AMD AOCC 4.0 Benchmarks AMD Ryzen 9 7950X compiler benchmarks by Michael Larabel for a future article. AOCC 4.0: Processor: AMD Ryzen 9 7950X 16-Core @ 5.88GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR X670E HERO (0703 BIOS), Chipset: AMD Device 14d8, Memory: 32GB, Disk: 1000GB Sabrent Rocket 4.0 Plus, Graphics: AMD Radeon RX 6800 16GB (2475/1000MHz), Audio: AMD Navi 21/23, Monitor: ASUS MG28U, Network: Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411 OS: Ubuntu 22.10, Kernel: 6.1.0-060100rc3daily20221103-generic (x86_64), Desktop: GNOME Shell 43.0, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.2.1 (LLVM 15.0.2 DRM 3.49), Vulkan: 1.3.224, Compiler: Clang 14.0.6, File-System: ext4, Screen Resolution: 3840x2160 GCC 12.2: Processor: AMD Ryzen 9 7950X 16-Core @ 5.88GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR X670E HERO (0703 BIOS), Chipset: AMD Device 14d8, Memory: 32GB, Disk: 1000GB Sabrent Rocket 4.0 Plus, Graphics: AMD Radeon RX 6800 16GB (2475/1000MHz), Audio: AMD Navi 21/23, Monitor: ASUS MG28U, Network: Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411 OS: Ubuntu 22.10, Kernel: 6.1.0-060100rc3daily20221103-generic (x86_64), Desktop: GNOME Shell 43.0, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.2.1 (LLVM 15.0.2 DRM 3.49), Vulkan: 1.3.224, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 3840x2160 GCC 13.0 14 Nov: Processor: AMD Ryzen 9 7950X 16-Core @ 5.88GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR X670E HERO (0703 BIOS), Chipset: AMD Device 14d8, Memory: 32GB, Disk: 1000GB Sabrent Rocket 4.0 Plus, Graphics: AMD Radeon RX 6800 16GB (2475/1000MHz), Audio: AMD Navi 21/23, Monitor: ASUS MG28U, Network: Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411 OS: Ubuntu 22.10, Kernel: 6.1.0-060100rc3daily20221103-generic (x86_64), Desktop: GNOME Shell 43.0, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.2.1 (LLVM 15.0.2 DRM 3.49), Vulkan: 1.3.224, Compiler: GCC 13.0.0 20221114 + clang (GCC) 13.0.0 20221114 (experimental), File-System: ext4, Screen Resolution: 3840x2160 LLVM Clang 14: Processor: AMD Ryzen 9 7950X 16-Core @ 5.88GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR X670E HERO (0703 BIOS), Chipset: AMD Device 14d8, Memory: 32GB, Disk: 1000GB Sabrent Rocket 4.0 Plus, Graphics: AMD Radeon RX 6800 16GB (2475/1000MHz), Audio: AMD Navi 21/23, Monitor: ASUS MG28U, Network: Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411 OS: Ubuntu 22.10, Kernel: 6.1.0-060100rc3daily20221103-generic (x86_64), Desktop: GNOME Shell 43.0, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.2.1 (LLVM 15.0.2 DRM 3.49), Vulkan: 1.3.224, Compiler: Clang 14.0.6-2, File-System: ext4, Screen Resolution: 3840x2160 LLVM Clang 15.0.2: Processor: AMD Ryzen 9 7950X 16-Core @ 5.88GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR X670E HERO (0703 BIOS), Chipset: AMD Device 14d8, Memory: 32GB, Disk: 1000GB Sabrent Rocket 4.0 Plus, Graphics: AMD Radeon RX 6800 16GB (2475/1000MHz), Audio: AMD Navi 21/23, Monitor: ASUS MG28U, Network: Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411 OS: Ubuntu 22.10, Kernel: 6.1.0-060100rc3daily20221103-generic (x86_64), Desktop: GNOME Shell 43.0, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.2.1 (LLVM 15.0.2 DRM 3.49), Vulkan: 1.3.224, Compiler: Clang 15.0.2-1, File-System: ext4, Screen Resolution: 3840x2160 AOBench Size: 2048 x 2048 - Total Time Seconds < Lower Is Better AOCC 4.0 .......... 21.58 |==================================================== GCC 12.2 .......... 19.76 |=============================================== GCC 13.0 14 Nov ... 19.90 |================================================ LLVM Clang 14 ..... 21.70 |==================================================== LLVM Clang 15.0.2 . 21.39 |=================================================== AOM AV1 3.5 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K Frames Per Second > Higher Is Better AOCC 4.0 .......... 20.75 |==================================================== GCC 12.2 .......... 19.32 |================================================ GCC 13.0 14 Nov ... 20.72 |==================================================== LLVM Clang 14 ..... 17.76 |============================================= LLVM Clang 15.0.2 . 17.92 |============================================= AOM AV1 3.5 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K Frames Per Second > Higher Is Better AOCC 4.0 .......... 65.44 |================================================== GCC 12.2 .......... 68.24 |==================================================== GCC 13.0 14 Nov ... 68.30 |==================================================== LLVM Clang 14 ..... 64.82 |================================================= LLVM Clang 15.0.2 . 68.71 |==================================================== ASTC Encoder 4.0 Preset: Fast MT/s > Higher Is Better AOCC 4.0 .......... 427.33 |=================================================== GCC 12.2 .......... 289.29 |=================================== GCC 13.0 14 Nov ... 294.57 |=================================== LLVM Clang 14 ..... 404.45 |================================================ LLVM Clang 15.0.2 . 412.21 |================================================= ASTC Encoder 4.0 Preset: Medium MT/s > Higher Is Better AOCC 4.0 .......... 160.96 |=================================================== GCC 12.2 .......... 132.59 |========================================== GCC 13.0 14 Nov ... 135.19 |=========================================== LLVM Clang 14 ..... 155.09 |================================================= LLVM Clang 15.0.2 . 156.45 |================================================== ASTC Encoder 4.0 Preset: Thorough MT/s > Higher Is Better AOCC 4.0 .......... 19.91 |==================================================== GCC 12.2 .......... 19.45 |=================================================== GCC 13.0 14 Nov ... 19.77 |==================================================== LLVM Clang 14 ..... 19.28 |================================================== LLVM Clang 15.0.2 . 19.43 |=================================================== ASTC Encoder 4.0 Preset: Exhaustive MT/s > Higher Is Better AOCC 4.0 .......... 2.0160 |=================================================== GCC 12.2 .......... 2.0071 |================================================== GCC 13.0 14 Nov ... 2.0276 |=================================================== LLVM Clang 14 ..... 1.9814 |================================================== LLVM Clang 15.0.2 . 1.9886 |================================================== C-Blosc 2.3 Test: blosclz shuffle MB/s > Higher Is Better AOCC 4.0 .......... 23837.2 |================================================ GCC 12.2 .......... 23526.4 |=============================================== GCC 13.0 14 Nov ... 24105.0 |================================================= LLVM Clang 14 ..... 24796.8 |================================================== LLVM Clang 15.0.2 . 24197.7 |================================================= C-Blosc 2.3 Test: blosclz bitshuffle MB/s > Higher Is Better AOCC 4.0 .......... 13508.9 |================================================ GCC 12.2 .......... 13017.3 |=============================================== GCC 13.0 14 Nov ... 13286.5 |=============================================== LLVM Clang 14 ..... 13989.2 |================================================== LLVM Clang 15.0.2 . 13694.2 |================================================= C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel Seconds < Lower Is Better AOCC 4.0 .......... 27.45 |==================================================== GCC 12.2 .......... 18.99 |==================================== GCC 13.0 14 Nov ... 19.33 |===================================== LLVM Clang 14 ..... 20.13 |====================================== LLVM Clang 15.0.2 . 20.15 |====================================== Caffe 2020-02-13 Model: AlexNet - Acceleration: CPU - Iterations: 100 Milli-Seconds < Lower Is Better AOCC 4.0 .......... 27653 |=============================== GCC 12.2 .......... 23685 |========================== GCC 13.0 14 Nov ... 23696 |========================== LLVM Clang 14 ..... 46976 |==================================================== LLVM Clang 15.0.2 . 46800 |==================================================== Caffe 2020-02-13 Model: GoogleNet - Acceleration: CPU - Iterations: 100 Milli-Seconds < Lower Is Better AOCC 4.0 .......... 73934 |=================================== GCC 12.2 .......... 65742 |=============================== GCC 13.0 14 Nov ... 64987 |=============================== LLVM Clang 14 ..... 104942 |================================================== LLVM Clang 15.0.2 . 107414 |=================================================== CLOMP 1.2 Static OMP Speedup Speedup > Higher Is Better AOCC 4.0 .......... 25.0 |===================================================== GCC 12.2 .......... 22.8 |================================================ GCC 13.0 14 Nov ... 22.6 |================================================ LLVM Clang 14 ..... 23.7 |================================================== LLVM Clang 15.0.2 . 23.6 |================================================== Coremark 1.0 CoreMark Size 666 - Iterations Per Second Iterations/Sec > Higher Is Better AOCC 4.0 .......... 946032.29 |======================================== GCC 12.2 .......... 1066150.95 |============================================= GCC 13.0 14 Nov ... 1117817.25 |=============================================== LLVM Clang 14 ..... 981244.49 |========================================= LLVM Clang 15.0.2 . 1004815.01 |========================================== CppPerformanceBenchmarks 9 Test: Atol Seconds < Lower Is Better AOCC 4.0 .......... 25.52 |=================================================== GCC 12.2 .......... 25.43 |=================================================== GCC 13.0 14 Nov ... 25.55 |=================================================== LLVM Clang 14 ..... 25.46 |=================================================== LLVM Clang 15.0.2 . 26.06 |==================================================== CppPerformanceBenchmarks 9 Test: Ctype Seconds < Lower Is Better AOCC 4.0 .......... 31.36 |================================================ GCC 12.2 .......... 32.28 |================================================== GCC 13.0 14 Nov ... 33.77 |==================================================== LLVM Clang 14 ..... 31.63 |================================================= LLVM Clang 15.0.2 . 31.78 |================================================= CppPerformanceBenchmarks 9 Test: Math Library Seconds < Lower Is Better AOCC 4.0 .......... 164.80 |================================================== GCC 12.2 .......... 166.58 |================================================== GCC 13.0 14 Nov ... 169.48 |=================================================== LLVM Clang 14 ..... 164.39 |================================================= LLVM Clang 15.0.2 . 165.55 |================================================== CppPerformanceBenchmarks 9 Test: Stepanov Vector Seconds < Lower Is Better AOCC 4.0 .......... 39.38 |=================================================== GCC 12.2 .......... 35.20 |============================================= GCC 13.0 14 Nov ... 35.51 |============================================== LLVM Clang 14 ..... 40.18 |==================================================== LLVM Clang 15.0.2 . 40.28 |==================================================== CppPerformanceBenchmarks 9 Test: Function Objects Seconds < Lower Is Better AOCC 4.0 .......... 9.023 |================================================== GCC 12.2 .......... 9.251 |==================================================== GCC 13.0 14 Nov ... 9.211 |=================================================== LLVM Clang 14 ..... 9.246 |==================================================== LLVM Clang 15.0.2 . 9.333 |==================================================== CppPerformanceBenchmarks 9 Test: Stepanov Abstraction Seconds < Lower Is Better AOCC 4.0 .......... 16.93 |=================================================== GCC 12.2 .......... 15.73 |================================================ GCC 13.0 14 Nov ... 15.73 |================================================ LLVM Clang 14 ..... 17.03 |==================================================== LLVM Clang 15.0.2 . 17.18 |==================================================== Crypto++ 8.2 Test: Keyed Algorithms MiB/second > Higher Is Better AOCC 4.0 .......... 1048.65 |================================================ GCC 12.2 .......... 881.35 |========================================= LLVM Clang 14 ..... 1086.85 |================================================== LLVM Clang 15.0.2 . 1075.79 |================================================= Crypto++ 8.2 Test: Unkeyed Algorithms MiB/second > Higher Is Better AOCC 4.0 .......... 591.18 |=================================================== GCC 12.2 .......... 481.76 |========================================= LLVM Clang 14 ..... 593.21 |=================================================== LLVM Clang 15.0.2 . 596.87 |=================================================== Dolfyn 0.527 Computational Fluid Dynamics Seconds < Lower Is Better AOCC 4.0 .......... 10.77 |==================================================== GCC 12.2 .......... 10.79 |==================================================== GCC 13.0 14 Nov ... 10.51 |=================================================== LLVM Clang 14 ..... 10.77 |==================================================== LLVM Clang 15.0.2 . 10.82 |==================================================== Dragonflydb 0.6 Clients: 50 - Set To Get Ratio: 1:5 Ops/sec > Higher Is Better AOCC 4.0 .......... 5728724.37 |=============================================== GCC 12.2 .......... 5707956.74 |=============================================== GCC 13.0 14 Nov ... 5757620.86 |=============================================== Dragonflydb 0.6 Clients: 50 - Set To Get Ratio: 5:1 Ops/sec > Higher Is Better AOCC 4.0 .......... 5242954.59 |=============================================== GCC 12.2 .......... 5275879.83 |=============================================== GCC 13.0 14 Nov ... 5271140.49 |=============================================== eSpeak-NG Speech Engine 20200907 Text-To-Speech Synthesis Seconds < Lower Is Better AOCC 4.0 .......... 14.54 |===================================== GCC 12.2 .......... 17.83 |============================================= GCC 13.0 14 Nov ... 17.77 |============================================= LLVM Clang 14 ..... 20.54 |==================================================== FLAC Audio Encoding 1.4 WAV To FLAC Seconds < Lower Is Better AOCC 4.0 .......... 9.928 |=============================================== GCC 12.2 .......... 10.520 |================================================== GCC 13.0 14 Nov ... 10.810 |=================================================== LLVM Clang 14 ..... 9.280 |============================================ LLVM Clang 15.0.2 . 9.435 |============================================= Google Draco 1.5.0 Model: Lion ms < Lower Is Better AOCC 4.0 .......... 3373 |================================================ GCC 12.2 .......... 3486 |================================================= LLVM Clang 14 ..... 3763 |===================================================== LLVM Clang 15.0.2 . 3700 |==================================================== Google Draco 1.5.0 Model: Church Facade ms < Lower Is Better AOCC 4.0 .......... 4324 |================================================ GCC 12.2 .......... 4424 |================================================= LLVM Clang 14 ..... 4782 |===================================================== LLVM Clang 15.0.2 . 4649 |==================================================== GraphicsMagick 1.3.38 Operation: Swirl Iterations Per Minute > Higher Is Better AOCC 4.0 .......... 1413 |=================================================== GCC 12.2 .......... 1473 |===================================================== GCC 13.0 14 Nov ... 1454 |==================================================== LLVM Clang 14 ..... 1382 |================================================== LLVM Clang 15.0.2 . 1390 |================================================== GraphicsMagick 1.3.38 Operation: Rotate Iterations Per Minute > Higher Is Better AOCC 4.0 .......... 952 |==================================================== GCC 12.2 .......... 988 |====================================================== GCC 13.0 14 Nov ... 912 |================================================== LLVM Clang 14 ..... 896 |================================================= LLVM Clang 15.0.2 . 876 |================================================ GraphicsMagick 1.3.38 Operation: Sharpen Iterations Per Minute > Higher Is Better AOCC 4.0 .......... 470 |====================================================== GCC 12.2 .......... 343 |======================================= GCC 13.0 14 Nov ... 341 |======================================= LLVM Clang 14 ..... 466 |====================================================== LLVM Clang 15.0.2 . 454 |==================================================== GraphicsMagick 1.3.38 Operation: Enhanced Iterations Per Minute > Higher Is Better AOCC 4.0 .......... 612 |==================================================== GCC 12.2 .......... 639 |====================================================== GCC 13.0 14 Nov ... 627 |===================================================== LLVM Clang 14 ..... 582 |================================================= LLVM Clang 15.0.2 . 583 |================================================= GraphicsMagick 1.3.38 Operation: Resizing Iterations Per Minute > Higher Is Better AOCC 4.0 .......... 2609 |================================================== GCC 12.2 .......... 2741 |===================================================== GCC 13.0 14 Nov ... 2625 |=================================================== LLVM Clang 14 ..... 2705 |==================================================== LLVM Clang 15.0.2 . 2655 |=================================================== GraphicsMagick 1.3.38 Operation: Noise-Gaussian Iterations Per Minute > Higher Is Better AOCC 4.0 .......... 607 |================================================= GCC 12.2 .......... 664 |====================================================== GCC 13.0 14 Nov ... 658 |====================================================== LLVM Clang 14 ..... 587 |================================================ LLVM Clang 15.0.2 . 583 |=============================================== GraphicsMagick 1.3.38 Operation: HWB Color Space Iterations Per Minute > Higher Is Better AOCC 4.0 .......... 1710 |===================================================== GCC 12.2 .......... 1533 |================================================ GCC 13.0 14 Nov ... 1438 |============================================= LLVM Clang 14 ..... 1580 |================================================= LLVM Clang 15.0.2 . 1584 |================================================= JPEG XL Decoding libjxl 0.7 CPU Threads: 1 MP/s > Higher Is Better AOCC 4.0 .......... 77.27 |==================================================== GCC 12.2 .......... 68.44 |============================================== GCC 13.0 14 Nov ... 66.30 |============================================= LLVM Clang 14 ..... 62.29 |========================================== LLVM Clang 15.0.2 . 63.32 |=========================================== JPEG XL Decoding libjxl 0.7 CPU Threads: All MP/s > Higher Is Better AOCC 4.0 .......... 286.44 |=================================================== GCC 12.2 .......... 263.90 |=============================================== GCC 13.0 14 Nov ... 245.54 |============================================ LLVM Clang 14 ..... 264.48 |=============================================== LLVM Clang 15.0.2 . 266.89 |================================================ JPEG XL libjxl 0.7 Input: PNG - Quality: 90 MP/s > Higher Is Better AOCC 4.0 .......... 13.50 |======================================= GCC 12.2 .......... 12.98 |====================================== GCC 13.0 14 Nov ... 13.01 |====================================== LLVM Clang 14 ..... 13.19 |====================================== LLVM Clang 15.0.2 . 17.98 |==================================================== JPEG XL libjxl 0.7 Input: JPEG - Quality: 90 MP/s > Higher Is Better AOCC 4.0 .......... 13.13 |======================================= GCC 12.2 .......... 12.54 |===================================== GCC 13.0 14 Nov ... 12.76 |====================================== LLVM Clang 14 ..... 13.10 |======================================= LLVM Clang 15.0.2 . 17.46 |==================================================== JPEG XL libjxl 0.7 Input: PNG - Quality: 100 MP/s > Higher Is Better AOCC 4.0 .......... 1.08 |==================================================== GCC 12.2 .......... 1.04 |================================================== GCC 13.0 14 Nov ... 1.03 |================================================= LLVM Clang 14 ..... 1.07 |=================================================== LLVM Clang 15.0.2 . 1.11 |===================================================== JPEG XL libjxl 0.7 Input: JPEG - Quality: 100 MP/s > Higher Is Better AOCC 4.0 .......... 0.85 |================================================ GCC 12.2 .......... 0.85 |================================================ GCC 13.0 14 Nov ... 0.85 |================================================ LLVM Clang 14 ..... 0.94 |===================================================== LLVM Clang 15.0.2 . 0.87 |================================================= Kripke 1.2.4 Throughput FoM > Higher Is Better AOCC 4.0 .......... 125045767 |============================================= GCC 12.2 .......... 133327173 |================================================ GCC 13.0 14 Nov ... 132057600 |================================================ LLVM Clang 14 ..... 109794900 |======================================== LLVM Clang 15.0.2 . 108977267 |======================================= KTX-Software toktx 4.0 Settings: UASTC 3 Seconds < Lower Is Better AOCC 4.0 .......... 4.957 |==================================================== GCC 12.2 .......... 4.958 |==================================================== GCC 13.0 14 Nov ... 4.961 |==================================================== LLVM Clang 14 ..... 4.982 |==================================================== LLVM Clang 15.0.2 . 4.983 |==================================================== KTX-Software toktx 4.0 Settings: Zstd Compression 19 Seconds < Lower Is Better AOCC 4.0 .......... 11.35 |================================================== GCC 12.2 .......... 11.69 |==================================================== GCC 13.0 14 Nov ... 11.62 |==================================================== LLVM Clang 14 ..... 11.64 |==================================================== LLVM Clang 15.0.2 . 11.46 |=================================================== Kvazaar 2.1 Video Input: Bosphorus 4K - Video Preset: Very Fast Frames Per Second > Higher Is Better AOCC 4.0 .......... 47.33 |==================================================== GCC 12.2 .......... 41.78 |============================================== GCC 13.0 14 Nov ... 44.12 |================================================ LLVM Clang 14 ..... 44.62 |================================================= LLVM Clang 15.0.2 . 45.31 |================================================== Kvazaar 2.1 Video Input: Bosphorus 4K - Video Preset: Ultra Fast Frames Per Second > Higher Is Better AOCC 4.0 .......... 85.35 |==================================================== GCC 12.2 .......... 80.31 |================================================= GCC 13.0 14 Nov ... 82.56 |================================================== LLVM Clang 14 ..... 82.97 |=================================================== LLVM Clang 15.0.2 . 84.80 |==================================================== Kvazaar 2.1 Video Input: Bosphorus 1080p - Video Preset: Very Fast Frames Per Second > Higher Is Better AOCC 4.0 .......... 113.72 |=================================================== GCC 12.2 .......... 102.77 |============================================== GCC 13.0 14 Nov ... 108.11 |================================================ LLVM Clang 14 ..... 107.85 |================================================ LLVM Clang 15.0.2 . 109.33 |================================================= Kvazaar 2.1 Video Input: Bosphorus 1080p - Video Preset: Ultra Fast Frames Per Second > Higher Is Better AOCC 4.0 .......... 219.33 |=================================================== GCC 12.2 .......... 204.74 |================================================ GCC 13.0 14 Nov ... 210.39 |================================================= LLVM Clang 14 ..... 210.45 |================================================= LLVM Clang 15.0.2 . 213.39 |================================================== LAME MP3 Encoding 3.100 WAV To MP3 Seconds < Lower Is Better AOCC 4.0 .......... 4.732 |==================================================== GCC 12.2 .......... 4.409 |================================================ GCC 13.0 14 Nov ... 4.387 |================================================ LLVM Clang 14 ..... 4.620 |=================================================== LLVM Clang 15.0.2 . 4.529 |================================================== LAMMPS Molecular Dynamics Simulator 23Jun2022 Model: Rhodopsin Protein ns/day > Higher Is Better AOCC 4.0 .......... 17.91 |==================================================== GCC 12.2 .......... 17.26 |================================================== GCC 13.0 14 Nov ... 17.04 |================================================= LLVM Clang 14 ..... 17.69 |=================================================== LLVM Clang 15.0.2 . 17.39 |================================================== LeelaChessZero 0.28 Backend: BLAS Nodes Per Second > Higher Is Better AOCC 4.0 .......... 1748 |===================================================== GCC 12.2 .......... 1697 |=================================================== LLVM Clang 14 ..... 1702 |==================================================== LLVM Clang 15.0.2 . 1686 |=================================================== LeelaChessZero 0.28 Backend: Eigen Nodes Per Second > Higher Is Better AOCC 4.0 .......... 1945 |===================================================== GCC 12.2 .......... 1725 |=============================================== LLVM Clang 14 ..... 1861 |=================================================== LLVM Clang 15.0.2 . 1863 |=================================================== libavif avifenc 0.11 Encoder Speed: 0 Seconds < Lower Is Better AOCC 4.0 .......... 63.11 |================================================ GCC 12.2 .......... 68.52 |==================================================== GCC 13.0 14 Nov ... 66.27 |================================================== LLVM Clang 14 ..... 65.23 |================================================== LLVM Clang 15.0.2 . 65.15 |================================================= libavif avifenc 0.11 Encoder Speed: 2 Seconds < Lower Is Better AOCC 4.0 .......... 31.38 |=============================================== GCC 12.2 .......... 34.85 |==================================================== GCC 13.0 14 Nov ... 33.75 |================================================== LLVM Clang 14 ..... 32.38 |================================================ LLVM Clang 15.0.2 . 32.31 |================================================ libavif avifenc 0.11 Encoder Speed: 6 Seconds < Lower Is Better AOCC 4.0 .......... 2.943 |================================================ GCC 12.2 .......... 3.202 |==================================================== GCC 13.0 14 Nov ... 3.105 |================================================== LLVM Clang 14 ..... 3.079 |================================================== LLVM Clang 15.0.2 . 3.044 |================================================= libavif avifenc 0.11 Encoder Speed: 6, Lossless Seconds < Lower Is Better AOCC 4.0 .......... 4.229 |============================================ GCC 12.2 .......... 5.028 |==================================================== GCC 13.0 14 Nov ... 4.675 |================================================ LLVM Clang 14 ..... 4.377 |============================================= LLVM Clang 15.0.2 . 4.363 |============================================= libavif avifenc 0.11 Encoder Speed: 10, Lossless Seconds < Lower Is Better AOCC 4.0 .......... 2.773 |=============================================== GCC 12.2 .......... 3.077 |==================================================== GCC 13.0 14 Nov ... 2.947 |================================================== LLVM Clang 14 ..... 2.761 |=============================================== LLVM Clang 15.0.2 . 2.725 |============================================== libjpeg-turbo tjbench 2.1.0 Test: Decompression Throughput Megapixels/sec > Higher Is Better AOCC 4.0 .......... 318.29 |================================================== GCC 12.2 .......... 314.76 |================================================== GCC 13.0 14 Nov ... 322.49 |=================================================== LLVM Clang 14 ..... 320.16 |=================================================== LLVM Clang 15.0.2 . 321.22 |=================================================== Liquid-DSP 2021.01.31 Threads: 8 - Buffer Length: 256 - Filter Length: 57 samples/s > Higher Is Better AOCC 4.0 .......... 896243333 |================================================ GCC 12.2 .......... 834506667 |============================================= GCC 13.0 14 Nov ... 650876667 |=================================== LLVM Clang 14 ..... 809823333 |=========================================== LLVM Clang 15.0.2 . 566352750 |============================== Liquid-DSP 2021.01.31 Threads: 16 - Buffer Length: 256 - Filter Length: 57 samples/s > Higher Is Better AOCC 4.0 .......... 1754833333 |=============================================== GCC 12.2 .......... 1627233333 |============================================ GCC 13.0 14 Nov ... 1274433333 |================================== LLVM Clang 14 ..... 1584066667 |========================================== LLVM Clang 15.0.2 . 1544266667 |========================================= Liquid-DSP 2021.01.31 Threads: 32 - Buffer Length: 256 - Filter Length: 57 samples/s > Higher Is Better AOCC 4.0 .......... 1886600000 |=============================================== GCC 12.2 .......... 1680600000 |========================================== GCC 13.0 14 Nov ... 1633166667 |========================================= LLVM Clang 14 ..... 1723600000 |=========================================== LLVM Clang 15.0.2 . 1654933333 |========================================= Monte Carlo Simulations of Ionised Nebulae 2019-03-24 Input: Dust 2D tau100.0 Seconds < Lower Is Better AOCC 4.0 .......... 64 |====================================================== GCC 12.2 .......... 64 |====================================================== GCC 13.0 14 Nov ... 63 |===================================================== LLVM Clang 14 ..... 65 |======================================================= LLVM Clang 15.0.2 . 64 |====================================================== NCNN 20220729 Target: CPU - Model: mobilenet ms < Lower Is Better AOCC 4.0 .......... 8.57 |================================================= GCC 12.2 .......... 8.68 |================================================== GCC 13.0 14 Nov ... 8.68 |================================================== LLVM Clang 14 ..... 9.09 |==================================================== LLVM Clang 15.0.2 . 9.24 |===================================================== NCNN 20220729 Target: CPU-v2-v2 - Model: mobilenet-v2 ms < Lower Is Better AOCC 4.0 .......... 3.17 |============================================= GCC 12.2 .......... 3.71 |===================================================== GCC 13.0 14 Nov ... 3.72 |===================================================== LLVM Clang 14 ..... 3.44 |================================================= LLVM Clang 15.0.2 . 3.53 |================================================== NCNN 20220729 Target: CPU-v3-v3 - Model: mobilenet-v3 ms < Lower Is Better AOCC 4.0 .......... 2.69 |============================================ GCC 12.2 .......... 3.24 |===================================================== GCC 13.0 14 Nov ... 3.26 |===================================================== LLVM Clang 14 ..... 2.99 |================================================= LLVM Clang 15.0.2 . 2.99 |================================================= NCNN 20220729 Target: CPU - Model: shufflenet-v2 ms < Lower Is Better AOCC 4.0 .......... 2.96 |========================================== GCC 12.2 .......... 3.76 |===================================================== GCC 13.0 14 Nov ... 3.74 |===================================================== LLVM Clang 14 ..... 3.25 |============================================== LLVM Clang 15.0.2 . 3.31 |=============================================== NCNN 20220729 Target: CPU - Model: mnasnet ms < Lower Is Better AOCC 4.0 .......... 2.72 |=========================================== GCC 12.2 .......... 3.30 |===================================================== GCC 13.0 14 Nov ... 3.32 |===================================================== LLVM Clang 14 ..... 3.02 |================================================ LLVM Clang 15.0.2 . 3.07 |================================================= NCNN 20220729 Target: CPU - Model: efficientnet-b0 ms < Lower Is Better AOCC 4.0 .......... 3.85 |============================================ GCC 12.2 .......... 4.64 |===================================================== GCC 13.0 14 Nov ... 4.57 |==================================================== LLVM Clang 14 ..... 4.25 |================================================= LLVM Clang 15.0.2 . 4.35 |================================================== NCNN 20220729 Target: CPU - Model: blazeface ms < Lower Is Better AOCC 4.0 .......... 1.21 |======================================== GCC 12.2 .......... 1.59 |===================================================== GCC 13.0 14 Nov ... 1.58 |===================================================== LLVM Clang 14 ..... 1.37 |============================================== LLVM Clang 15.0.2 . 1.41 |=============================================== NCNN 20220729 Target: CPU - Model: googlenet ms < Lower Is Better AOCC 4.0 .......... 7.85 |================================================ GCC 12.2 .......... 8.66 |===================================================== GCC 13.0 14 Nov ... 8.65 |===================================================== LLVM Clang 14 ..... 8.54 |==================================================== LLVM Clang 15.0.2 . 8.63 |===================================================== NCNN 20220729 Target: CPU - Model: vgg16 ms < Lower Is Better AOCC 4.0 .......... 24.83 |================================================== GCC 12.2 .......... 24.56 |================================================= GCC 13.0 14 Nov ... 24.39 |================================================= LLVM Clang 14 ..... 25.96 |==================================================== LLVM Clang 15.0.2 . 25.52 |=================================================== NCNN 20220729 Target: CPU - Model: resnet18 ms < Lower Is Better AOCC 4.0 .......... 7.18 |=================================================== GCC 12.2 .......... 7.49 |===================================================== GCC 13.0 14 Nov ... 7.49 |===================================================== LLVM Clang 14 ..... 7.36 |==================================================== LLVM Clang 15.0.2 . 7.41 |==================================================== NCNN 20220729 Target: CPU - Model: alexnet ms < Lower Is Better AOCC 4.0 .......... 4.79 |================================================= GCC 12.2 .......... 4.99 |=================================================== GCC 13.0 14 Nov ... 4.98 |=================================================== LLVM Clang 14 ..... 5.19 |===================================================== LLVM Clang 15.0.2 . 5.16 |===================================================== NCNN 20220729 Target: CPU - Model: resnet50 ms < Lower Is Better AOCC 4.0 .......... 11.62 |================================================= GCC 12.2 .......... 12.25 |==================================================== GCC 13.0 14 Nov ... 12.26 |==================================================== LLVM Clang 14 ..... 12.32 |==================================================== LLVM Clang 15.0.2 . 12.32 |==================================================== NCNN 20220729 Target: CPU - Model: yolov4-tiny ms < Lower Is Better AOCC 4.0 .......... 14.38 |================================================== GCC 12.2 .......... 14.03 |================================================= GCC 13.0 14 Nov ... 14.12 |================================================= LLVM Clang 14 ..... 14.68 |=================================================== LLVM Clang 15.0.2 . 14.94 |==================================================== NCNN 20220729 Target: CPU - Model: squeezenet_ssd ms < Lower Is Better AOCC 4.0 .......... 11.27 |=============================================== GCC 12.2 .......... 12.43 |==================================================== GCC 13.0 14 Nov ... 12.27 |=================================================== LLVM Clang 14 ..... 11.80 |================================================= LLVM Clang 15.0.2 . 11.97 |================================================== NCNN 20220729 Target: CPU - Model: regnety_400m ms < Lower Is Better AOCC 4.0 .......... 7.55 |================================= GCC 12.2 .......... 12.01 |==================================================== GCC 13.0 14 Nov ... 11.84 |=================================================== LLVM Clang 14 ..... 9.41 |========================================= LLVM Clang 15.0.2 . 9.62 |========================================== Nettle 3.8 Test: aes256 Mbyte/s > Higher Is Better AOCC 4.0 .......... 8564.64 |================================================== GCC 12.2 .......... 8521.41 |================================================== GCC 13.0 14 Nov ... 8490.46 |================================================== LLVM Clang 14 ..... 8566.17 |================================================== LLVM Clang 15.0.2 . 8553.15 |================================================== Nettle 3.8 Test: chacha Mbyte/s > Higher Is Better AOCC 4.0 .......... 1528.82 |================================================== GCC 12.2 .......... 1515.12 |================================================= GCC 13.0 14 Nov ... 1507.34 |================================================= LLVM Clang 14 ..... 1536.84 |================================================== LLVM Clang 15.0.2 . 1523.44 |================================================== Nettle 3.8 Test: sha512 Mbyte/s > Higher Is Better AOCC 4.0 .......... 864.29 |=================================================== GCC 12.2 .......... 858.30 |=================================================== GCC 13.0 14 Nov ... 859.96 |=================================================== LLVM Clang 14 ..... 864.87 |=================================================== LLVM Clang 15.0.2 . 863.73 |=================================================== Nettle 3.8 Test: poly1305-aes Mbyte/s > Higher Is Better AOCC 4.0 .......... 4255.91 |================================================== GCC 12.2 .......... 4182.31 |================================================= GCC 13.0 14 Nov ... 4188.14 |================================================= LLVM Clang 14 ..... 4231.63 |================================================== LLVM Clang 15.0.2 . 4214.74 |================================================== Ngspice 34 Circuit: C2670 Seconds < Lower Is Better AOCC 4.0 .......... 58.26 |================================================ GCC 12.2 .......... 63.75 |==================================================== GCC 13.0 14 Nov ... 62.68 |=================================================== LLVM Clang 14 ..... 57.71 |=============================================== LLVM Clang 15.0.2 . 58.96 |================================================ Ngspice 34 Circuit: C7552 Seconds < Lower Is Better AOCC 4.0 .......... 56.31 |================================================= GCC 12.2 .......... 58.28 |================================================== GCC 13.0 14 Nov ... 60.29 |==================================================== LLVM Clang 14 ..... 56.22 |================================================ LLVM Clang 15.0.2 . 56.48 |================================================= oneDNN 2.7 Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better AOCC 4.0 .......... 0.338581 |================================ GCC 12.2 .......... 0.479906 |============================================= GCC 13.0 14 Nov ... 0.524283 |================================================= LLVM Clang 14 ..... 0.343595 |================================ LLVM Clang 15.0.2 . 0.343749 |================================ oneDNN 2.7 Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better AOCC 4.0 .......... 0.347017 |=============================================== GCC 12.2 .......... 0.356546 |================================================ GCC 13.0 14 Nov ... 0.356407 |================================================ LLVM Clang 14 ..... 0.337760 |============================================== LLVM Clang 15.0.2 . 0.362285 |================================================= oneDNN 2.7 Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better AOCC 4.0 .......... 0.577834 |======================================= GCC 12.2 .......... 0.725978 |================================================= GCC 13.0 14 Nov ... 0.662905 |============================================= LLVM Clang 14 ..... 0.583377 |======================================= LLVM Clang 15.0.2 . 0.582381 |======================================= oneDNN 2.7 Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better AOCC 4.0 .......... 1.50016 |================================================= GCC 12.2 .......... 1.54379 |================================================== GCC 13.0 14 Nov ... 1.53305 |================================================== LLVM Clang 14 ..... 1.51176 |================================================= LLVM Clang 15.0.2 . 1.51369 |================================================= oneDNN 2.7 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better AOCC 4.0 .......... 5.28785 |================================================== GCC 12.2 .......... 5.32780 |================================================== GCC 13.0 14 Nov ... 5.30501 |================================================== LLVM Clang 14 ..... 5.31930 |================================================== LLVM Clang 15.0.2 . 5.30337 |================================================== oneDNN 2.7 Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better AOCC 4.0 .......... 0.405097 |=============================================== GCC 12.2 .......... 0.421419 |================================================= GCC 13.0 14 Nov ... 0.418860 |================================================= LLVM Clang 14 ..... 0.408432 |=============================================== LLVM Clang 15.0.2 . 0.408594 |================================================ oneDNN 2.7 Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better AOCC 4.0 .......... 1138.83 |================================================== GCC 12.2 .......... 1140.34 |================================================== GCC 13.0 14 Nov ... 1139.64 |================================================== LLVM Clang 14 ..... 1138.26 |================================================== LLVM Clang 15.0.2 . 1137.49 |================================================== oneDNN 2.7 Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better AOCC 4.0 .......... 3.27367 |=============================================== GCC 12.2 .......... 3.36412 |================================================= GCC 13.0 14 Nov ... 3.42686 |================================================= LLVM Clang 14 ..... 3.39581 |================================================= LLVM Clang 15.0.2 . 3.46444 |================================================== oneDNN 2.7 Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better AOCC 4.0 .......... 1.38455 |================================================= GCC 12.2 .......... 1.38696 |================================================= GCC 13.0 14 Nov ... 1.40503 |================================================== LLVM Clang 14 ..... 1.40766 |================================================== LLVM Clang 15.0.2 . 1.39579 |================================================== oneDNN 2.7 Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better AOCC 4.0 .......... 574.01 |================================================== GCC 12.2 .......... 580.33 |=================================================== GCC 13.0 14 Nov ... 582.00 |=================================================== LLVM Clang 14 ..... 575.60 |================================================== LLVM Clang 15.0.2 . 575.13 |================================================== oneDNN 2.7 Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better AOCC 4.0 .......... 1139.44 |================================================== GCC 12.2 .......... 1139.55 |================================================== GCC 13.0 14 Nov ... 1142.22 |================================================== LLVM Clang 14 ..... 1142.95 |================================================== LLVM Clang 15.0.2 . 1142.00 |================================================== oneDNN 2.7 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better AOCC 4.0 .......... 574.91 |=================================================== GCC 12.2 .......... 578.82 |=================================================== GCC 13.0 14 Nov ... 578.96 |=================================================== LLVM Clang 14 ..... 575.94 |=================================================== LLVM Clang 15.0.2 . 576.22 |=================================================== oneDNN 2.7 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better AOCC 4.0 .......... 0.133169 |============================================== GCC 12.2 .......... 0.139862 |================================================= GCC 13.0 14 Nov ... 0.140349 |================================================= LLVM Clang 14 ..... 0.136343 |================================================ LLVM Clang 15.0.2 . 0.136310 |================================================ oneDNN 2.7 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better AOCC 4.0 .......... 0.222378 |================================================ GCC 12.2 .......... 0.228514 |================================================= GCC 13.0 14 Nov ... 0.227918 |================================================= LLVM Clang 14 ..... 0.224101 |================================================ LLVM Clang 15.0.2 . 0.224449 |================================================ ONNX Runtime 1.11 Model: GPT-2 - Device: CPU - Executor: Parallel Inferences Per Minute > Higher Is Better AOCC 4.0 .......... 7135 |===================================================== GCC 12.2 .......... 7126 |===================================================== LLVM Clang 14 ..... 7062 |==================================================== ONNX Runtime 1.11 Model: GPT-2 - Device: CPU - Executor: Standard Inferences Per Minute > Higher Is Better AOCC 4.0 .......... 9147 |===================================================== GCC 12.2 .......... 8940 |==================================================== LLVM Clang 14 ..... 8980 |==================================================== ONNX Runtime 1.11 Model: yolov4 - Device: CPU - Executor: Parallel Inferences Per Minute > Higher Is Better AOCC 4.0 .......... 569 |====================================================== GCC 12.2 .......... 568 |====================================================== LLVM Clang 14 ..... 563 |===================================================== ONNX Runtime 1.11 Model: yolov4 - Device: CPU - Executor: Standard Inferences Per Minute > Higher Is Better AOCC 4.0 .......... 576 |====================================================== GCC 12.2 .......... 564 |===================================================== LLVM Clang 14 ..... 566 |===================================================== ONNX Runtime 1.11 Model: bertsquad-12 - Device: CPU - Executor: Parallel Inferences Per Minute > Higher Is Better AOCC 4.0 .......... 917 |====================================================== GCC 12.2 .......... 918 |====================================================== LLVM Clang 14 ..... 923 |====================================================== ONNX Runtime 1.11 Model: fcn-resnet101-11 - Device: CPU - Executor: Parallel Inferences Per Minute > Higher Is Better AOCC 4.0 .......... 135 |====================================================== GCC 12.2 .......... 134 |====================================================== LLVM Clang 14 ..... 135 |====================================================== ONNX Runtime 1.11 Model: ArcFace ResNet-100 - Device: CPU - Executor: Parallel Inferences Per Minute > Higher Is Better AOCC 4.0 .......... 2356 |===================================================== GCC 12.2 .......... 2361 |===================================================== LLVM Clang 14 ..... 2353 |===================================================== ONNX Runtime 1.11 Model: ArcFace ResNet-100 - Device: CPU - Executor: Standard Inferences Per Minute > Higher Is Better AOCC 4.0 .......... 2207 |========================================== GCC 12.2 .......... 2222 |========================================== LLVM Clang 14 ..... 2781 |===================================================== ONNX Runtime 1.11 Model: super-resolution-10 - Device: CPU - Executor: Parallel Inferences Per Minute > Higher Is Better AOCC 4.0 .......... 9113 |=================================================== GCC 12.2 .......... 9395 |===================================================== LLVM Clang 14 ..... 8949 |================================================== OpenJPEG 2.4 Encode: NASA Curiosity Panorama M34 ms < Lower Is Better AOCC 4.0 .......... 73991 |==================================================== GCC 12.2 .......... 71222 |================================================== GCC 13.0 14 Nov ... 69608 |================================================= LLVM Clang 14 ..... 64416 |============================================= LLVM Clang 15.0.2 . 62689 |============================================ OpenSSL 3.0 Algorithm: SHA256 byte/s > Higher Is Better AOCC 4.0 .......... 40426163787 |============================================== GCC 12.2 .......... 36864232800 |========================================== GCC 13.0 14 Nov ... 37385343083 |=========================================== LLVM Clang 14 ..... 39948818857 |============================================= LLVM Clang 15.0.2 . 37916267870 |=========================================== OpenSSL 3.0 Algorithm: RSA4096 sign/s > Higher Is Better AOCC 4.0 .......... 6026.2 |=================================================== GCC 12.2 .......... 6014.6 |=================================================== GCC 13.0 14 Nov ... 6029.1 |=================================================== LLVM Clang 14 ..... 6016.6 |=================================================== LLVM Clang 15.0.2 . 6030.9 |=================================================== OpenSSL 3.0 Algorithm: RSA4096 verify/s > Higher Is Better AOCC 4.0 .......... 394139.1 |================================================= GCC 12.2 .......... 393865.8 |================================================= GCC 13.0 14 Nov ... 394339.8 |================================================= LLVM Clang 14 ..... 395258.1 |================================================= LLVM Clang 15.0.2 . 395636.4 |================================================= OpenVINO 2022.2.dev Model: Face Detection FP16 - Device: CPU FPS > Higher Is Better AOCC 4.0 .......... 14.29 |==================================================== GCC 12.2 .......... 14.29 |==================================================== LLVM Clang 14 ..... 14.37 |==================================================== LLVM Clang 15.0.2 . 14.35 |==================================================== OpenVINO 2022.2.dev Model: Face Detection FP16 - Device: CPU ms < Lower Is Better AOCC 4.0 .......... 557.01 |=================================================== GCC 12.2 .......... 557.36 |=================================================== LLVM Clang 14 ..... 555.00 |=================================================== LLVM Clang 15.0.2 . 555.05 |=================================================== OpenVINO 2022.2.dev Model: Person Detection FP16 - Device: CPU FPS > Higher Is Better AOCC 4.0 .......... 7.53 |===================================================== GCC 12.2 .......... 7.50 |===================================================== LLVM Clang 14 ..... 7.46 |===================================================== LLVM Clang 15.0.2 . 7.47 |===================================================== OpenVINO 2022.2.dev Model: Person Detection FP16 - Device: CPU ms < Lower Is Better AOCC 4.0 .......... 1055.29 |================================================= GCC 12.2 .......... 1060.87 |================================================== LLVM Clang 14 ..... 1066.07 |================================================== LLVM Clang 15.0.2 . 1065.31 |================================================== OpenVINO 2022.2.dev Model: Person Detection FP32 - Device: CPU FPS > Higher Is Better AOCC 4.0 .......... 7.37 |===================================================== GCC 12.2 .......... 7.34 |==================================================== LLVM Clang 14 ..... 7.29 |==================================================== LLVM Clang 15.0.2 . 7.43 |===================================================== OpenVINO 2022.2.dev Model: Person Detection FP32 - Device: CPU ms < Lower Is Better AOCC 4.0 .......... 1081.01 |================================================= GCC 12.2 .......... 1085.33 |================================================== LLVM Clang 14 ..... 1092.87 |================================================== LLVM Clang 15.0.2 . 1071.50 |================================================= OpenVINO 2022.2.dev Model: Vehicle Detection FP16 - Device: CPU FPS > Higher Is Better AOCC 4.0 .......... 752.27 |=================================================== GCC 12.2 .......... 757.51 |=================================================== LLVM Clang 14 ..... 752.41 |=================================================== LLVM Clang 15.0.2 . 742.44 |================================================== OpenVINO 2022.2.dev Model: Vehicle Detection FP16 - Device: CPU ms < Lower Is Better AOCC 4.0 .......... 10.62 |=================================================== GCC 12.2 .......... 10.55 |=================================================== LLVM Clang 14 ..... 10.62 |=================================================== LLVM Clang 15.0.2 . 10.76 |==================================================== OpenVINO 2022.2.dev Model: Face Detection FP16-INT8 - Device: CPU FPS > Higher Is Better AOCC 4.0 .......... 28.19 |==================================================== GCC 12.2 .......... 28.09 |==================================================== LLVM Clang 14 ..... 28.15 |==================================================== LLVM Clang 15.0.2 . 28.27 |==================================================== OpenVINO 2022.2.dev Model: Face Detection FP16-INT8 - Device: CPU ms < Lower Is Better AOCC 4.0 .......... 283.07 |=================================================== GCC 12.2 .......... 284.35 |=================================================== LLVM Clang 14 ..... 283.72 |=================================================== LLVM Clang 15.0.2 . 282.59 |=================================================== OpenVINO 2022.2.dev Model: Vehicle Detection FP16-INT8 - Device: CPU FPS > Higher Is Better AOCC 4.0 .......... 1847.73 |================================================== GCC 12.2 .......... 1837.19 |================================================== LLVM Clang 14 ..... 1851.09 |================================================== LLVM Clang 15.0.2 . 1845.31 |================================================== OpenVINO 2022.2.dev Model: Vehicle Detection FP16-INT8 - Device: CPU ms < Lower Is Better AOCC 4.0 .......... 4.32 |===================================================== GCC 12.2 .......... 4.35 |===================================================== LLVM Clang 14 ..... 4.32 |===================================================== LLVM Clang 15.0.2 . 4.33 |===================================================== OpenVINO 2022.2.dev Model: Weld Porosity Detection FP16 - Device: CPU FPS > Higher Is Better AOCC 4.0 .......... 1439.92 |================================================== GCC 12.2 .......... 1429.51 |================================================== LLVM Clang 14 ..... 1434.10 |================================================== LLVM Clang 15.0.2 . 1438.37 |================================================== OpenVINO 2022.2.dev Model: Weld Porosity Detection FP16 - Device: CPU ms < Lower Is Better AOCC 4.0 .......... 5.55 |===================================================== GCC 12.2 .......... 5.59 |===================================================== LLVM Clang 14 ..... 5.57 |===================================================== LLVM Clang 15.0.2 . 5.56 |===================================================== OpenVINO 2022.2.dev Model: Machine Translation EN To DE FP16 - Device: CPU FPS > Higher Is Better AOCC 4.0 .......... 133.33 |================================================== GCC 12.2 .......... 134.42 |=================================================== LLVM Clang 14 ..... 133.59 |================================================== LLVM Clang 15.0.2 . 135.00 |=================================================== OpenVINO 2022.2.dev Model: Machine Translation EN To DE FP16 - Device: CPU ms < Lower Is Better AOCC 4.0 .......... 59.95 |==================================================== GCC 12.2 .......... 59.47 |==================================================== LLVM Clang 14 ..... 59.83 |==================================================== LLVM Clang 15.0.2 . 59.20 |=================================================== OpenVINO 2022.2.dev Model: Weld Porosity Detection FP16-INT8 - Device: CPU FPS > Higher Is Better AOCC 4.0 .......... 2896.90 |================================================== GCC 12.2 .......... 2879.68 |================================================== LLVM Clang 14 ..... 2887.98 |================================================== LLVM Clang 15.0.2 . 2895.02 |================================================== OpenVINO 2022.2.dev Model: Weld Porosity Detection FP16-INT8 - Device: CPU ms < Lower Is Better AOCC 4.0 .......... 5.52 |===================================================== GCC 12.2 .......... 5.55 |===================================================== LLVM Clang 14 ..... 5.54 |===================================================== LLVM Clang 15.0.2 . 5.52 |===================================================== OpenVINO 2022.2.dev Model: Person Vehicle Bike Detection FP16 - Device: CPU FPS > Higher Is Better AOCC 4.0 .......... 1664.77 |================================================== GCC 12.2 .......... 1658.30 |================================================== LLVM Clang 14 ..... 1654.66 |================================================== LLVM Clang 15.0.2 . 1645.57 |================================================= OpenVINO 2022.2.dev Model: Person Vehicle Bike Detection FP16 - Device: CPU ms < Lower Is Better AOCC 4.0 .......... 4.80 |==================================================== GCC 12.2 .......... 4.82 |===================================================== LLVM Clang 14 ..... 4.83 |===================================================== LLVM Clang 15.0.2 . 4.85 |===================================================== OpenVINO 2022.2.dev Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU FPS > Higher Is Better AOCC 4.0 .......... 44290.73 |================================================= GCC 12.2 .......... 44470.59 |================================================= LLVM Clang 14 ..... 44085.25 |================================================= LLVM Clang 15.0.2 . 44075.75 |================================================= OpenVINO 2022.2.dev Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU ms < Lower Is Better AOCC 4.0 .......... 0.36 |===================================================== GCC 12.2 .......... 0.36 |===================================================== LLVM Clang 14 ..... 0.36 |===================================================== LLVM Clang 15.0.2 . 0.36 |===================================================== OpenVINO 2022.2.dev Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU FPS > Higher Is Better AOCC 4.0 .......... 63880.92 |================================================= GCC 12.2 .......... 63460.60 |================================================= LLVM Clang 14 ..... 63331.04 |================================================= LLVM Clang 15.0.2 . 63386.57 |================================================= OpenVINO 2022.2.dev Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU ms < Lower Is Better AOCC 4.0 .......... 0.25 |===================================================== GCC 12.2 .......... 0.25 |===================================================== LLVM Clang 14 ..... 0.25 |===================================================== LLVM Clang 15.0.2 . 0.25 |===================================================== PJSIP 2.11 Method: INVITE Responses Per Second > Higher Is Better AOCC 4.0 .......... 5344 |===================================================== GCC 12.2 .......... 5279 |==================================================== GCC 13.0 14 Nov ... 5338 |===================================================== PJSIP 2.11 Method: OPTIONS, Stateful Responses Per Second > Higher Is Better AOCC 4.0 .......... 9370 |===================================================== GCC 12.2 .......... 9330 |===================================================== GCC 13.0 14 Nov ... 9389 |===================================================== PJSIP 2.11 Method: OPTIONS, Stateless Responses Per Second > Higher Is Better AOCC 4.0 .......... 124523 |================================================== GCC 12.2 .......... 125825 |=================================================== GCC 13.0 14 Nov ... 124028 |================================================== POV-Ray 3.7.0.7 Trace Time Seconds < Lower Is Better AOCC 4.0 .......... 16.10 |=================================================== GCC 12.2 .......... 15.96 |================================================== LLVM Clang 14 ..... 16.57 |==================================================== LLVM Clang 15.0.2 . 14.83 |=============================================== Primesieve 8.0 Length: 1e12 Seconds < Lower Is Better AOCC 4.0 .......... 6.245 |==================================================== GCC 12.2 .......... 6.187 |==================================================== LLVM Clang 14 ..... 6.172 |=================================================== LLVM Clang 15.0.2 . 6.243 |==================================================== QuadRay 2022.05.25 Scene: 1 - Resolution: 4K FPS > Higher Is Better AOCC 4.0 .......... 26.43 |==================================================== GCC 12.2 .......... 26.48 |==================================================== GCC 13.0 14 Nov ... 26.30 |==================================================== LLVM Clang 14 ..... 26.23 |==================================================== LLVM Clang 15.0.2 . 26.15 |=================================================== QuadRay 2022.05.25 Scene: 5 - Resolution: 4K FPS > Higher Is Better AOCC 4.0 .......... 2.05 |==================================================== GCC 12.2 .......... 2.08 |===================================================== GCC 13.0 14 Nov ... 1.93 |================================================= LLVM Clang 14 ..... 2.04 |==================================================== LLVM Clang 15.0.2 . 1.92 |================================================= QuadRay 2022.05.25 Scene: 1 - Resolution: 1080p FPS > Higher Is Better AOCC 4.0 .......... 102.79 |=================================================== GCC 12.2 .......... 102.31 |=================================================== GCC 13.0 14 Nov ... 102.06 |=================================================== LLVM Clang 14 ..... 101.89 |=================================================== LLVM Clang 15.0.2 . 102.34 |=================================================== QuadRay 2022.05.25 Scene: 5 - Resolution: 1080p FPS > Higher Is Better AOCC 4.0 .......... 8.12 |==================================================== GCC 12.2 .......... 8.22 |===================================================== GCC 13.0 14 Nov ... 7.65 |================================================= LLVM Clang 14 ..... 8.07 |==================================================== LLVM Clang 15.0.2 . 7.60 |================================================= QuantLib 1.21 MFLOPS > Higher Is Better AOCC 4.0 .......... 4532.3 |================================================ GCC 12.2 .......... 4501.1 |================================================ LLVM Clang 14 ..... 4669.5 |================================================== LLVM Clang 15.0.2 . 4783.9 |=================================================== Redis 7.0.4 Test: SET - Parallel Connections: 50 Requests Per Second > Higher Is Better AOCC 4.0 .......... 4701871.50 |=============================================== GCC 12.2 .......... 4498344.35 |============================================= GCC 13.0 14 Nov ... 4675697.00 |=============================================== SecureMark 1.0.4 Benchmark: SecureMark-TLS marks > Higher Is Better AOCC 4.0 .......... 475941 |=================================================== GCC 12.2 .......... 432810 |============================================== GCC 13.0 14 Nov ... 432896 |============================================== LLVM Clang 14 ..... 472624 |=================================================== LLVM Clang 15.0.2 . 454957 |================================================= simdjson 2.0 Throughput Test: Kostya GB/s > Higher Is Better AOCC 4.0 .......... 5.85 |=================================================== GCC 12.2 .......... 5.89 |==================================================== GCC 13.0 14 Nov ... 6.03 |===================================================== LLVM Clang 14 ..... 5.62 |================================================= simdjson 2.0 Throughput Test: TopTweet GB/s > Higher Is Better AOCC 4.0 .......... 10.85 |==================================================== GCC 12.2 .......... 9.90 |=============================================== GCC 13.0 14 Nov ... 10.22 |================================================= LLVM Clang 14 ..... 10.60 |=================================================== simdjson 2.0 Throughput Test: LargeRandom GB/s > Higher Is Better AOCC 4.0 .......... 1.59 |============================================== GCC 12.2 .......... 1.83 |===================================================== GCC 13.0 14 Nov ... 1.80 |==================================================== LLVM Clang 14 ..... 1.56 |============================================= simdjson 2.0 Throughput Test: PartialTweets GB/s > Higher Is Better AOCC 4.0 .......... 11.82 |==================================================== GCC 12.2 .......... 9.85 |=========================================== GCC 13.0 14 Nov ... 9.63 |========================================== LLVM Clang 14 ..... 11.15 |================================================= simdjson 2.0 Throughput Test: DistinctUserID GB/s > Higher Is Better AOCC 4.0 .......... 11.97 |==================================================== GCC 12.2 .......... 9.80 |=========================================== GCC 13.0 14 Nov ... 9.28 |======================================== LLVM Clang 14 ..... 10.77 |=============================================== Sockperf 3.7 Test: Throughput Messages Per Second > Higher Is Better AOCC 4.0 .......... 879095 |================================================== GCC 12.2 .......... 889103 |================================================== GCC 13.0 14 Nov ... 900310 |=================================================== LLVM Clang 14 ..... 883756 |================================================== LLVM Clang 15.0.2 . 897661 |=================================================== SQLite Speedtest 3.30 Timed Time - Size 1,000 Seconds < Lower Is Better AOCC 4.0 .......... 34.77 |================================================== GCC 12.2 .......... 34.23 |================================================= GCC 13.0 14 Nov ... 34.73 |================================================== LLVM Clang 14 ..... 36.07 |==================================================== LLVM Clang 15.0.2 . 36.02 |==================================================== SVT-AV1 1.2 Encoder Mode: Preset 4 - Input: Bosphorus 4K Frames Per Second > Higher Is Better AOCC 4.0 .......... 3.078 |=================================================== GCC 12.2 .......... 3.067 |=================================================== GCC 13.0 14 Nov ... 3.109 |==================================================== LLVM Clang 14 ..... 3.020 |=================================================== LLVM Clang 15.0.2 . 3.058 |=================================================== SVT-AV1 1.2 Encoder Mode: Preset 8 - Input: Bosphorus 4K Frames Per Second > Higher Is Better AOCC 4.0 .......... 76.51 |==================================================== GCC 12.2 .......... 76.40 |==================================================== GCC 13.0 14 Nov ... 76.36 |==================================================== LLVM Clang 14 ..... 74.61 |=================================================== LLVM Clang 15.0.2 . 74.82 |=================================================== SVT-AV1 1.2 Encoder Mode: Preset 10 - Input: Bosphorus 4K Frames Per Second > Higher Is Better AOCC 4.0 .......... 131.29 |================================================== GCC 12.2 .......... 131.04 |================================================== GCC 13.0 14 Nov ... 132.75 |=================================================== LLVM Clang 14 ..... 130.11 |================================================== LLVM Clang 15.0.2 . 129.16 |================================================== SVT-AV1 1.2 Encoder Mode: Preset 12 - Input: Bosphorus 4K Frames Per Second > Higher Is Better AOCC 4.0 .......... 188.40 |=================================================== GCC 12.2 .......... 187.37 |=================================================== GCC 13.0 14 Nov ... 188.62 |=================================================== LLVM Clang 14 ..... 183.19 |================================================== LLVM Clang 15.0.2 . 184.15 |================================================== SVT-HEVC 1.5.0 Tuning: 1 - Input: Bosphorus 4K Frames Per Second > Higher Is Better AOCC 4.0 .......... 6.68 |===================================================== GCC 12.2 .......... 6.64 |===================================================== GCC 13.0 14 Nov ... 6.63 |===================================================== LLVM Clang 14 ..... 6.64 |===================================================== LLVM Clang 15.0.2 . 6.63 |===================================================== SVT-HEVC 1.5.0 Tuning: 7 - Input: Bosphorus 4K Frames Per Second > Higher Is Better AOCC 4.0 .......... 108.32 |=================================================== GCC 12.2 .......... 106.57 |================================================== GCC 13.0 14 Nov ... 107.16 |================================================== LLVM Clang 14 ..... 105.55 |================================================== LLVM Clang 15.0.2 . 105.69 |================================================== SVT-HEVC 1.5.0 Tuning: 10 - Input: Bosphorus 4K Frames Per Second > Higher Is Better AOCC 4.0 .......... 171.09 |================================================== GCC 12.2 .......... 172.53 |=================================================== GCC 13.0 14 Nov ... 173.36 |=================================================== LLVM Clang 14 ..... 170.79 |================================================== LLVM Clang 15.0.2 . 171.02 |================================================== SVT-VP9 0.3 Tuning: VMAF Optimized - Input: Bosphorus 4K Frames Per Second > Higher Is Better AOCC 4.0 .......... 112.44 |=================================================== GCC 12.2 .......... 109.95 |================================================== GCC 13.0 14 Nov ... 110.49 |================================================== LLVM Clang 14 ..... 111.35 |=================================================== LLVM Clang 15.0.2 . 110.81 |================================================== SVT-VP9 0.3 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 4K Frames Per Second > Higher Is Better AOCC 4.0 .......... 121.78 |=================================================== GCC 12.2 .......... 122.32 |=================================================== GCC 13.0 14 Nov ... 122.36 |=================================================== LLVM Clang 14 ..... 120.82 |================================================== LLVM Clang 15.0.2 . 121.26 |=================================================== SVT-VP9 0.3 Tuning: Visual Quality Optimized - Input: Bosphorus 4K Frames Per Second > Higher Is Better AOCC 4.0 .......... 111.78 |=================================================== GCC 12.2 .......... 111.60 |=================================================== GCC 13.0 14 Nov ... 111.61 |=================================================== LLVM Clang 14 ..... 110.25 |================================================== LLVM Clang 15.0.2 . 111.24 |=================================================== Tachyon 0.99.2 Total Time Seconds < Lower Is Better AOCC 4.0 .......... 55.49 |==================================================== GCC 12.2 .......... 55.48 |==================================================== GCC 13.0 14 Nov ... 55.28 |==================================================== LLVM Clang 14 ..... 54.97 |==================================================== LLVM Clang 15.0.2 . 55.08 |==================================================== TNN 0.3 Target: CPU - Model: DenseNet ms < Lower Is Better AOCC 4.0 .......... 2455.64 |=========================== GCC 12.2 .......... 4513.71 |================================================== LLVM Clang 14 ..... 3634.33 |======================================== LLVM Clang 15.0.2 . 2499.82 |============================ TNN 0.3 Target: CPU - Model: MobileNet v2 ms < Lower Is Better AOCC 4.0 .......... 244.05 |============================= GCC 12.2 .......... 412.94 |================================================= LLVM Clang 14 ..... 431.56 |=================================================== LLVM Clang 15.0.2 . 337.48 |======================================== TNN 0.3 Target: CPU - Model: SqueezeNet v2 ms < Lower Is Better AOCC 4.0 .......... 44.59 |================================= GCC 12.2 .......... 69.92 |==================================================== LLVM Clang 14 ..... 61.11 |============================================= LLVM Clang 15.0.2 . 49.43 |===================================== TNN 0.3 Target: CPU - Model: SqueezeNet v1.1 ms < Lower Is Better AOCC 4.0 .......... 217.84 |============================= GCC 12.2 .......... 388.44 |=================================================== LLVM Clang 14 ..... 302.02 |======================================== LLVM Clang 15.0.2 . 232.44 |=============================== TSCP 1.81 AI Chess Performance Nodes Per Second > Higher Is Better AOCC 4.0 .......... 2477063 |========================================== GCC 12.2 .......... 2496821 |========================================== GCC 13.0 14 Nov ... 2492228 |========================================== LLVM Clang 14 ..... 2942579 |================================================== LLVM Clang 15.0.2 . 2857022 |================================================= WebP Image Encode 1.2.4 Encode Settings: Default MP/s > Higher Is Better AOCC 4.0 .......... 29.65 |==================================================== GCC 12.2 .......... 28.02 |================================================= GCC 13.0 14 Nov ... 28.18 |================================================= LLVM Clang 14 ..... 29.45 |==================================================== LLVM Clang 15.0.2 . 29.23 |=================================================== WebP Image Encode 1.2.4 Encode Settings: Quality 100 MP/s > Higher Is Better AOCC 4.0 .......... 17.67 |==================================================== GCC 12.2 .......... 16.61 |================================================= GCC 13.0 14 Nov ... 16.80 |================================================= LLVM Clang 14 ..... 17.51 |==================================================== LLVM Clang 15.0.2 . 17.00 |================================================== WebP Image Encode 1.2.4 Encode Settings: Quality 100, Lossless MP/s > Higher Is Better AOCC 4.0 .......... 2.17 |================================================== GCC 12.2 .......... 2.23 |==================================================== GCC 13.0 14 Nov ... 2.28 |===================================================== LLVM Clang 14 ..... 2.06 |================================================ LLVM Clang 15.0.2 . 2.15 |================================================== WebP Image Encode 1.2.4 Encode Settings: Quality 100, Highest Compression MP/s > Higher Is Better AOCC 4.0 .......... 6.33 |===================================================== GCC 12.2 .......... 4.48 |====================================== GCC 13.0 14 Nov ... 4.60 |======================================= LLVM Clang 14 ..... 6.16 |==================================================== LLVM Clang 15.0.2 . 6.21 |==================================================== WebP Image Encode 1.2.4 Encode Settings: Quality 100, Lossless, Highest Compression MP/s > Higher Is Better AOCC 4.0 .......... 0.89 |==================================================== GCC 12.2 .......... 0.88 |==================================================== GCC 13.0 14 Nov ... 0.89 |==================================================== LLVM Clang 14 ..... 0.83 |================================================= LLVM Clang 15.0.2 . 0.90 |===================================================== Xsbench 2017-07-06 Lookups/s > Higher Is Better AOCC 4.0 .......... 7081346 |================================================== GCC 12.2 .......... 6234508 |============================================ GCC 13.0 14 Nov ... 6318262 |============================================= LLVM Clang 14 ..... 6756001 |================================================ LLVM Clang 15.0.2 . 6752238 |================================================ yquake2 8.10 Renderer: Software CPU - AF: On - MSAA: On - Resolution: 1920 x 1080 Frames Per Second > Higher Is Better AOCC 4.0 .......... 196.3 |==================================================== GCC 12.2 .......... 185.1 |================================================= GCC 13.0 14 Nov ... 191.4 |=================================================== yquake2 8.10 Renderer: Software CPU - AF: Off - MSAA: On - Resolution: 1920 x 1080 Frames Per Second > Higher Is Better AOCC 4.0 .......... 205.6 |==================================================== GCC 12.2 .......... 192.0 |================================================= GCC 13.0 14 Nov ... 199.0 |================================================== yquake2 8.10 Renderer: Software CPU - AF: On - MSAA: Off - Resolution: 1920 x 1080 Frames Per Second > Higher Is Better AOCC 4.0 .......... 198.1 |==================================================== GCC 12.2 .......... 185.9 |================================================= GCC 13.0 14 Nov ... 192.3 |================================================== yquake2 8.10 Renderer: Software CPU - AF: Off - MSAA: Off - Resolution: 1920 x 1080 Frames Per Second > Higher Is Better AOCC 4.0 .......... 205.7 |==================================================== GCC 12.2 .......... 192.2 |================================================= GCC 13.0 14 Nov ... 201.0 |=================================================== yquake2 8.10 Renderer: Software CPU Color Light - AF: On - MSAA: On - Resolution: 1920 x 1080 Frames Per Second > Higher Is Better AOCC 4.0 .......... 185.3 |==================================================== GCC 12.2 .......... 171.2 |================================================ GCC 13.0 14 Nov ... 181.3 |=================================================== yquake2 8.10 Renderer: Software CPU Color Light - AF: Off - MSAA: On - Resolution: 1920 x 1080 Frames Per Second > Higher Is Better AOCC 4.0 .......... 189.9 |==================================================== GCC 12.2 .......... 176.2 |================================================ GCC 13.0 14 Nov ... 184.4 |================================================== yquake2 8.10 Renderer: Software CPU Color Light - AF: On - MSAA: Off - Resolution: 1920 x 1080 Frames Per Second > Higher Is Better AOCC 4.0 .......... 188.2 |==================================================== GCC 12.2 .......... 172.5 |================================================ GCC 13.0 14 Nov ... 178.5 |================================================= yquake2 8.10 Renderer: Software CPU Color Light - AF: Off - MSAA: Off - Resolution: 1920 x 1080 Frames Per Second > Higher Is Better AOCC 4.0 .......... 191.8 |==================================================== GCC 12.2 .......... 176.4 |================================================ GCC 13.0 14 Nov ... 185.3 |================================================== Zstd Compression 1.5.0 Compression Level: 3 - Compression Speed MB/s > Higher Is Better AOCC 4.0 .......... 5092.2 |=================================================== GCC 12.2 .......... 4978.6 |================================================== GCC 13.0 14 Nov ... 4976.2 |================================================== LLVM Clang 14 ..... 5045.2 |=================================================== LLVM Clang 15.0.2 . 5030.8 |================================================== Zstd Compression 1.5.0 Compression Level: 3 - Decompression Speed MB/s > Higher Is Better AOCC 4.0 .......... 5461.2 |=================================================== GCC 12.2 .......... 5432.1 |=================================================== GCC 13.0 14 Nov ... 5485.8 |=================================================== LLVM Clang 14 ..... 5463.5 |=================================================== LLVM Clang 15.0.2 . 5458.6 |=================================================== Zstd Compression 1.5.0 Compression Level: 19 - Compression Speed MB/s > Higher Is Better AOCC 4.0 .......... 79.1 |=================================================== GCC 12.2 .......... 80.3 |==================================================== GCC 13.0 14 Nov ... 82.2 |===================================================== LLVM Clang 14 ..... 78.9 |=================================================== LLVM Clang 15.0.2 . 78.8 |=================================================== Zstd Compression 1.5.0 Compression Level: 19 - Decompression Speed MB/s > Higher Is Better AOCC 4.0 .......... 5018.6 |=================================================== GCC 12.2 .......... 4932.0 |================================================== GCC 13.0 14 Nov ... 4988.9 |=================================================== LLVM Clang 14 ..... 4913.2 |================================================== LLVM Clang 15.0.2 . 4894.6 |================================================== Zstd Compression 1.5.0 Compression Level: 19, Long Mode - Compression Speed MB/s > Higher Is Better AOCC 4.0 .......... 54.6 |==================================================== GCC 12.2 .......... 54.6 |==================================================== GCC 13.0 14 Nov ... 54.3 |==================================================== LLVM Clang 14 ..... 54.6 |==================================================== LLVM Clang 15.0.2 . 55.3 |===================================================== Zstd Compression 1.5.0 Compression Level: 19, Long Mode - Decompression Speed MB/s > Higher Is Better AOCC 4.0 .......... 5117.8 |=================================================== GCC 12.2 .......... 4974.4 |================================================== GCC 13.0 14 Nov ... 5073.8 |=================================================== LLVM Clang 14 ..... 5055.8 |================================================== LLVM Clang 15.0.2 . 4928.9 |=================================================