AMD EPYC 9654 EOY2024

2 x AMD EPYC 9654 96-Core testing with a AMD Titanite_4G (RTI1007B BIOS) and ASPEED on Ubuntu 23.10 via the Phoronix Test Suite.

HTML result view exported from: https://openbenchmarking.org/result/2412270-NE-AMDEPYC9698&grs.

Stress-NG

Test: Crypto

LiteRT

Model: Quantized COCO SSD MobileNet v1

LiteRT

Model: Mobilenet Quant

XNNPACK

Model: FP32MobileNetV3Large

XNNPACK

Model: FP16MobileNetV3Large

XNNPACK

Model: FP16MobileNetV3Small

XNNPACK

Model: FP32MobileNetV1

XNNPACK

Model: QS8MobileNetV2

LiteRT

Model: Mobilenet Float

XNNPACK

Model: FP32MobileNetV3Small

XNNPACK

Model: FP16MobileNetV1

LiteRT

Model: DeepLab V3

LiteRT

Model: Inception ResNet V2

oneDNN

Harness: Convolution Batch Shapes Auto - Engine: CPU

XNNPACK

Model: FP32MobileNetV2

Stockfish

Chess Benchmark

LiteRT

Model: SqueezeNet

Graph500

Scale: 26

C-Ray

Resolution: 1080p - Rays Per Pixel: 16

SVT-AV1

Encoder Mode: Preset 8 - Input: Bosphorus 4K

srsRAN Project

Test: PUSCH Processor Benchmark, Throughput Total

simdjson

Throughput Test: DistinctUserID

SVT-AV1

Encoder Mode: Preset 13 - Input: Bosphorus 4K

uvg266

Video Input: Bosphorus 4K - Video Preset: Ultra Fast

Renaissance

Test: Apache Spark PageRank

Graph500

Scale: 26

SVT-AV1

Encoder Mode: Preset 5 - Input: Bosphorus 4K

uvg266

Video Input: Bosphorus 4K - Video Preset: Slow

Laghos

Test: Triple Point Problem

GraphicsMagick

Operation: Enhanced

GraphicsMagick

Operation: Swirl

OpenVINO

Model: Handwritten English Recognition FP16 - Device: CPU

OpenVINO

Model: Handwritten English Recognition FP16 - Device: CPU

Timed Linux Kernel Compilation

Build: allmodconfig

oneDNN

Harness: Deconvolution Batch shapes_1d - Engine: CPU

Graph500

Scale: 26

GraphicsMagick

Operation: Resizing

CP2K Molecular Dynamics

Input: H20-64

XNNPACK

Model: FP16MobileNetV2

Stress-NG

Test: Context Switching

OpenVINO GenAI

Model: Phi-3-mini-128k-instruct-int4-ov - Device: CPU

uvg266

Video Input: Bosphorus 4K - Video Preset: Very Fast

LiteRT

Model: Inception V4

LiteRT

Model: NASNet Mobile

GraphicsMagick

Operation: Sharpen

uvg266

Video Input: Bosphorus 4K - Video Preset: Super Fast

Rustls

Benchmark: handshake-ticket - Suite: TLS_ECDHE_ECDSA_WITH_AES_256_GCM_SHA384

SVT-AV1

Encoder Mode: Preset 3 - Input: Bosphorus 4K

OpenVINO

Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU

simdjson

Throughput Test: TopTweet

Stress-NG

Test: Semaphores

GraphicsMagick

Operation: Noise-Gaussian

Renaissance

Test: Apache Spark Bayes

uvg266

Video Input: Bosphorus 4K - Video Preset: Medium

WebP Image Encode

Encode Settings: Quality 100

oneDNN

Harness: Deconvolution Batch shapes_3d - Engine: CPU

7-Zip Compression

Test: Compression Rating

Blender

Blend File: BMW27 - Compute: CPU-Only

OpenVINO GenAI

Model: TinyLlama-1.1B-Chat-v1.0 - Device: CPU

Rustls

Benchmark: handshake-resume - Suite: TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256

CP2K Molecular Dynamics

Input: Fayalite-FIST

Build2

Time To Compile

Palabos

Grid Size: 100

ASTC Encoder

Preset: Medium

Rustls

Benchmark: handshake-ticket - Suite: TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256

x265

Video Input: Bosphorus 1080p

C-Ray

Resolution: 5K - Rays Per Pixel: 16

CP2K Molecular Dynamics

Input: H20-256

NAMD

Input: ATPase with 327,506 Atoms

srsRAN Project

Test: PDSCH Processor Benchmark, Throughput Total

Blender

Blend File: Pabellon Barcelona - Compute: CPU-Only

OpenVINO

Model: Noise Suppression Poconet-Like FP16 - Device: CPU

OSPRay Studio

Camera: 3 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU

x265

Video Input: Bosphorus 4K

WebP Image Encode

Encode Settings: Quality 100, Lossless

Y-Cruncher

Pi Digits To Calculate: 500M

C-Ray

Resolution: 4K - Rays Per Pixel: 16

OSPRay Studio

Camera: 3 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU

Blender

Blend File: Fishy Cat - Compute: CPU-Only

Intel Open Image Denoise

Run: RTLightmap.hdr.4096x4096 - Device: CPU-Only

Y-Cruncher

Pi Digits To Calculate: 5B

Intel Open Image Denoise

Run: RT.ldr_alb_nrm.3840x2160 - Device: CPU-Only

Intel Open Image Denoise

Run: RT.hdr_alb_nrm.3840x2160 - Device: CPU-Only

ASTC Encoder

Preset: Thorough

OSPRay Studio

Camera: 1 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU

OSPRay Studio

Camera: 1 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU

OSPRay Studio

Camera: 2 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU

OSPRay Studio

Camera: 2 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU

OpenVINO

Model: Weld Porosity Detection FP16-INT8 - Device: CPU

OpenVINO

Model: Person Detection FP16 - Device: CPU

Timed FFmpeg Compilation

Time To Compile

GraphicsMagick

Operation: HWB Color Space

OpenVINO

Model: Person Detection FP16 - Device: CPU

Blender

Blend File: Classroom - Compute: CPU-Only

OpenVINO GenAI

Model: Falcon-7b-instruct-int4-ov - Device: CPU

Timed Node.js Compilation

Time To Compile

OSPRay Studio

Camera: 2 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU

OSPRay Studio

Camera: 1 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU

OpenVINO

Model: Noise Suppression Poconet-Like FP16 - Device: CPU

GROMACS

Input: water_GMX50_bare

Primesieve

Length: 1e12

OpenVINO

Model: Weld Porosity Detection FP16-INT8 - Device: CPU

WarpX

Input: Uniform Plasma

ASTC Encoder

Preset: Fast

OpenVINO

Model: Handwritten English Recognition FP16-INT8 - Device: CPU

OSPRay Studio

Camera: 2 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU

OpenVINO

Model: Machine Translation EN To DE FP16 - Device: CPU

OpenVINO

Model: Handwritten English Recognition FP16-INT8 - Device: CPU

OpenVINO

Model: Machine Translation EN To DE FP16 - Device: CPU

WarpX

Input: Plasma Acceleration

simdjson

Throughput Test: PartialTweets

Blender

Blend File: Junkshop - Compute: CPU-Only

oneDNN

Harness: IP Shapes 1D - Engine: CPU

WebP Image Encode

Encode Settings: Quality 100, Highest Compression

Palabos

Grid Size: 500

OpenVINO

Model: Road Segmentation ADAS FP16-INT8 - Device: CPU

OpenVINO

Model: Road Segmentation ADAS FP16 - Device: CPU

OpenVINO

Model: Road Segmentation ADAS FP16-INT8 - Device: CPU

NAMD

Input: STMV with 1,066,628 Atoms

OSPRay Studio

Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU

oneDNN

Harness: IP Shapes 3D - Engine: CPU

OpenVINO

Model: Road Segmentation ADAS FP16 - Device: CPU

Stress-NG

Test: CPU Stress

Rustls

Benchmark: handshake - Suite: TLS_ECDHE_ECDSA_WITH_AES_256_GCM_SHA384

Rustls

Benchmark: handshake-resume - Suite: TLS_ECDHE_ECDSA_WITH_AES_256_GCM_SHA384

WebP Image Encode

Encode Settings: Default

OpenVINO GenAI

Model: Gemma-7b-int4-ov - Device: CPU

ACES DGEMM

Sustained Floating-Point Rate

Graph500

Scale: 26

OpenSSL

Algorithm: AES-256-GCM

Timed Linux Kernel Compilation

Build: defconfig

Blender

Blend File: Barbershop - Compute: CPU-Only

Timed PHP Compilation

Time To Compile

OpenSSL

Algorithm: SHA512

Palabos

Grid Size: 400

Stress-NG

Test: Socket Activity

Y-Cruncher

Pi Digits To Calculate: 1B

OpenSSL

Algorithm: ChaCha20-Poly1305

QuantLib

Size: S

OpenVINO

Model: Weld Porosity Detection FP16 - Device: CPU

BRL-CAD

VGR Performance Metric

Primesieve

Length: 1e13

oneDNN

Harness: Recurrent Neural Network Training - Engine: CPU

OpenSSL

Algorithm: AES-128-GCM

Stress-NG

Test: Radix String Sort

OpenVINO

Model: Person Vehicle Bike Detection FP16 - Device: CPU

OpenVINO

Model: Person Re-Identification Retail FP16 - Device: CPU

OpenSSL

Algorithm: SHA256

ASTC Encoder

Preset: Very Thorough

OpenVINO

Model: Vehicle Detection FP16-INT8 - Device: CPU

Timed Eigen Compilation

Time To Compile

OpenSSL

Algorithm: ChaCha20-Poly1305

Rustls

Benchmark: handshake - Suite: TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256

7-Zip Compression

Test: Decompression Rating

OpenVINO

Model: Face Detection FP16-INT8 - Device: CPU

OpenVINO

Model: Face Detection FP16-INT8 - Device: CPU

OpenVINO

Model: Face Detection Retail FP16 - Device: CPU

Epoch