CachyOS Threadripper AMD Linux
Benchmarks by Michael Larabel for a future article.


No v3 or v4: 

	Processor: AMD Ryzen Threadripper PRO 7995WX 96-Cores @ 5.19GHz (96 Cores / 192 Threads), Motherboard: HP Z6 G5 A Workstation 8B24 (U65 Ver. 01.01.04 BIOS), Chipset: AMD Device 14a4, Memory: 8 x 16GB DDR5-5200MT/s Hynix HMCG78AGBRA190N, Disk: 2 x 1024GB SAMSUNG MZVL21T0HCLR-00BH1, Graphics: NVIDIA RTX A4000 16GB, Audio: NVIDIA GA104 HD Audio, Monitor: ASUS VP28U, Network: Realtek RTL8111/8168/8411

	OS: CachyOS rolling, Kernel: 6.7.2-1-cachyos (x86_64), Desktop: GNOME Shell 45.3, Display Server: X Server 1.21.1.11, Display Driver: NVIDIA 545.29.06, OpenGL: 4.6.0, Compiler: GCC 13.2.1 20230801 + Clang 16.0.6 + LLVM 16.0.6, File-System: xfs, Screen Resolution: 3840x2160

x86-64-v3: 

	Processor: AMD Ryzen Threadripper PRO 7995WX 96-Cores @ 5.19GHz (96 Cores / 192 Threads), Motherboard: HP Z6 G5 A Workstation 8B24 (U65 Ver. 01.01.04 BIOS), Chipset: AMD Device 14a4, Memory: 8 x 16GB DDR5-5200MT/s Hynix HMCG78AGBRA190N, Disk: 2 x 1024GB SAMSUNG MZVL21T0HCLR-00BH1, Graphics: NVIDIA RTX A4000 16GB, Audio: NVIDIA GA104 HD Audio, Monitor: ASUS VP28U, Network: Realtek RTL8111/8168/8411

	OS: CachyOS rolling, Kernel: 6.7.2-2-cachyos (x86_64), Desktop: GNOME Shell 45.3, Display Server: X Server 1.21.1.11, Display Driver: NVIDIA 545.29.06, OpenGL: 4.6.0, Compiler: GCC 13.2.1 20231026 + Clang 16.0.6 + LLVM 16.0.6, File-System: xfs, Screen Resolution: 3840x2160

x86-64-v4: 

	Processor: AMD Ryzen Threadripper PRO 7995WX 96-Cores @ 5.19GHz (96 Cores / 192 Threads), Motherboard: HP Z6 G5 A Workstation 8B24 (U65 Ver. 01.01.04 BIOS), Chipset: AMD Device 14a4, Memory: 8 x 16GB DDR5-5200MT/s Hynix HMCG78AGBRA190N, Disk: 2 x 1024GB SAMSUNG MZVL21T0HCLR-00BH1, Graphics: NVIDIA RTX A4000 16GB, Audio: NVIDIA GA104 HD Audio, Monitor: ASUS VP28U, Network: Realtek RTL8111/8168/8411

	OS: CachyOS rolling, Kernel: 6.7.2-2-cachyos (x86_64), Desktop: GNOME Shell 45.3, Display Server: X Server 1.21.1.11, Display Driver: NVIDIA 545.29.06, OpenGL: 4.6.0, Compiler: GCC 13.2.1 20240118 + Clang 16.0.6 + LLVM 16.0.6, File-System: xfs, Screen Resolution: 3840x2160


TensorFlow 2.12
Device: CPU - Batch Size: 512 - Model: VGG-16
images/sec > Higher Is Better
No v3 or v4 . 72.40 |==========================================================
x86-64-v3 ... 72.45 |==========================================================
x86-64-v4 ... 72.52 |==========================================================


G'MIC
Test: 3D Elevated Function In Random Colors, 100 Times
Seconds < Lower Is Better
No v3 or v4 . 564.95 |=========================================================
x86-64-v4 ... 560.94 |=========================================================


OSPRay 2.12
Benchmark: particle_volume/pathtracer/real_time
Items Per Second > Higher Is Better
No v3 or v4 . 173.25 |====================================================
x86-64-v3 ... 181.64 |=======================================================
x86-64-v4 ... 189.67 |=========================================================


TensorFlow 2.12
Device: CPU - Batch Size: 512 - Model: ResNet-50
images/sec > Higher Is Better
No v3 or v4 . 132.32 |=========================================================
x86-64-v3 ... 132.42 |=========================================================
x86-64-v4 ... 132.30 |=========================================================


GNU Radio
Test: Hilbert Transform
MiB/s > Higher Is Better
No v3 or v4 . 666.1 |==========================================================
x86-64-v3 ... 648.0 |========================================================
x86-64-v4 ... 655.7 |=========================================================


GNU Radio
Test: FM Deemphasis Filter
MiB/s > Higher Is Better
No v3 or v4 . 641.3 |=========================================
x86-64-v3 ... 879.2 |========================================================
x86-64-v4 ... 911.5 |==========================================================


GNU Radio
Test: IIR Filter
MiB/s > Higher Is Better
No v3 or v4 . 391.9 |==========================================================
x86-64-v3 ... 388.0 |=========================================================
x86-64-v4 ... 385.5 |=========================================================


GNU Radio
Test: FIR Filter
MiB/s > Higher Is Better
No v3 or v4 . 1101.5 |==============================================
x86-64-v3 ... 1248.0 |=====================================================
x86-64-v4 ... 1353.9 |=========================================================


GNU Radio
Test: Signal Source (Cosine)
MiB/s > Higher Is Better
No v3 or v4 . 4535.0 |========================================================
x86-64-v3 ... 3813.4 |===============================================
x86-64-v4 ... 4635.1 |=========================================================


GNU Radio
Test: Five Back to Back FIR Filters
MiB/s > Higher Is Better
No v3 or v4 . 618.7 |================================================
x86-64-v3 ... 744.1 |==========================================================
x86-64-v4 ... 748.6 |==========================================================


Stockfish 15
Total Time
Nodes Per Second > Higher Is Better
No v3 or v4 . 312914005 |=====================================================
x86-64-v3 ... 310805004 |=====================================================
x86-64-v4 ... 316348838 |======================================================


WireGuard + Linux Networking Stack Stress Test

Seconds < Lower Is Better
No v3 or v4 . 224.41 |========================================================
x86-64-v3 ... 223.44 |========================================================
x86-64-v4 ... 226.94 |=========================================================


Renaissance 0.14
Test: Scala Dotty
ms < Lower Is Better
No v3 or v4 . 537.2 |========================================================
x86-64-v3 ... 541.3 |=========================================================
x86-64-v4 ... 552.5 |==========================================================


DaCapo Benchmark 23.11
Java Test: Tradebeans
msec < Lower Is Better
No v3 or v4 . 9696  |===================================================
x86-64-v3 ... 9391  |==================================================
x86-64-v4 ... 10971 |==========================================================


OpenSSL
Algorithm: SHA256
byte/s > Higher Is Better
No v3 or v4 . 129865031843 |==================================================
x86-64-v3 ... 131490753083 |===================================================
x86-64-v4 ... 131972819647 |===================================================


OpenSSL
Algorithm: SHA512
byte/s > Higher Is Better
No v3 or v4 . 42903236637 |===================================================
x86-64-v3 ... 43384654080 |====================================================
x86-64-v4 ... 43303214713 |====================================================


Zstd Compression
Compression Level: 3 - Decompression Speed
MB/s > Higher Is Better
No v3 or v4 . 5740.2 |=========================================================
x86-64-v3 ... 5653.0 |========================================================
x86-64-v4 ... 5712.1 |=========================================================


Zstd Compression
Compression Level: 3 - Compression Speed
MB/s > Higher Is Better
No v3 or v4 . 10052.0 |========================================================
x86-64-v3 ... 9847.8  |=======================================================
x86-64-v4 ... 9635.5  |======================================================


Timed LLVM Compilation 16.0
Build System: Unix Makefiles
Seconds < Lower Is Better
No v3 or v4 . 183.80 |=========================================================
x86-64-v3 ... 162.58 |==================================================
x86-64-v4 ... 161.96 |==================================================


Speedb 2.7
Test: Read While Writing
Op/s > Higher Is Better
No v3 or v4 . 15767159 |=====================================================
x86-64-v3 ... 16312709 |=======================================================
x86-64-v4 ... 15742769 |=====================================================


Selenium
Benchmark: Jetstream 2 - Browser: Firefox
Score > Higher Is Better
No v3 or v4 . 215.57 |=========================================================
x86-64-v3 ... 215.81 |=========================================================
x86-64-v4 ... 217.30 |=========================================================


OSPRay 2.12
Benchmark: particle_volume/scivis/real_time
Items Per Second > Higher Is Better
No v3 or v4 . 25.98 |==========================================================
x86-64-v3 ... 25.54 |=========================================================
x86-64-v4 ... 25.57 |=========================================================


Blender
Blend File: Barbershop - Compute: CPU-Only
Seconds < Lower Is Better
No v3 or v4 . 130.45 |=========================================================
x86-64-v3 ... 130.71 |=========================================================
x86-64-v4 ... 130.89 |=========================================================


Timed Linux Kernel Compilation 6.1
Build: defconfig
Seconds < Lower Is Better
No v3 or v4 . 28.94 |==========================================================
x86-64-v3 ... 27.34 |=======================================================
x86-64-v4 ... 27.58 |=======================================================


Apache Cassandra 4.1.3
Test: Writes
Op/s > Higher Is Better
No v3 or v4 . 310529 |=========================================================
x86-64-v3 ... 311540 |=========================================================
x86-64-v4 ... 309725 |=========================================================


G'MIC
Test: 2D Function Plotting, 1000 Times
Seconds < Lower Is Better
No v3 or v4 . 122.04 |=========================================================
x86-64-v4 ... 103.54 |================================================


Timed LLVM Compilation 16.0
Build System: Ninja
Seconds < Lower Is Better
No v3 or v4 . 116.79 |=========================================================
x86-64-v3 ... 103.71 |===================================================
x86-64-v4 ... 104.21 |===================================================


Zstd Compression
Compression Level: 19 - Decompression Speed
MB/s > Higher Is Better
No v3 or v4 . 4882.5 |========================================================
x86-64-v3 ... 4977.7 |=========================================================
x86-64-v4 ... 4977.0 |=========================================================


Zstd Compression
Compression Level: 19 - Compression Speed
MB/s > Higher Is Better
No v3 or v4 . 135.3 |==========================================================
x86-64-v3 ... 132.8 |=========================================================
x86-64-v4 ... 131.6 |========================================================


Renaissance 0.14
Test: ALS Movie Lens
ms < Lower Is Better
No v3 or v4 . 4744.5 |=========================================================
x86-64-v3 ... 4768.6 |=========================================================
x86-64-v4 ... 4742.2 |=========================================================


Appleseed 2.0 Beta
Scene: Material Tester
Seconds < Lower Is Better
No v3 or v4 . 149.64 |========================================================
x86-64-v3 ... 152.29 |=========================================================
x86-64-v4 ... 151.00 |=========================================================


OSPRay 2.12
Benchmark: particle_volume/ao/real_time
Items Per Second > Higher Is Better
No v3 or v4 . 25.58 |==========================================================
x86-64-v3 ... 25.07 |=========================================================
x86-64-v4 ... 25.04 |=========================================================


Chaos Group V-RAY 5.02
Mode: CPU
vsamples > Higher Is Better
No v3 or v4 . 106099 |========================================================
x86-64-v3 ... 107830 |=========================================================
x86-64-v4 ... 106670 |========================================================


Zstd Compression
Compression Level: 19, Long Mode - Decompression Speed
MB/s > Higher Is Better
No v3 or v4 . 4843.2 |========================================================
x86-64-v3 ... 4833.7 |========================================================
x86-64-v4 ... 4903.2 |=========================================================


Zstd Compression
Compression Level: 19, Long Mode - Compression Speed
MB/s > Higher Is Better
No v3 or v4 . 63.4 |==========================================================
x86-64-v3 ... 64.8 |===========================================================
x86-64-v4 ... 62.5 |=========================================================


VVenC 1.9
Video Input: Bosphorus 4K - Video Preset: Fast
Frames Per Second > Higher Is Better
No v3 or v4 . 9.292 |==========================================================
x86-64-v3 ... 9.269 |==========================================================
x86-64-v4 ... 9.234 |==========================================================


Appleseed 2.0 Beta
Scene: Emily
Seconds < Lower Is Better
No v3 or v4 . 97.67 |==========================================================
x86-64-v3 ... 98.30 |==========================================================
x86-64-v4 ... 97.96 |==========================================================


QuantLib 1.32
Configuration: Multi-Threaded
MFLOPS > Higher Is Better
No v3 or v4 . 319756.0 |=======================================================
x86-64-v3 ... 314870.6 |======================================================
x86-64-v4 ... 316288.2 |======================================================


Speedb 2.7
Test: Update Random
Op/s > Higher Is Better
No v3 or v4 . 508411 |=========================================================
x86-64-v3 ... 499920 |========================================================
x86-64-v4 ... 501800 |========================================================


Speedb 2.7
Test: Read Random Write Random
Op/s > Higher Is Better
No v3 or v4 . 2479906 |=======================================================
x86-64-v3 ... 2515641 |========================================================
x86-64-v4 ... 2515862 |========================================================


Speedb 2.7
Test: Random Read
Op/s > Higher Is Better
No v3 or v4 . 522454918 |=====================================================
x86-64-v3 ... 531958100 |======================================================
x86-64-v4 ... 529243821 |======================================================


PyPerformance 1.0.0
Benchmark: python_startup
Milliseconds < Lower Is Better
No v3 or v4 . 6.92 |==========================================================
x86-64-v3 ... 6.97 |===========================================================
x86-64-v4 ... 6.98 |===========================================================


DaCapo Benchmark 23.11
Java Test: Apache Cassandra
msec < Lower Is Better
No v3 or v4 . 5653 |===========================================================
x86-64-v3 ... 5679 |===========================================================
x86-64-v4 ... 5686 |===========================================================


OpenSCAD
Render: Pistol
Seconds < Lower Is Better
No v3 or v4 . 48.13 |==========================================================
x86-64-v3 ... 48.44 |==========================================================
x86-64-v4 ... 46.56 |========================================================


OpenSCAD
Render: Leonardo Phone Case Slim
Seconds < Lower Is Better
No v3 or v4 . 14.73 |=======================================================
x86-64-v3 ... 15.48 |==========================================================
x86-64-v4 ... 14.65 |=======================================================


OpenSCAD
Render: Mini-ITX Case
Seconds < Lower Is Better
No v3 or v4 . 22.62 |=========================================================
x86-64-v3 ... 23.22 |==========================================================
x86-64-v4 ... 22.01 |=======================================================


DaCapo Benchmark 23.11
Java Test: Spring Boot
msec < Lower Is Better
No v3 or v4 . 1891 |===========================================================
x86-64-v3 ... 1894 |===========================================================
x86-64-v4 ... 1880 |===========================================================


OSPRay 2.12
Benchmark: gravity_spheres_volume/dim_512/ao/real_time
Items Per Second > Higher Is Better
No v3 or v4 . 26.24 |==========================================================
x86-64-v3 ... 26.35 |==========================================================
x86-64-v4 ... 26.16 |==========================================================


Llamafile 0.6
Test: mistral-7b-instruct-v0.2.Q8_0 - Acceleration: CPU
Tokens Per Second > Higher Is Better
No v3 or v4 . 18.09 |==========================================================
x86-64-v3 ... 18.08 |==========================================================
x86-64-v4 ... 18.01 |==========================================================


OSPRay 2.12
Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time
Items Per Second > Higher Is Better
No v3 or v4 . 27.36 |==========================================================
x86-64-v3 ... 27.23 |==========================================================
x86-64-v4 ... 27.34 |==========================================================


OSPRay 2.12
Benchmark: gravity_spheres_volume/dim_512/scivis/real_time
Items Per Second > Higher Is Better
No v3 or v4 . 25.52 |==========================================================
x86-64-v3 ... 25.51 |==========================================================
x86-64-v4 ... 25.33 |==========================================================


PyBench 2018-02-16
Total For Average Test Times
Milliseconds < Lower Is Better
No v3 or v4 . 585 |===========================================================
x86-64-v3 ... 578 |==========================================================
x86-64-v4 ... 593 |============================================================


PyPerformance 1.0.0
Benchmark: django_template
Milliseconds < Lower Is Better
No v3 or v4 . 21.7 |===========================================================
x86-64-v3 ... 21.5 |==========================================================
x86-64-v4 ... 21.7 |===========================================================


Z3 Theorem Prover 4.12.1
SMT File: 2.smt2
Seconds < Lower Is Better
No v3 or v4 . 48.13 |========================================================
x86-64-v3 ... 48.64 |=========================================================
x86-64-v4 ... 49.46 |==========================================================


RawTherapee
Total Benchmark Time
Seconds < Lower Is Better
No v3 or v4 . 46.91 |==========================================================
x86-64-v3 ... 46.02 |=========================================================
x86-64-v4 ... 45.54 |========================================================


Renaissance 0.14
Test: Apache Spark ALS
ms < Lower Is Better
No v3 or v4 . 1238.4 |=========================================================
x86-64-v3 ... 1236.5 |=========================================================
x86-64-v4 ... 1237.8 |=========================================================


GEGL
Operation: Cartoon
Seconds < Lower Is Better
No v3 or v4 . 66.66 |==========================================================
x86-64-v4 ... 63.83 |========================================================


Selenium
Benchmark: Speedometer - Browser: Firefox
Runs Per Minute > Higher Is Better
No v3 or v4 . 320 |==========================================================
x86-64-v3 ... 325 |===========================================================
x86-64-v4 ... 329 |============================================================


Blender
Blend File: Pabellon Barcelona - Compute: CPU-Only
Seconds < Lower Is Better
No v3 or v4 . 39.72 |==========================================================
x86-64-v3 ... 39.79 |==========================================================
x86-64-v4 ... 39.92 |==========================================================


DaCapo Benchmark 23.11
Java Test: BioJava Biological Data Framework
msec < Lower Is Better
No v3 or v4 . 4634 |==========================================================
x86-64-v3 ... 4678 |===========================================================
x86-64-v4 ... 4615 |==========================================================


VVenC 1.9
Video Input: Bosphorus 4K - Video Preset: Faster
Frames Per Second > Higher Is Better
No v3 or v4 . 16.41 |==========================================================
x86-64-v3 ... 16.23 |=========================================================
x86-64-v4 ... 16.24 |=========================================================


Git
Time To Complete Common Git Commands
Seconds < Lower Is Better
No v3 or v4 . 36.06 |==========================================================
x86-64-v3 ... 34.65 |========================================================
x86-64-v4 ... 35.02 |========================================================


Zstd Compression
Compression Level: 8, Long Mode - Decompression Speed
MB/s > Higher Is Better
No v3 or v4 . 6298.1 |========================================================
x86-64-v3 ... 6284.1 |========================================================
x86-64-v4 ... 6368.4 |=========================================================


Zstd Compression
Compression Level: 8, Long Mode - Compression Speed
MB/s > Higher Is Better
No v3 or v4 . 1089.9 |=========================================================
x86-64-v3 ... 1088.1 |=========================================================
x86-64-v4 ... 1086.6 |=========================================================


Zstd Compression
Compression Level: 3, Long Mode - Decompression Speed
MB/s > Higher Is Better
No v3 or v4 . 6024.0 |========================================================
x86-64-v3 ... 6132.1 |=========================================================
x86-64-v4 ... 6137.1 |=========================================================


Zstd Compression
Compression Level: 3, Long Mode - Compression Speed
MB/s > Higher Is Better
No v3 or v4 . 982.6 |=========================================================
x86-64-v3 ... 996.9 |==========================================================
x86-64-v4 ... 995.9 |==========================================================


Zstd Compression
Compression Level: 8 - Decompression Speed
MB/s > Higher Is Better
No v3 or v4 . 5966.2 |========================================================
x86-64-v3 ... 6035.4 |=========================================================
x86-64-v4 ... 5979.5 |========================================================


Zstd Compression
Compression Level: 8 - Compression Speed
MB/s > Higher Is Better
No v3 or v4 . 6220.8 |=========================================================
x86-64-v3 ... 6154.3 |========================================================
x86-64-v4 ... 6109.8 |========================================================


PyPerformance 1.0.0
Benchmark: raytrace
Milliseconds < Lower Is Better
No v3 or v4 . 183 |============================================================
x86-64-v3 ... 183 |============================================================
x86-64-v4 ... 179 |===========================================================


PHPBench 0.8.1
PHP Benchmark Suite
Score > Higher Is Better
No v3 or v4 . 890814  |========================
x86-64-v3 ... 2058126 |========================================================
x86-64-v4 ... 2021067 |=======================================================


7-Zip Compression 22.01
Test: Decompression Rating
MIPS > Higher Is Better
No v3 or v4 . 664597 |=========================================================
x86-64-v3 ... 662490 |=========================================================
x86-64-v4 ... 652847 |========================================================


7-Zip Compression 22.01
Test: Compression Rating
MIPS > Higher Is Better
No v3 or v4 . 715590 |=========================================================
x86-64-v3 ... 715823 |=========================================================
x86-64-v4 ... 719120 |=========================================================


Selenium
Benchmark: PSPDFKit WASM - Browser: Firefox
Score < Lower Is Better
No v3 or v4 . 2369 |===========================================================
x86-64-v3 ... 2352 |===========================================================
x86-64-v4 ... 2351 |===========================================================


DaCapo Benchmark 23.11
Java Test: Avrora AVR Simulation Framework
msec < Lower Is Better
No v3 or v4 . 5140 |=======================================================
x86-64-v3 ... 5503 |===========================================================
x86-64-v4 ... 5271 |=========================================================


DaCapo Benchmark 23.11
Java Test: Eclipse
msec < Lower Is Better
No v3 or v4 . 7934 |==========================================================
x86-64-v3 ... 8074 |===========================================================
x86-64-v4 ... 8004 |==========================================================


Llamafile 0.6
Test: wizardcoder-python-34b-v1.0.Q6_K - Acceleration: CPU
Tokens Per Second > Higher Is Better
No v3 or v4 . 7.64 |===========================================================
x86-64-v3 ... 7.64 |===========================================================
x86-64-v4 ... 7.64 |===========================================================


GEGL
Operation: Wavelet Blur
Seconds < Lower Is Better
No v3 or v4 . 44.48 |==========================================================
x86-64-v4 ... 43.19 |========================================================


PyPerformance 1.0.0
Benchmark: 2to3
Milliseconds < Lower Is Better
No v3 or v4 . 151 |============================================================
x86-64-v3 ... 151 |============================================================
x86-64-v4 ... 150 |============================================================


DaCapo Benchmark 23.11
Java Test: Tradesoap
msec < Lower Is Better
No v3 or v4 . 6509 |=========================================================
x86-64-v3 ... 6684 |===========================================================
x86-64-v4 ... 6705 |===========================================================


Blender
Blend File: Classroom - Compute: CPU-Only
Seconds < Lower Is Better
No v3 or v4 . 27.20 |==========================================================
x86-64-v3 ... 27.11 |==========================================================
x86-64-v4 ... 27.31 |==========================================================


Renaissance 0.14
Test: Random Forest
ms < Lower Is Better
No v3 or v4 . 469.9 |==========================================================
x86-64-v3 ... 471.9 |==========================================================
x86-64-v4 ... 471.0 |==========================================================


PyPerformance 1.0.0
Benchmark: regex_compile
Milliseconds < Lower Is Better
No v3 or v4 . 68.1 |===========================================================
x86-64-v3 ... 68.1 |===========================================================
x86-64-v4 ... 67.7 |===========================================================


GEGL
Operation: Rotate 90 Degrees
Seconds < Lower Is Better
No v3 or v4 . 40.07 |==========================================================
x86-64-v4 ... 36.97 |======================================================


TensorFlow 2.12
Device: CPU - Batch Size: 1 - Model: ResNet-50
images/sec > Higher Is Better
No v3 or v4 . 4.77 |===========================================================
x86-64-v3 ... 4.79 |===========================================================
x86-64-v4 ... 4.74 |==========================================================


PyPerformance 1.0.0
Benchmark: go
Milliseconds < Lower Is Better
No v3 or v4 . 88.0 |===========================================================
x86-64-v3 ... 86.4 |==========================================================
x86-64-v4 ... 85.7 |=========================================================


DaCapo Benchmark 23.11
Java Test: H2 Database Engine
msec < Lower Is Better
No v3 or v4 . 2357 |===========================================================
x86-64-v3 ... 2338 |==========================================================
x86-64-v4 ... 2368 |===========================================================


DaCapo Benchmark 23.11
Java Test: H2O In-Memory Platform For Machine Learning
msec < Lower Is Better
No v3 or v4 . 3972 |=========================================================
x86-64-v3 ... 3912 |========================================================
x86-64-v4 ... 4126 |===========================================================


Java SciMark 2.2
Computational Test: Composite
Mflops > Higher Is Better
No v3 or v4 . 5396.91 |=======================================================
x86-64-v3 ... 5413.22 |=======================================================
x86-64-v4 ... 5542.47 |========================================================


PyPerformance 1.0.0
Benchmark: pathlib
Milliseconds < Lower Is Better
No v3 or v4 . 12.5 |==========================================================
x86-64-v3 ... 12.7 |===========================================================
x86-64-v4 ... 12.8 |===========================================================


librsvg
Operation: SVG Files To PNG
Seconds < Lower Is Better
No v3 or v4 . 25.37 |==========================================================
x86-64-v3 ... 24.40 |========================================================
x86-64-v4 ... 24.51 |========================================================


GEGL
Operation: Color Enhance
Seconds < Lower Is Better
No v3 or v4 . 37.35 |==========================================================
x86-64-v4 ... 34.79 |======================================================


Appleseed 2.0 Beta
Scene: Disney Material
Seconds < Lower Is Better
No v3 or v4 . 37.01 |==========================================================
x86-64-v3 ... 36.86 |==========================================================
x86-64-v4 ... 37.18 |==========================================================


DaCapo Benchmark 23.11
Java Test: Apache Lucene Search Index
msec < Lower Is Better
No v3 or v4 . 2696 |=========================================================
x86-64-v3 ... 2800 |===========================================================
x86-64-v4 ... 2731 |==========================================================


PyPerformance 1.0.0
Benchmark: json_loads
Milliseconds < Lower Is Better
No v3 or v4 . 12.2 |===========================================================
x86-64-v3 ... 11.9 |==========================================================
x86-64-v4 ... 11.9 |==========================================================


Selenium
Benchmark: Kraken - Browser: Firefox
ms < Lower Is Better
No v3 or v4 . 494.6 |==========================================================
x86-64-v3 ... 494.3 |==========================================================
x86-64-v4 ... 487.5 |=========================================================


DaCapo Benchmark 23.11
Java Test: Apache Kafka
msec < Lower Is Better
No v3 or v4 . 6268 |==========================================================
x86-64-v3 ... 6364 |===========================================================
x86-64-v4 ... 6335 |===========================================================


PyPerformance 1.0.0
Benchmark: nbody
Milliseconds < Lower Is Better
No v3 or v4 . 62.4 |===========================================================
x86-64-v3 ... 62.9 |===========================================================
x86-64-v4 ... 62.9 |===========================================================


DaCapo Benchmark 23.11
Java Test: jMonkeyEngine
msec < Lower Is Better
No v3 or v4 . 6882 |===========================================================
x86-64-v3 ... 6886 |===========================================================
x86-64-v4 ... 6880 |===========================================================


PyPerformance 1.0.0
Benchmark: pickle_pure_python
Milliseconds < Lower Is Better
No v3 or v4 . 176 |===========================================================
x86-64-v3 ... 176 |===========================================================
x86-64-v4 ... 180 |============================================================


Blender
Blend File: Fishy Cat - Compute: CPU-Only
Seconds < Lower Is Better
No v3 or v4 . 19.32 |==========================================================
x86-64-v3 ... 19.34 |==========================================================
x86-64-v4 ... 19.39 |==========================================================


Z3 Theorem Prover 4.12.1
SMT File: 1.smt2
Seconds < Lower Is Better
No v3 or v4 . 20.65 |==========================================================
x86-64-v3 ... 20.45 |=========================================================
x86-64-v4 ... 20.65 |==========================================================


Renaissance 0.14
Test: Apache Spark Bayes
ms < Lower Is Better
No v3 or v4 . 440.3 |==========================================================
x86-64-v3 ... 437.2 |==========================================================
x86-64-v4 ... 431.5 |=========================================================


GEGL
Operation: Antialias
Seconds < Lower Is Better
No v3 or v4 . 29.32 |==========================================================
x86-64-v4 ... 27.83 |=======================================================


uvg266 0.4.1
Video Input: Bosphorus 4K - Video Preset: Slow
Frames Per Second > Higher Is Better
No v3 or v4 . 31.44 |==========================================================
x86-64-v3 ... 31.53 |==========================================================
x86-64-v4 ... 31.34 |==========================================================


SVT-AV1 1.8
Encoder Mode: Preset 4 - Input: Bosphorus 4K
Frames Per Second > Higher Is Better
No v3 or v4 . 9.877 |==========================================================
x86-64-v3 ... 9.942 |==========================================================
x86-64-v4 ... 9.907 |==========================================================


GEGL
Operation: Scale
Seconds < Lower Is Better
No v3 or v4 . 8.541 |==========================================================
x86-64-v4 ... 7.873 |=====================================================


PyPerformance 1.0.0
Benchmark: float
Milliseconds < Lower Is Better
No v3 or v4 . 44.9 |===========================================================
x86-64-v3 ... 45.0 |===========================================================
x86-64-v4 ... 45.2 |===========================================================


uvg266 0.4.1
Video Input: Bosphorus 4K - Video Preset: Medium
Frames Per Second > Higher Is Better
No v3 or v4 . 35.10 |==========================================================
x86-64-v3 ... 35.21 |==========================================================
x86-64-v4 ... 35.11 |==========================================================


x265 3.4
Video Input: Bosphorus 4K
Frames Per Second > Higher Is Better
No v3 or v4 . 35.18 |=========================================================
x86-64-v3 ... 35.55 |==========================================================
x86-64-v4 ... 35.27 |==========================================================


PyPerformance 1.0.0
Benchmark: crypto_pyaes
Milliseconds < Lower Is Better
No v3 or v4 . 43.1 |===========================================================
x86-64-v3 ... 43.1 |===========================================================
x86-64-v4 ... 43.2 |===========================================================


PyPerformance 1.0.0
Benchmark: chaos
Milliseconds < Lower Is Better
No v3 or v4 . 41.2 |==========================================================
x86-64-v3 ... 41.9 |===========================================================
x86-64-v4 ... 41.4 |==========================================================


Blender
Blend File: BMW27 - Compute: CPU-Only
Seconds < Lower Is Better
No v3 or v4 . 15.26 |==========================================================
x86-64-v3 ... 15.26 |==========================================================
x86-64-v4 ... 15.33 |==========================================================


DaCapo Benchmark 23.11
Java Test: GraphChi
msec < Lower Is Better
No v3 or v4 . 2229 |==========================================================
x86-64-v3 ... 2258 |===========================================================
x86-64-v4 ... 2178 |=========================================================


GEGL
Operation: Tile Glass
Seconds < Lower Is Better
No v3 or v4 . 23.75 |==========================================================
x86-64-v4 ... 22.06 |======================================================


GEGL
Operation: Reflect
Seconds < Lower Is Better
No v3 or v4 . 23.51 |==========================================================
x86-64-v4 ... 22.05 |======================================================


Selenium
Benchmark: WASM collisionDetection - Browser: Firefox
ms < Lower Is Better
No v3 or v4 . 269.0 |=========================================================
x86-64-v3 ... 271.5 |==========================================================
x86-64-v4 ... 266.2 |=========================================================


SVT-AV1 1.8
Encoder Mode: Preset 13 - Input: Bosphorus 4K
Frames Per Second > Higher Is Better
No v3 or v4 . 235.26 |=======================================================
x86-64-v3 ... 236.44 |========================================================
x86-64-v4 ... 242.35 |=========================================================


Crafty 25.2
Elapsed Time
Nodes Per Second > Higher Is Better
No v3 or v4 . 13921913 |=======================================================
x86-64-v3 ... 13790708 |======================================================
x86-64-v4 ... 13918436 |=======================================================


DaCapo Benchmark 23.11
Java Test: Jython
msec < Lower Is Better
No v3 or v4 . 3817 |==========================================================
x86-64-v3 ... 3829 |==========================================================
x86-64-v4 ... 3883 |===========================================================


Kvazaar 2.2
Video Input: Bosphorus 4K - Video Preset: Slow
Frames Per Second > Higher Is Better
No v3 or v4 . 43.86 |==========================================================
x86-64-v3 ... 43.91 |==========================================================
x86-64-v4 ... 44.20 |==========================================================


Kvazaar 2.2
Video Input: Bosphorus 4K - Video Preset: Medium
Frames Per Second > Higher Is Better
No v3 or v4 . 44.05 |==========================================================
x86-64-v3 ... 44.07 |==========================================================
x86-64-v4 ... 44.15 |==========================================================


Llamafile 0.6
Test: llava-v1.5-7b-q4 - Acceleration: CPU
Tokens Per Second > Higher Is Better
No v3 or v4 . 27.23 |==========================================================
x86-64-v3 ... 27.15 |==========================================================
x86-64-v4 ... 27.27 |==========================================================


GIMP 2.10.36
Test: unsharp-mask
Seconds < Lower Is Better
No v3 or v4 . 16.79 |==========================================================
x86-64-v4 ... 16.82 |==========================================================


TensorFlow 2.12
Device: CPU - Batch Size: 1 - Model: VGG-16
images/sec > Higher Is Better
No v3 or v4 . 12.23 |==========================================================
x86-64-v3 ... 12.16 |==========================================================
x86-64-v4 ... 12.08 |=========================================================


G'MIC
Test: Plotting Isosurface Of A 3D Volume, 1000 Times
Seconds < Lower Is Better
No v3 or v4 . 10.63 |=========================================================
x86-64-v4 ... 10.75 |==========================================================


DaCapo Benchmark 23.11
Java Test: Zxing 1D/2D Barcode Image Processing
msec < Lower Is Better
No v3 or v4 . 354 |============================================================
x86-64-v3 ... 334 |=========================================================
x86-64-v4 ... 348 |===========================================================


DaCapo Benchmark 23.11
Java Test: PMD Source Code Analyzer
msec < Lower Is Better
No v3 or v4 . 1690 |===========================================================
x86-64-v3 ... 1679 |===========================================================
x86-64-v4 ... 1649 |==========================================================


GIMP 2.10.36
Test: resize
Seconds < Lower Is Better
No v3 or v4 . 15.31 |=========================================================
x86-64-v4 ... 15.51 |==========================================================


Selenium
Benchmark: WASM imageConvolute - Browser: Firefox
ms < Lower Is Better
No v3 or v4 . 26.9 |===========================================================
x86-64-v3 ... 27.0 |===========================================================
x86-64-v4 ... 26.9 |===========================================================


GIMP 2.10.36
Test: rotate
Seconds < Lower Is Better
No v3 or v4 . 10.19 |=========================================================
x86-64-v4 ... 10.36 |==========================================================


GIMP 2.10.36
Test: auto-levels
Seconds < Lower Is Better
No v3 or v4 . 13.01 |==========================================================
x86-64-v4 ... 13.04 |==========================================================


uvg266 0.4.1
Video Input: Bosphorus 4K - Video Preset: Very Fast
Frames Per Second > Higher Is Better
No v3 or v4 . 70.22 |==========================================================
x86-64-v3 ... 70.49 |==========================================================
x86-64-v4 ... 69.74 |=========================================================


uvg266 0.4.1
Video Input: Bosphorus 4K - Video Preset: Super Fast
Frames Per Second > Higher Is Better
No v3 or v4 . 71.04 |==========================================================
x86-64-v3 ... 71.20 |==========================================================
x86-64-v4 ... 70.52 |=========================================================


DaCapo Benchmark 23.11
Java Test: Apache Xalan XSLT
msec < Lower Is Better
No v3 or v4 . 802 |===========================================================
x86-64-v3 ... 812 |============================================================
x86-64-v4 ... 800 |===========================================================


uvg266 0.4.1
Video Input: Bosphorus 4K - Video Preset: Ultra Fast
Frames Per Second > Higher Is Better
No v3 or v4 . 72.90 |==========================================================
x86-64-v3 ... 73.40 |==========================================================
x86-64-v4 ... 72.72 |=========================================================


DaCapo Benchmark 23.11
Java Test: Apache Lucene Search Engine
msec < Lower Is Better
No v3 or v4 . 1490 |===========================================================
x86-64-v3 ... 1486 |===========================================================
x86-64-v4 ... 1478 |===========================================================


DaCapo Benchmark 23.11
Java Test: Apache Tomcat
msec < Lower Is Better
No v3 or v4 . 1357 |=========================================================
x86-64-v3 ... 1408 |===========================================================
x86-64-v4 ... 1401 |===========================================================


SVT-AV1 1.8
Encoder Mode: Preset 8 - Input: Bosphorus 4K
Frames Per Second > Higher Is Better
No v3 or v4 . 103.39 |=========================================================
x86-64-v3 ... 104.08 |=========================================================
x86-64-v4 ... 102.13 |========================================================


SVT-AV1 1.8
Encoder Mode: Preset 12 - Input: Bosphorus 4K
Frames Per Second > Higher Is Better
No v3 or v4 . 247.27 |=========================================================
x86-64-v3 ... 238.57 |=======================================================
x86-64-v4 ... 245.07 |========================================================


Kvazaar 2.2
Video Input: Bosphorus 4K - Video Preset: Very Fast
Frames Per Second > Higher Is Better
No v3 or v4 . 85.95 |==========================================================
x86-64-v3 ... 86.49 |==========================================================
x86-64-v4 ... 86.48 |==========================================================


GNU Octave Benchmark 8.4.0

Seconds < Lower Is Better
No v3 or v4 . 4.435 |==========================================================
x86-64-v4 ... 4.015 |=====================================================


Kvazaar 2.2
Video Input: Bosphorus 4K - Video Preset: Super Fast
Frames Per Second > Higher Is Better
No v3 or v4 . 89.07 |==========================================================
x86-64-v3 ... 89.27 |==========================================================
x86-64-v4 ... 89.69 |==========================================================


Darktable 4.6.0
Test: Server Room - Acceleration: CPU-only
Seconds < Lower Is Better
No v3 or v4 . 1.366 |==========================================================
x86-64-v4 ... 1.352 |=========================================================


Kvazaar 2.2
Video Input: Bosphorus 4K - Video Preset: Ultra Fast
Frames Per Second > Higher Is Better
No v3 or v4 . 92.79 |==========================================================
x86-64-v3 ... 93.31 |==========================================================
x86-64-v4 ... 92.84 |==========================================================


ctx_clock
Context Switch Time
Clocks < Lower Is Better
No v3 or v4 . 208 |===========================================================
x86-64-v3 ... 200 |=========================================================
x86-64-v4 ... 210 |============================================================


DaCapo Benchmark 23.11
Java Test: Batik SVG Toolkit
msec < Lower Is Better
No v3 or v4 . 1177 |===========================================================
x86-64-v3 ... 1173 |===========================================================
x86-64-v4 ... 1133 |=========================================================


GEGL
Operation: Crop
Seconds < Lower Is Better
No v3 or v4 . 7.410 |==========================================================
x86-64-v4 ... 6.254 |=================================================


Darktable 4.6.0
Test: Server Rack - Acceleration: CPU-only
Seconds < Lower Is Better
No v3 or v4 . 0.162 |==========================================================
x86-64-v4 ... 0.154 |=======================================================


OpenSCAD
Render: Projector Mount Swivel
Seconds < Lower Is Better
No v3 or v4 . 3.881 |==========================================================
x86-64-v3 ... 3.853 |==========================================================
x86-64-v4 ... 3.801 |=========================================================


Darktable 4.6.0
Test: Masskrug - Acceleration: CPU-only
Seconds < Lower Is Better
No v3 or v4 . 1.247 |==========================================================
x86-64-v4 ... 1.187 |=======================================================


Darktable 4.6.0
Test: Boat - Acceleration: CPU-only
Seconds < Lower Is Better
No v3 or v4 . 1.399 |==========================================================
x86-64-v4 ... 1.329 |=======================================================


DaCapo Benchmark 23.11
Java Test: FOP Print Formatter
msec < Lower Is Better
No v3 or v4 . 401 |============================================================
x86-64-v3 ... 396 |===========================================================
x86-64-v4 ... 392 |===========================================================


OpenSCAD
Render: Retro Car
Seconds < Lower Is Better
No v3 or v4 . 2.368 |==========================================================
x86-64-v3 ... 2.319 |=========================================================
x86-64-v4 ... 2.304 |========================================================


Java SciMark 2.2
Computational Test: Jacobi Successive Over-Relaxation
Mflops > Higher Is Better
No v3 or v4 . 2370.81 |========================================================
x86-64-v3 ... 2373.57 |========================================================
x86-64-v4 ... 2373.57 |========================================================


Java SciMark 2.2
Computational Test: Dense LU Matrix Factorization
Mflops > Higher Is Better
No v3 or v4 . 18369.55 |=====================================================
x86-64-v3 ... 18453.29 |======================================================
x86-64-v4 ... 18931.45 |=======================================================


Java SciMark 2.2
Computational Test: Sparse Matrix Multiply
Mflops > Higher Is Better
No v3 or v4 . 3668.67 |======================================================
x86-64-v3 ... 3666.80 |======================================================
x86-64-v4 ... 3781.55 |========================================================


Java SciMark 2.2
Computational Test: Fast Fourier Transform
Mflops > Higher Is Better
No v3 or v4 . 419.23 |=========================================================
x86-64-v3 ... 417.06 |=========================================================
x86-64-v4 ... 418.89 |=========================================================


Java SciMark 2.2
Computational Test: Monte Carlo
Mflops > Higher Is Better
No v3 or v4 . 2156.28 |=======================================================
x86-64-v3 ... 2155.40 |=======================================================
x86-64-v4 ... 2206.89 |========================================================