastest.

uvg266 benchmark with settings of Video Input: Bosphorus 4K, Video Preset: Slow. Ryzen 9 9950X was the fastest.

uvg266 benchmark with settings of Video Input: Bosphorus 4K, Video Preset: Very Fast. Ryzen 9 9950X was the fastest.

uvg266 benchmark with settings of Video Input: Bosphorus 4K, Video Preset: Ultra Fast. Ryzen 9 9950X was the fastest.

VVenC benchmark with settings of Video Input: Bosphorus 4K, Video Preset: Fast. Core Ultra 9 285K was the fastest.

VVenC benchmark with settings of Video Input: Bosphorus 4K, Video Preset: Faster. Core Ultra 9 285K was the fastest.

FFmpeg benchmark with settings of Encoder: libx265, Scenario: Live. Ryzen 9 9950X was the fastest.

FFmpeg benchmark with settings of Encoder: libx265, Scenario: Platform. Ryzen 9 9950X was the fastest.

x265 benchmark with settings of Video Input: Bosphorus 4K. Core Ultra 9 285K was the fastest.

NeatBench benchmark with settings of Acceleration: CPU. Ryzen 9 9950X was the fastest.

Across a variety of video coding benchmarks and different formats, the AMD Ryzen 9 9950X tended to hold onto the top spot.

FLAC Audio Encoding benchmark with settings of WAV To FLAC. Ryzen 9 9950X was the fastest.

WavPack Audio Encoding benchmark with settings of WAV To WavPack. Ryzen 9 9950X was the fastest.

LAME MP3 Encoding benchmark with settings of WAV To MP3. Ryzen 9 9950X was the fastest.

Ogg Audio Encoding benchmark with settings of WAV To Ogg. Core Ultra 9 285K was the fastest.

Opus Codec Encoding benchmark with settings of WAV To Opus Encode. Ryzen 9 9950X was the fastest.

With the exception of Ogg audio encoding, in other audio encoding benchmarks the Ryzen 9 9900 series was the leader.

libavif avifenc benchmark with settings of Encoder Speed: 2. Ryzen 9 9950X was the fastest.

ASTC Encoder benchmark with settings of Preset: Medium. Ryzen 9 9950X was the fastest.

ASTC Encoder benchmark with settings of Preset: Thorough. Core Ultra 9 285K was the fastest.

ASTC Encoder benchmark with settings of Preset: Exhaustive. Core Ultra 9 285K was the fastest.

Basis Universal benchmark with settings of Settings: ETC1S Level 3. Ryzen 9 9950X was the fastest.

Basis Universal benchmark with settings of Settings: UASTC Level 3. Core Ultra 9 285K was the fastest.

Etcpak benchmark with settings of Benchmark: Multi-Threaded, Configuration: ETC2. Ryzen 9 9950X was the fastest.

JPEG-XL libjxl benchmark with settings of Input: PNG, Quality: 100. Ryzen 9 9950X was the fastest.

JPEG-XL Decoding libjxl benchmark with settings of CPU Threads: All. Ryzen 7 9800X3D was the fastest.

WebP Image Encode benchmark with settings of Encode Settings: Quality 100, Highest Compression. Core Ultra 9 285K was the fastest.

It was a fun benchmarking battle to say the least.

TensorFlow benchmark with settings of Device: CPU, Batch Size: 64, Model: ResNet-50. Ryzen 9 9950X was the fastest.

For AI workloads the Ryzen 9000 series tends to perform very well with its AVX-512 support having a full 512-bit data path.

OpenVINO benchmark with settings of Model: Face Detection FP16-INT8, Device: CPU. Ryzen 9 9950X was the fastest.

OpenVINO benchmark with settings of Model: Age Gender Recognition Retail 0013 FP16-INT8, Device: CPU. Ryzen 9 9950X was the fastest.

OpenVINO benchmark with settings of Model: Person Detection FP16, Device: CPU. Ryzen 7 9800X3D was the fastest.

OpenVINO benchmark with settings of Model: Weld Porosity Detection FP16-INT8, Device: CPU. Ryzen 9 9950X was the fastest.

OpenVINO benchmark with settings of Model: Vehicle Detection FP16-INT8, Device: CPU. Ryzen 9 9950X was the fastest.

OpenVINO benchmark with settings of Model: Person Vehicle Bike Detection FP16, Device: CPU. Ryzen 9 9950X was the fastest.

OpenVINO benchmark with settings of Model: Face Detection Retail FP16-INT8, Device: CPU. Ryzen 9 9950X was the fastest.

With Intel’s own OpenVINO software the best performance was with the Ryzen 9000 series, again thanks in large part to AVX-512 on these desktop CPUs/

OpenVINO GenAI benchmark with settings of Model: TinyLlama-1.1B-Chat-v1.0, Device: CPU. Core Ultra 5 245K was the fastest.

OpenVINO GenAI benchmark with settings of Model: Phi-3-mini-128k-instruct-int4-ov, Device: CPU. Core Ultra 9 285K was the fastest.

OpenVINO GenAI benchmark with settings of Model: Gemma-7b-int4-ov, Device: CPU. Core Ultra 9 285K was the fastest.

In some of the OpenVINO GenAI benchmarks the 24-core Core Ultra 9 285K was able to pull off an upset.

LiteRT benchmark with settings of Model: Mobilenet Quant. Ryzen 7 9800X3D was the fastest.

LiteRT benchmark with settings of Model: Inception ResNet V2. Ryzen 7 9800X3D was the fastest.

LiteRT benchmark with settings of Model: Inception V4. Ryzen 9 9950X was the fastest.

NCNN benchmark with settings of Target: CPU, Model: vision_transformer. Ryzen 9 9950X was the fastest.

Whisper.cpp benchmark with settings of Model: ggml-base.en, Input: 2016 State of the Union. Ryzen 9 9950X was the fastest.

Whisper.cpp benchmark with settings of Model: ggml-small.en, Input: 2016 State of the Union. Ryzen 9 9950X was the fastest.

Llama.cpp benchmark with settings of Backend: CPU BLAS, Model: Llama-3.1-Tulu-3-8B-Q8_0, Test: Prompt Processing 512. Ryzen 9 9950X was the fastest.

Llama.cpp benchmark with settings of Backend: CPU BLAS, Model: Llama-3.1-Tulu-3-8B-Q8_0, Test: Prompt Processing 1024. Ryzen 9 9950X was the fastest.

Llama.cpp benchmark with settings of Backend: CPU BLAS, Model: granite-3.0-3b-a800m-instruct-Q8_0, Test: Text Generation 128. Ryzen 9 9900X was the fastest.

Llama.cpp benchmark with settings of Backend: CPU BLAS, Model: granite-3.0-3b-a800m-instruct-Q8_0, Test: Prompt Processing 512. Ryzen 9 9950X was the fastest.

Llama.cpp benchmark with settings of Backend: CPU BLAS, Model: Mistral-7B-Instruct-v0.3-Q8_0, Test: Prompt Processing 512. Ryzen 9 9950X was the fastest.

ONNX Runtime benchmark with settings of Model: ResNet101_DUC_HDC-12, Device: CPU, Executor: Standard. Ryzen 9 9950X was the fastest.

Across nearly all of the CPU-based AI benchmarks carried out, the Ryzen 9000 series were the standout leaders.

RawTherapee benchmark with settings of Total Benchmark Time. Ryzen 9 9950X was the fastest.

Darktable benchmark with settings of Test: Boat, Acceleration: CPU-only. Ryzen 9 9950X was the fastest.

Darktable benchmark with settings of Test: Masskrug, Acceleration: CPU-only. Ryzen 9 9950X was the fastest.

GIMP benchmark with settings of Test: rotate. Ryzen 9 9900X was the fastest.

GIMP benchmark with settings of Test: auto-levels. Ryzen 7 9700X was the fastest.

GEGL benchmark with settings of Operation: Scale. Ryzen 7 9700X was the fastest.

There was healthy competition between Arrow Lake and Zen 5 with these Linux imaging application benchmarks but overall the Ryzen 9000 series had the most wins.

OpenSCAD benchmark with settings of Render: Leonardo Phone Case Slim. Ryzen 9 9900X was the fastest.

OpenSCAD benchmark with settings of Render: Pistol. Ryzen 9 9950X was the fastest.

OpenSCAD benchmark with settings of Render: Mini-ITX Case. Core Ultra 9 285K was the fastest.

Healthy competition in OpenSCAD too for open-source CAD.

GNU Octave Benchmark benchmark with settings of . Ryzen 7 9800X3D was the fastest.

The Ryzen 9000 series swept the race for the GNU Octave program as an open-source alternative to MATLAB.

3DMark Wild Life Extreme benchmark with settings of Resolution: 3840 x 2160. Core Ultra 5 245K was the fastest.

GPUScore: Breaking Limit benchmark with settings of Resolution: 1920 x 1080, Ray-Tracing: Off. Core Ultra 5 245K was the fastest.

With some of the graphics/gaming benchmarks the Intel Arrow Lake CPUs had the advantage.

Xonotic benchmark with settings of Resolution: 3840 x 2160, Effects Quality: Low. Ryzen 9 9950X was the fastest.

Xonotic benchmark with settings of Resolution: 3840 x 2160, Effects Quality: High. Ryzen 9 9950X was the fastest.

Xonotic benchmark with settings of Resolution: 3840 x 2160, Effects Quality: Ultra. Ryzen 9 9950X was the fastest.

Xonotic benchmark with settings of Resolution: 3840 x 2160, Effects Quality: Ultimate. Ryzen 9 9950X was the fastest.

Unvanquished benchmark with settings of Resolution: 3840 x 2160, Effects Quality: Medium. Ryzen 9 9950X was the fastest.

Unvanquished benchmark with settings of Resolution: 3840 x 2160, Effects Quality: Ultra. Ryzen 9 9950X was the fastest.

But in other lightweight, CPU-bound games the AMD Ryzen 9000 series had the advantage. The results from Unvanquished are rather interesting as it would appear that even with the Linux 6.13 kernel there are still scheduler issues where at times some workloads are being punted to the E cores rather than the P cores.

ParaView benchmark with settings of Test: Wavelet Volume, Frames: 600, Resolution: 3840 x 2160. Ryzen 9 9950X was the fastest.

VKMark benchmark with settings of Resolution: 1920 x 1080, Present Mode: Mailbox. Ryzen 7 9700X was the fastest.

VKMark benchmark with settings of Resolution: 3840 x 2160, Present Mode: Mailbox. Ryzen 7 9700X was the fastest.

Number Of First Place Finishes benchmark with settings of Wins, 402 Tests.

When looking at the wins and losses breakdown from 402 test cases, the Ryzen 9 9950X straight-up won 50% of the time with a first place finish. The Core Ultra 9 285K was the first place finisher 28% of the time and the Ryzen 7 9800X3D in first around 14% of the time with its 3D V-Cache. With a mix of single and multi-threaded workloads explored, in a handful of cases the Ryzen 9 9900X and other CPUs also won.

Geometric Mean Of All Test Results benchmark with settings of Result Composite, Intel Core Ultra, AMD Ryzen 9000, Linux 2025. Ryzen 9 9950X was the fastest.

When taking the geometric mean of all the benchmarks tested on all of the CPUs, the Ryzen 9 9950X was the hands-down winner from this wide-reaching Linux performance comparison. The AMD Ryzen 9 9950X was 18% faster than the Intel Core Ultra 9 285K on average and even the Ryzen 9 9900X was 6% faster than the Core Ultra 9 285K. As of writing the Intel Core Ultra 9 285K was retailing for around $599 USD while the Ryzen 9 9950X can be found for about $560 USD!

The Ryzen 7 9700X meanwhile was 6% faster than the Core Ultra 5 245K. The Core Ultra 5 245K currently is retailing for $299 USD while the Ryzen 7 9700X goes for around $309.

CPU Power Consumption Monitoring Overview benchmark with settings of Accumulated CPU Power Consumption Monitoring.

On average across all the benchmarks conducted, the Ryzen 9 9950X had a CPU power consumption average of 148 Watts while the Core Ultra 9 285K had a 140 Watt average. Where the power consumption is concerning for Arrow Lake is on the top-end at peaking at 312 Watts compared to 210 Watts with the Ryzen 9 9950X. The Ryzen 7 9700X meanwhile consumed far less power than the Core Ultra 5 245K while enjoying better performance.

For those wondering if BIOS updates and/or the newest Linux kernel and other open-source software updates have changed the picture for Zen 5 vs. Arrow Lake, in relative terms it’s similar to what we have seen in prior months. Albeit in some cases some healthy performance uplift at large thanks to the Linux software improvements. Those wanting to go through all ~400 benchmark results and power data in full can do so via this result page.