AMD Ryzen 9 7950X3D Modes On Linux
Ryzen 9 7950X3D benchmarks for a future article by Michael Larabel.
HTML result view exported from: https://openbenchmarking.org/result/2302261-NE-7950X3DMO02&sor&grs&export=pdf.
oneDNN
Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU
srsRAN
Test: 4G PHY_DL_Test 100 PRB SISO 256-QAM
srsRAN
Test: OFDM_Test
srsRAN
Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM
KTX-Software toktx
Settings: UASTC 3 + Zstd Compression 19
srsRAN
Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM
simdjson
Throughput Test: PartialTweets
PyBench
Total For Average Test Times
srsRAN
Test: 4G PHY_DL_Test 100 PRB SISO 64-QAM
KTX-Software toktx
Settings: Zstd Compression 19
srsRAN
Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAM
srsRAN
Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM
simdjson
Throughput Test: LargeRandom
Google Draco
Model: Lion
srsRAN
Test: 4G PHY_DL_Test 100 PRB SISO 256-QAM
srsRAN
Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM
Radiance Benchmark
Test: Serial
simdjson
Throughput Test: TopTweet
simdjson
Throughput Test: DistinctUserID
PHPBench
PHP Benchmark Suite
Pennant
Test: leblancbig
QuantLib
GNU Radio
Test: FIR Filter
simdjson
Throughput Test: Kostya
ASKAP
Test: tConvolve MT - Gridding
Numpy Benchmark
LZ4 Compression
Compression Level: 3 - Compression Speed
Zstd Compression
Compression Level: 19, Long Mode - Decompression Speed
LAME MP3 Encoding
WAV To MP3
Numenta Anomaly Benchmark
Detector: Earthgecko Skyline
libavif avifenc
Encoder Speed: 0
ACES DGEMM
Sustained Floating-Point Rate
AOM AV1
Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K
WebP Image Encode
Encode Settings: Default
oneDNN
Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU
Zstd Compression
Compression Level: 3, Long Mode - Decompression Speed
GNU Radio
Test: Hilbert Transform
Numenta Anomaly Benchmark
Detector: Bayesian Changepoint
AOM AV1
Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K
OpenFOAM
Input: drivaerFastback, Small Mesh Size - Execution Time
OpenFOAM
Input: drivaerFastback, Small Mesh Size - Mesh Time
SQLite Speedtest
Timed Time - Size 1,000
Selenium
Benchmark: ARES-6 - Browser: Google Chrome
RNNoise
Numenta Anomaly Benchmark
Detector: Contextual Anomaly Detector OSE
GraphicsMagick
Operation: HWB Color Space
GNU Radio
Test: Signal Source (Cosine)
GIMP
Test: rotate
WebP Image Encode
Encode Settings: Quality 100, Highest Compression
Stress-NG
Test: CPU Stress
GNU Radio
Test: Five Back to Back FIR Filters
Timed LLVM Compilation
Build System: Unix Makefiles
libavif avifenc
Encoder Speed: 2
VP9 libvpx Encoding
Speed: Speed 5 - Input: Bosphorus 4K
ClickHouse
100M Rows Hits Dataset, Second Run
ClickHouse
100M Rows Hits Dataset, Third Run
OpenVKL
Benchmark: vklBenchmark ISPC
LuaRadio
Test: Five Back to Back FIR Filters
KTX-Software toktx
Settings: UASTC 3
ONNX Runtime
Model: GPT-2 - Device: CPU - Executor: Standard
WebP Image Encode
Encode Settings: Quality 100
Zstd Compression
Compression Level: 19 - Decompression Speed
AOM AV1
Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K
AOM AV1
Encoder Mode: Speed 9 Realtime - Input: Bosphorus 1080p
OSPRay Studio
Camera: 2 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer
NCNN
Target: CPU - Model: mobilenet
Unpacking Firefox
Extracting: firefox-84.0.source.tar.xz
WebP Image Encode
Encode Settings: Quality 100, Lossless
GraphicsMagick
Operation: Resizing
TNN
Target: CPU - Model: SqueezeNet v2
Darktable
Test: Server Rack - Acceleration: CPU-only
OSPRay Studio
Camera: 1 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer
NCNN
Target: CPU - Model: efficientnet-b0
Unpacking The Linux Kernel
linux-5.19.tar.xz
PyHPC Benchmarks
Device: CPU - Backend: Numpy - Project Size: 4194304 - Benchmark: Isoneutral Mixing
GEGL
Operation: Color Enhance
OSPRay Studio
Camera: 2 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer
OSPRay Studio
Camera: 3 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer
LZ4 Compression
Compression Level: 3 - Decompression Speed
Zstd Compression
Compression Level: 19 - Compression Speed
LuxCoreRender
Scene: Danish Mood - Acceleration: CPU
Xcompact3d Incompact3d
Input: input.i3d 129 Cells Per Direction
OSPRay Studio
Camera: 3 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer
Pennant
Test: sedovbig
Selenium
Benchmark: Jetstream 2 - Browser: Google Chrome
OSPRay Studio
Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer
Node.js V8 Web Tooling Benchmark
GEGL
Operation: Rotate 90 Degrees
OSPRay Studio
Camera: 2 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer
NCNN
Target: CPU-v3-v3 - Model: mobilenet-v3
OSPRay Studio
Camera: 1 - Resolution: 1080p - Samples Per Pixel: 32 - Renderer: Path Tracer
LZ4 Compression
Compression Level: 9 - Compression Speed
Zstd Compression
Compression Level: 19, Long Mode - Compression Speed
Cpuminer-Opt
Algorithm: Myriad-Groestl
Stargate Digital Audio Workstation
Sample Rate: 192000 - Buffer Size: 512
rav1e
Speed: 5
GNU Radio
Test: FM Deemphasis Filter
ClickHouse
100M Rows Hits Dataset, First Run / Cold Cache
libavif avifenc
Encoder Speed: 6
NCNN
Target: CPU - Model: resnet18
GEGL
Operation: Antialias
NCNN
Target: CPU - Model: blazeface
libjpeg-turbo tjbench
Test: Decompression Throughput
NCNN
Target: CPU-v2-v2 - Model: mobilenet-v2
rav1e
Speed: 10
oneDNN
Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU
NCNN
Target: CPU - Model: vision_transformer
GEGL
Operation: Reflect
OpenEMS
Test: pyEMS Coupler
GraphicsMagick
Operation: Rotate
libavif avifenc
Encoder Speed: 6, Lossless
Radiance Benchmark
Test: SMP Parallel
OSPRay Studio
Camera: 1 - Resolution: 1080p - Samples Per Pixel: 1 - Renderer: Path Tracer
Numenta Anomaly Benchmark
Detector: Relative Entropy
LuaRadio
Test: Complex Phase
Node.js Express HTTP Load Test
OSPRay Studio
Camera: 1 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer
Natron
Input: Spaceship
NCNN
Target: CPU - Model: alexnet
LeelaChessZero
Backend: Eigen
OpenVINO
Model: Vehicle Detection FP16 - Device: CPU
Zstd Compression
Compression Level: 3 - Compression Speed
PyHPC Benchmarks
Device: CPU - Backend: Numpy - Project Size: 1048576 - Benchmark: Equation of State
NCNN
Target: CPU - Model: googlenet
OpenVINO
Model: Vehicle Detection FP16 - Device: CPU
GIMP
Test: resize
GraphicsMagick
Operation: Swirl
Stress-NG
Test: Semaphores
GEGL
Operation: Scale
Gcrypt Library
NCNN
Target: CPU - Model: regnety_400m
NCNN
Target: CPU - Model: mnasnet
GNU Octave Benchmark
Renaissance
Test: ALS Movie Lens
LZ4 Compression
Compression Level: 9 - Decompression Speed
NCNN
Target: CPU - Model: squeezenet_ssd
libavif avifenc
Encoder Speed: 10, Lossless
AOM AV1
Encoder Mode: Speed 8 Realtime - Input: Bosphorus 1080p
Renaissance
Test: Savina Reactors.IO
Xcompact3d Incompact3d
Input: input.i3d 193 Cells Per Direction
Cpuminer-Opt
Algorithm: Ringcoin
Stargate Digital Audio Workstation
Sample Rate: 96000 - Buffer Size: 512
Renaissance
Test: Apache Spark PageRank
Renaissance
Test: Genetic Algorithm Using Jenetics + Futures
Stress-NG
Test: Glibc Qsort Data Sorting
VP9 libvpx Encoding
Speed: Speed 5 - Input: Bosphorus 1080p
Stress-NG
Test: Glibc C String Functions
Stress-NG
Test: Memory Copying
GIMP
Test: auto-levels
Zstd Compression
Compression Level: 12 - Compression Speed
OSPRay
Benchmark: gravity_spheres_volume/dim_512/scivis/real_time
Cpuminer-Opt
Algorithm: Deepcoin
Zstd Compression
Compression Level: 3, Long Mode - Compression Speed
PyHPC Benchmarks
Device: CPU - Backend: Numpy - Project Size: 1048576 - Benchmark: Isoneutral Mixing
Stress-NG
Test: Socket Activity
oneDNN
Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU
WireGuard + Linux Networking Stack Stress Test
GNU Radio
Test: IIR Filter
LAMMPS Molecular Dynamics Simulator
Model: Rhodopsin Protein
OpenVINO
Model: Vehicle Detection FP16-INT8 - Device: CPU
LuaRadio
Test: Hilbert Transform
GraphicsMagick
Operation: Enhanced
PyHPC Benchmarks
Device: CPU - Backend: Numpy - Project Size: 4194304 - Benchmark: Equation of State
WebP Image Encode
Encode Settings: Quality 100, Lossless, Highest Compression
VP9 libvpx Encoding
Speed: Speed 0 - Input: Bosphorus 4K
Selenium
Benchmark: PSPDFKit WASM - Browser: Google Chrome
GraphicsMagick
Operation: Noise-Gaussian
Stress-NG
Test: MEMFD
GEGL
Operation: Cartoon
VP9 libvpx Encoding
Speed: Speed 0 - Input: Bosphorus 1080p
Stress-NG
Test: Futex
AOM AV1
Encoder Mode: Speed 10 Realtime - Input: Bosphorus 1080p
Liquid-DSP
Threads: 16 - Buffer Length: 256 - Filter Length: 57
OSPRay
Benchmark: gravity_spheres_volume/dim_512/ao/real_time
AOM AV1
Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 1080p
SVT-AV1
Encoder Mode: Preset 13 - Input: Bosphorus 1080p
x265
Video Input: Bosphorus 4K
Timed Linux Kernel Compilation
Build: allmodconfig
OpenVINO
Model: Vehicle Detection FP16-INT8 - Device: CPU
SVT-AV1
Encoder Mode: Preset 8 - Input: Bosphorus 1080p
NCNN
Target: CPU - Model: FastestDet
Timed Mesa Compilation
Time To Compile
OpenVINO
Model: Person Vehicle Bike Detection FP16 - Device: CPU
NCNN
Target: CPU - Model: resnet50
NCNN
Target: CPU - Model: shufflenet-v2
Crafty
Elapsed Time
AOM AV1
Encoder Mode: Speed 6 Realtime - Input: Bosphorus 1080p
LuxCoreRender
Scene: DLSC - Acceleration: CPU
Stargate Digital Audio Workstation
Sample Rate: 96000 - Buffer Size: 1024
Sysbench
Test: RAM / Memory
Darktable
Test: Boat - Acceleration: CPU-only
SVT-VP9
Tuning: VMAF Optimized - Input: Bosphorus 1080p
Stress-NG
Test: SENDFILE
Zstd Compression
Compression Level: 8, Long Mode - Decompression Speed
oneDNN
Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU
oneDNN
Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU
Embree
Binary: Pathtracer - Model: Crown
Renaissance
Test: Apache Spark ALS
TNN
Target: CPU - Model: SqueezeNet v1.1
ASKAP
Test: tConvolve OpenMP - Gridding
Stargate Digital Audio Workstation
Sample Rate: 44100 - Buffer Size: 512
Numenta Anomaly Benchmark
Detector: KNN CAD
AOM AV1
Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K
OpenVINO
Model: Person Vehicle Bike Detection FP16 - Device: CPU
Stargate Digital Audio Workstation
Sample Rate: 44100 - Buffer Size: 1024
OSPRay
Benchmark: particle_volume/scivis/real_time
Darktable
Test: Server Room - Acceleration: CPU-only
NCNN
Target: CPU - Model: vgg16
AOM AV1
Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4K
Timed Linux Kernel Compilation
Build: defconfig
Renaissance
Test: Akka Unbalanced Cobwebbed Tree
AOM AV1
Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 1080p
Liquid-DSP
Threads: 8 - Buffer Length: 256 - Filter Length: 57
Stress-NG
Test: NUMA
Stress-NG
Test: Matrix Math
Appleseed
Scene: Emily
Zstd Compression
Compression Level: 3 - Decompression Speed
LuxCoreRender
Scene: Orange Juice - Acceleration: CPU
Xmrig
Variant: Monero - Hash Count: 1M
Renaissance
Test: Finagle HTTP Requests
GEGL
Operation: Wavelet Blur
LZ4 Compression
Compression Level: 1 - Decompression Speed
SVT-AV1
Encoder Mode: Preset 12 - Input: Bosphorus 4K
SVT-HEVC
Tuning: 7 - Input: Bosphorus 1080p
RocksDB
Test: Random Fill
GEGL
Operation: Crop
Stress-NG
Test: Forking
KTX-Software toktx
Settings: UASTC 4 + Zstd Compression 19
OpenVINO
Model: Person Detection FP32 - Device: CPU
Numenta Anomaly Benchmark
Detector: Windowed Gaussian
ASTC Encoder
Preset: Thorough
Renaissance
Test: Random Forest
DaCapo Benchmark
Java Test: Jython
Timed FFmpeg Compilation
Time To Compile
Selenium
Benchmark: WASM collisionDetection - Browser: Google Chrome
Zstd Compression
Compression Level: 12 - Decompression Speed
RocksDB
Test: Random Fill Sync
NCNN
Target: CPU - Model: yolov4-tiny
Embree
Binary: Pathtracer ISPC - Model: Crown
RocksDB
Test: Sequential Fill
Stargate Digital Audio Workstation
Sample Rate: 192000 - Buffer Size: 1024
Algebraic Multi-Grid Benchmark
Stress-NG
Test: Malloc
Zstd Compression
Compression Level: 8 - Decompression Speed
dav1d
Video Input: Summer Nature 4K
oneDNN
Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU
Selenium
Benchmark: Octane - Browser: Google Chrome
LuxCoreRender
Scene: Rainbow Colors and Prism - Acceleration: CPU
SVT-AV1
Encoder Mode: Preset 8 - Input: Bosphorus 4K
Timed MPlayer Compilation
Time To Compile
GROMACS
Implementation: MPI CPU - Input: water_GMX50_bare
x264
Video Input: Bosphorus 4K
IndigoBench
Acceleration: CPU - Scene: Supercar
Blender
Blend File: BMW27 - Compute: CPU-Only
OpenVINO
Model: Person Detection FP32 - Device: CPU
GEGL
Operation: Tile Glass
Cpuminer-Opt
Algorithm: x25x
Selenium
Benchmark: Speedometer - Browser: Google Chrome
Stress-NG
Test: Mutex
7-Zip Compression
Test: Compression Rating
rav1e
Speed: 6
OpenVKL
Benchmark: vklBenchmark Scalar
BRL-CAD
VGR Performance Metric
Renaissance
Test: Apache Spark Bayes
Darktable
Test: Masskrug - Acceleration: CPU-only
Stress-NG
Test: MMAP