dddas AMD Ryzen Threadripper 3970X 32-Core testing with a ASUS ROG ZENITH II EXTREME (1603 BIOS) and AMD Radeon RX 5700 8GB on Ubuntu 22.04 via the Phoronix Test Suite. a: Processor: AMD Ryzen Threadripper 3970X 32-Core @ 3.70GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH II EXTREME (1603 BIOS), Chipset: AMD Starship/Matisse, Memory: 64GB, Disk: Samsung SSD 980 PRO 500GB, Graphics: AMD Radeon RX 5700 8GB (1750/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: ASUS VP28U, Network: Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200 OS: Ubuntu 22.04, Kernel: 5.19.0-051900rc7-generic (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.1 (LLVM 13.0.1 DRM 3.47), Vulkan: 1.2.204, Compiler: GCC 11.3.0, File-System: ext4, Screen Resolution: 3840x2160 b: Processor: AMD Ryzen Threadripper 3970X 32-Core @ 3.70GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH II EXTREME (1603 BIOS), Chipset: AMD Starship/Matisse, Memory: 64GB, Disk: Samsung SSD 980 PRO 500GB, Graphics: AMD Radeon RX 5700 8GB (1750/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: ASUS VP28U, Network: Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200 OS: Ubuntu 22.04, Kernel: 5.19.0-051900rc7-generic (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.1 (LLVM 13.0.1 DRM 3.47), Vulkan: 1.2.204, Compiler: GCC 11.3.0, File-System: ext4, Screen Resolution: 3840x2160 CP2K Molecular Dynamics 2023.1 Input: H20-64 Seconds < Lower Is Better a . 42.97 |==================================================================== b . 42.11 |=================================================================== CP2K Molecular Dynamics 2023.1 Input: H2O-DFT-LS Seconds < Lower Is Better CP2K Molecular Dynamics 2023.1 Input: Fayalite-FIST Seconds < Lower Is Better a . 123.83 |=================================================================== b . 122.98 |=================================================================== dav1d 1.2.1 Video Input: Chimera 1080p FPS > Higher Is Better a . 398.39 |=================================================================== b . 398.20 |=================================================================== dav1d 1.2.1 Video Input: Summer Nature 4K FPS > Higher Is Better a . 222.52 |=================================================================== b . 222.24 |=================================================================== dav1d 1.2.1 Video Input: Summer Nature 1080p FPS > Higher Is Better a . 597.02 |=================================================================== b . 597.19 |=================================================================== dav1d 1.2.1 Video Input: Chimera 1080p 10-bit FPS > Higher Is Better a . 374.79 |=================================================================== b . 374.13 |=================================================================== Embree 4.1 Binary: Pathtracer - Model: Crown Frames Per Second > Higher Is Better a . 38.47 |==================================================================== b . 38.45 |==================================================================== Embree 4.1 Binary: Pathtracer ISPC - Model: Crown Frames Per Second > Higher Is Better a . 34.41 |==================================================================== b . 34.53 |==================================================================== Embree 4.1 Binary: Pathtracer - Model: Asian Dragon Frames Per Second > Higher Is Better a . 41.59 |==================================================================== b . 41.78 |==================================================================== Embree 4.1 Binary: Pathtracer - Model: Asian Dragon Obj Frames Per Second > Higher Is Better a . 37.39 |==================================================================== b . 37.48 |==================================================================== Embree 4.1 Binary: Pathtracer ISPC - Model: Asian Dragon Frames Per Second > Higher Is Better a . 39.40 |==================================================================== b . 39.40 |==================================================================== Embree 4.1 Binary: Pathtracer ISPC - Model: Asian Dragon Obj Frames Per Second > Higher Is Better a . 33.83 |==================================================================== b . 33.96 |==================================================================== eSpeak-NG Speech Engine 1.51 Text-To-Speech Synthesis Seconds < Lower Is Better a . 31.08 |=================================================================== b . 31.49 |==================================================================== GPAW 23.6 Input: Carbon Nanotube Seconds < Lower Is Better a . 110.85 |=================================================================== b . 110.95 |=================================================================== HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: c2c - Backend: FFTW - Precision: double-long - X Y Z: 128 GFLOP/s > Higher Is Better a . 30.80 |==================================================================== b . 30.86 |==================================================================== HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: c2c - Backend: FFTW - Precision: double-long - X Y Z: 256 GFLOP/s > Higher Is Better a . 13.76 |==================================================================== b . 13.79 |==================================================================== HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: c2c - Backend: FFTW - Precision: double-long - X Y Z: 512 GFLOP/s > Higher Is Better a . 15.35 |==================================================================== b . 15.35 |==================================================================== HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: r2c - Backend: FFTW - Precision: double-long - X Y Z: 128 GFLOP/s > Higher Is Better a . 56.45 |==================================================================== b . 55.85 |=================================================================== HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: r2c - Backend: FFTW - Precision: double-long - X Y Z: 256 GFLOP/s > Higher Is Better a . 27.43 |==================================================================== b . 27.27 |==================================================================== HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: r2c - Backend: FFTW - Precision: double-long - X Y Z: 512 GFLOP/s > Higher Is Better a . 27.71 |==================================================================== b . 27.73 |==================================================================== HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: c2c - Backend: Stock - Precision: double-long - X Y Z: 128 GFLOP/s > Higher Is Better a . 26.55 |=================================================================== b . 26.81 |==================================================================== HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: c2c - Backend: Stock - Precision: double-long - X Y Z: 256 GFLOP/s > Higher Is Better a . 13.88 |==================================================================== b . 13.85 |==================================================================== HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: c2c - Backend: Stock - Precision: double-long - X Y Z: 512 GFLOP/s > Higher Is Better a . 15.41 |==================================================================== b . 15.41 |==================================================================== HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: r2c - Backend: Stock - Precision: double-long - X Y Z: 128 GFLOP/s > Higher Is Better a . 51.88 |==================================================================== b . 50.93 |=================================================================== HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: r2c - Backend: Stock - Precision: double-long - X Y Z: 256 GFLOP/s > Higher Is Better a . 30.14 |==================================================================== b . 30.29 |==================================================================== HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: r2c - Backend: Stock - Precision: double-long - X Y Z: 512 GFLOP/s > Higher Is Better a . 30.01 |==================================================================== b . 30.04 |==================================================================== High Performance Conjugate Gradient 3.1 X Y Z: 104 104 104 - RT: 60 GFLOP/s > Higher Is Better a . 10.96 |==================================================================== b . 11.02 |==================================================================== High Performance Conjugate Gradient 3.1 X Y Z: 144 144 144 - RT: 60 GFLOP/s > Higher Is Better Intel Open Image Denoise 2.0 Run: RT.hdr_alb_nrm.3840x2160 - Device: CPU-Only Images / Sec > Higher Is Better a . 1.22 |===================================================================== b . 1.22 |===================================================================== Intel Open Image Denoise 2.0 Run: RT.ldr_alb_nrm.3840x2160 - Device: CPU-Only Images / Sec > Higher Is Better a . 1.22 |==================================================================== b . 1.23 |===================================================================== Intel Open Image Denoise 2.0 Run: RTLightmap.hdr.4096x4096 - Device: CPU-Only Images / Sec > Higher Is Better a . 0.60 |===================================================================== b . 0.60 |===================================================================== Intel Open Image Denoise 2.0 Run: RT.hdr_alb_nrm.3840x2160 - Device: Radeon HIP Images / Sec > Higher Is Better Intel Open Image Denoise 2.0 Run: RT.ldr_alb_nrm.3840x2160 - Device: Radeon HIP Images / Sec > Higher Is Better Intel Open Image Denoise 2.0 Run: RTLightmap.hdr.4096x4096 - Device: Radeon HIP Images / Sec > Higher Is Better Intel Open Image Denoise 2.0 Run: RT.hdr_alb_nrm.3840x2160 - Device: Intel oneAPI SYCL Images / Sec > Higher Is Better Intel Open Image Denoise 2.0 Run: RT.ldr_alb_nrm.3840x2160 - Device: Intel oneAPI SYCL Images / Sec > Higher Is Better Intel Open Image Denoise 2.0 Run: RTLightmap.hdr.4096x4096 - Device: Intel oneAPI SYCL Images / Sec > Higher Is Better Kripke 1.2.6 Throughput FoM > Higher Is Better a . 148243333 |================================================================ b . 146215600 |=============================================================== Laghos 3.1 Test: Triple Point Problem Major Kernels Total Rate > Higher Is Better a . 220.46 |=================================================================== b . 219.12 |=================================================================== Laghos 3.1 Test: Sedov Blast Wave, ube_922_hex.mesh Major Kernels Total Rate > Higher Is Better a . 264.34 |=================================================================== b . 265.39 |=================================================================== LevelDB 1.23 Benchmark: Hot Read Microseconds Per Op < Lower Is Better a . 43.14 |==================================================================== b . 42.76 |=================================================================== LevelDB 1.23 Benchmark: Fill Sync MB/s > Higher Is Better a . 0.6 |====================================================================== b . 0.4 |=============================================== LevelDB 1.23 Benchmark: Fill Sync Microseconds Per Op < Lower Is Better a . 10866.01 |=========================================== b . 16348.37 |================================================================= LevelDB 1.23 Benchmark: Overwrite MB/s > Higher Is Better a . 27.0 |===================================================================== b . 27.0 |===================================================================== LevelDB 1.23 Benchmark: Overwrite Microseconds Per Op < Lower Is Better a . 262.35 |=================================================================== b . 262.28 |=================================================================== LevelDB 1.23 Benchmark: Random Fill MB/s > Higher Is Better a . 26.9 |===================================================================== b . 26.7 |==================================================================== LevelDB 1.23 Benchmark: Random Fill Microseconds Per Op < Lower Is Better a . 262.98 |=================================================================== b . 264.87 |=================================================================== LevelDB 1.23 Benchmark: Random Read Microseconds Per Op < Lower Is Better a . 43.49 |==================================================================== b . 43.55 |==================================================================== LevelDB 1.23 Benchmark: Seek Random Microseconds Per Op < Lower Is Better a . 65.84 |==================================================================== b . 64.56 |=================================================================== LevelDB 1.23 Benchmark: Random Delete Microseconds Per Op < Lower Is Better a . 245.16 |=================================================================== b . 245.18 |=================================================================== LevelDB 1.23 Benchmark: Sequential Fill MB/s > Higher Is Better a . 27.8 |===================================================================== b . 27.7 |===================================================================== LevelDB 1.23 Benchmark: Sequential Fill Microseconds Per Op < Lower Is Better a . 254.94 |=================================================================== b . 255.49 |=================================================================== libxsmm 2-1.17-3645 M N K: 128 GFLOPS/s > Higher Is Better a . 635.8 |==================================================================== b . 635.4 |==================================================================== libxsmm 2-1.17-3645 M N K: 256 GFLOPS/s > Higher Is Better a . 910.4 |==================================================================== b . 907.4 |==================================================================== libxsmm 2-1.17-3645 M N K: 32 GFLOPS/s > Higher Is Better a . 160.5 |==================================================================== b . 160.7 |==================================================================== libxsmm 2-1.17-3645 M N K: 64 GFLOPS/s > Higher Is Better a . 318.5 |==================================================================== b . 318.7 |==================================================================== Liquid-DSP 1.6 Threads: 1 - Buffer Length: 256 - Filter Length: 32 samples/s > Higher Is Better a . 45075000 |================================================================= b . 45023000 |================================================================= Liquid-DSP 1.6 Threads: 1 - Buffer Length: 256 - Filter Length: 57 samples/s > Higher Is Better a . 51993333 |================================================================= b . 51814000 |================================================================= Liquid-DSP 1.6 Threads: 2 - Buffer Length: 256 - Filter Length: 32 samples/s > Higher Is Better a . 89896333 |================================================================= b . 89686000 |================================================================= Liquid-DSP 1.6 Threads: 2 - Buffer Length: 256 - Filter Length: 57 samples/s > Higher Is Better a . 103806667 |================================================================ b . 103260000 |================================================================ Liquid-DSP 1.6 Threads: 4 - Buffer Length: 256 - Filter Length: 32 samples/s > Higher Is Better a . 178100000 |================================================================ b . 179170000 |================================================================ Liquid-DSP 1.6 Threads: 4 - Buffer Length: 256 - Filter Length: 57 samples/s > Higher Is Better a . 206086667 |================================================================ b . 206140000 |================================================================ Liquid-DSP 1.6 Threads: 8 - Buffer Length: 256 - Filter Length: 32 samples/s > Higher Is Better a . 354570000 |================================================================ b . 355110000 |================================================================ Liquid-DSP 1.6 Threads: 8 - Buffer Length: 256 - Filter Length: 57 samples/s > Higher Is Better a . 409183333 |================================================================ b . 410090000 |================================================================ Liquid-DSP 1.6 Threads: 1 - Buffer Length: 256 - Filter Length: 512 samples/s > Higher Is Better a . 10537667 |================================================================= b . 10560000 |================================================================= Liquid-DSP 1.6 Threads: 16 - Buffer Length: 256 - Filter Length: 32 samples/s > Higher Is Better a . 690800000 |================================================================ b . 692130000 |================================================================ Liquid-DSP 1.6 Threads: 16 - Buffer Length: 256 - Filter Length: 57 samples/s > Higher Is Better a . 795103333 |================================================================ b . 799490000 |================================================================ Liquid-DSP 1.6 Threads: 2 - Buffer Length: 256 - Filter Length: 512 samples/s > Higher Is Better a . 20851667 |================================================================= b . 20989000 |================================================================= Liquid-DSP 1.6 Threads: 32 - Buffer Length: 256 - Filter Length: 32 samples/s > Higher Is Better a . 1343200000 |=============================================================== b . 1350000000 |=============================================================== Liquid-DSP 1.6 Threads: 32 - Buffer Length: 256 - Filter Length: 57 samples/s > Higher Is Better a . 1506266667 |=============================================================== b . 1512100000 |=============================================================== Liquid-DSP 1.6 Threads: 4 - Buffer Length: 256 - Filter Length: 512 samples/s > Higher Is Better a . 41277667 |================================================================= b . 41475000 |================================================================= Liquid-DSP 1.6 Threads: 64 - Buffer Length: 256 - Filter Length: 32 samples/s > Higher Is Better a . 2250733333 |============================================================== b . 2269700000 |=============================================================== Liquid-DSP 1.6 Threads: 64 - Buffer Length: 256 - Filter Length: 57 samples/s > Higher Is Better a . 1836033333 |=============================================================== b . 1837000000 |=============================================================== Liquid-DSP 1.6 Threads: 8 - Buffer Length: 256 - Filter Length: 512 samples/s > Higher Is Better a . 82123667 |================================================================= b . 82224000 |================================================================= Liquid-DSP 1.6 Threads: 16 - Buffer Length: 256 - Filter Length: 512 samples/s > Higher Is Better a . 160113333 |================================================================ b . 160130000 |================================================================ Liquid-DSP 1.6 Threads: 32 - Buffer Length: 256 - Filter Length: 512 samples/s > Higher Is Better a . 313753333 |================================================================ b . 314560000 |================================================================ Liquid-DSP 1.6 Threads: 64 - Buffer Length: 256 - Filter Length: 512 samples/s > Higher Is Better a . 506326667 |================================================================ b . 506050000 |================================================================ Monte Carlo Simulations of Ionised Nebulae 2.02.73.3 Input: Gas HII40 Seconds < Lower Is Better a . 12.68 |==================================================================== b . 12.60 |==================================================================== Monte Carlo Simulations of Ionised Nebulae 2.02.73.3 Input: Dust 2D tau100.0 Seconds < Lower Is Better a . 181.27 |=================================================================== b . 180.73 |=================================================================== nekRS 23.0 Input: Kershaw flops/rank > Higher Is Better a . 2123046667 |=============================================================== b . 2109640000 |=============================================================== nekRS 23.0 Input: TurboPipe Periodic flops/rank > Higher Is Better a . 3444566667 |=============================================================== b . 3441770000 |=============================================================== Neural Magic DeepSparse 1.5 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 28.88 |==================================================================== b . 28.92 |==================================================================== Neural Magic DeepSparse 1.5 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 552.28 |=================================================================== b . 551.16 |=================================================================== Neural Magic DeepSparse 1.5 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 16.35 |=================================================================== b . 16.48 |==================================================================== Neural Magic DeepSparse 1.5 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 61.17 |==================================================================== b . 60.66 |=================================================================== Neural Magic DeepSparse 1.5 Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 255.99 |=================================================================== b . 256.64 |=================================================================== Neural Magic DeepSparse 1.5 Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 62.44 |==================================================================== b . 62.31 |==================================================================== Neural Magic DeepSparse 1.5 Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 77.58 |=================================================================== b . 78.40 |==================================================================== Neural Magic DeepSparse 1.5 Model: NLP Sentiment Analysis, 80% Pruned Quantized BERT Base Uncased - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 12.88 |==================================================================== b . 12.75 |=================================================================== Neural Magic DeepSparse 1.5 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 86.13 |=================================================================== b . 87.02 |==================================================================== Neural Magic DeepSparse 1.5 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 185.74 |=================================================================== b . 183.81 |================================================================== Neural Magic DeepSparse 1.5 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 28.73 |================================================================= b . 29.86 |==================================================================== Neural Magic DeepSparse 1.5 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 34.80 |==================================================================== b . 33.48 |================================================================= Neural Magic DeepSparse 1.5 Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 148.97 |=================================================================== b . 149.96 |=================================================================== Neural Magic DeepSparse 1.5 Model: CV Detection, YOLOv5s COCO - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 107.37 |=================================================================== b . 106.66 |=================================================================== Neural Magic DeepSparse 1.5 Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 91.57 |==================================================================== b . 91.98 |==================================================================== Neural Magic DeepSparse 1.5 Model: CV Detection, YOLOv5s COCO - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 10.91 |==================================================================== b . 10.86 |==================================================================== Neural Magic DeepSparse 1.5 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 323.09 |=================================================================== b . 324.27 |=================================================================== Neural Magic DeepSparse 1.5 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 49.49 |==================================================================== b . 49.32 |==================================================================== Neural Magic DeepSparse 1.5 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 141.30 |================================================================= b . 146.19 |=================================================================== Neural Magic DeepSparse 1.5 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 7.0704 |=================================================================== b . 6.8330 |================================================================= Neural Magic DeepSparse 1.5 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 235.81 |=================================================================== b . 236.82 |=================================================================== Neural Magic DeepSparse 1.5 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 67.82 |==================================================================== b . 67.54 |==================================================================== Neural Magic DeepSparse 1.5 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 81.34 |================================================================== b . 83.43 |==================================================================== Neural Magic DeepSparse 1.5 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 12.29 |==================================================================== b . 11.98 |================================================================== Neural Magic DeepSparse 1.5 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 33.10 |==================================================================== b . 32.91 |==================================================================== Neural Magic DeepSparse 1.5 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 482.75 |=================================================================== b . 485.95 |=================================================================== Neural Magic DeepSparse 1.5 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 21.57 |==================================================================== b . 21.49 |==================================================================== Neural Magic DeepSparse 1.5 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 46.35 |==================================================================== b . 46.51 |==================================================================== Neural Magic DeepSparse 1.5 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 119.01 |=================================================================== b . 119.09 |=================================================================== Neural Magic DeepSparse 1.5 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 134.41 |=================================================================== b . 134.32 |=================================================================== Neural Magic DeepSparse 1.5 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 43.88 |==================================================================== b . 43.05 |=================================================================== Neural Magic DeepSparse 1.5 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 22.78 |=================================================================== b . 23.22 |==================================================================== Neural Magic DeepSparse 1.5 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 28.61 |=================================================================== b . 28.99 |==================================================================== Neural Magic DeepSparse 1.5 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 556.61 |=================================================================== b . 550.43 |================================================================== Neural Magic DeepSparse 1.5 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 16.34 |==================================================================== b . 16.43 |==================================================================== Neural Magic DeepSparse 1.5 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 61.20 |==================================================================== b . 60.86 |==================================================================== oneDNN 3.1 Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU ms < Lower Is Better a . 1.55099 |================================================================== b . 1.30181 |======================================================= oneDNN 3.1 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU ms < Lower Is Better a . 4.26624 |================================================================== b . 4.21682 |================================================================= oneDNN 3.1 Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a . 1.177229 |================================================================= b . 1.089440 |============================================================ oneDNN 3.1 Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a . 0.948450 |=============================================================== b . 0.985098 |================================================================= oneDNN 3.1 Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better oneDNN 3.1 Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better oneDNN 3.1 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU ms < Lower Is Better a . 4.81893 |================================================================== b . 4.83565 |================================================================== oneDNN 3.1 Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU ms < Lower Is Better a . 5.69206 |================================================================== b . 5.73181 |================================================================== oneDNN 3.1 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU ms < Lower Is Better a . 2.68566 |================================================================== b . 2.69872 |================================================================== oneDNN 3.1 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a . 5.76769 |================================================================== b . 5.70574 |================================================================= oneDNN 3.1 Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a . 1.36630 |=============================================================== b . 1.43664 |================================================================== oneDNN 3.1 Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a . 1.57740 |================================================================== b . 1.57190 |================================================================== oneDNN 3.1 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU ms < Lower Is Better a . 3252.18 |================================================================== b . 3275.76 |================================================================== oneDNN 3.1 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU ms < Lower Is Better a . 976.47 |================================================================== b . 987.99 |=================================================================== oneDNN 3.1 Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a . 3244.46 |================================================================== b . 3227.02 |================================================================== oneDNN 3.1 Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better oneDNN 3.1 Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better oneDNN 3.1 Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better oneDNN 3.1 Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a . 938.10 |================================================================== b . 958.84 |=================================================================== oneDNN 3.1 Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better a . 3235.41 |================================================================== b . 3226.70 |================================================================== oneDNN 3.1 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better a . 935.61 |=================================================================== b . 932.99 |=================================================================== Opus Codec Encoding 1.4 WAV To Opus Encode Seconds < Lower Is Better a . 28.70 |==================================================================== b . 28.83 |==================================================================== OSPRay 2.12 Benchmark: particle_volume/ao/real_time Items Per Second > Higher Is Better a . 9.86893 |================================================================== b . 9.89124 |================================================================== OSPRay 2.12 Benchmark: particle_volume/scivis/real_time Items Per Second > Higher Is Better a . 9.74548 |================================================================== b . 9.76771 |================================================================== OSPRay 2.12 Benchmark: particle_volume/pathtracer/real_time Items Per Second > Higher Is Better a . 128.57 |=================================================================== b . 128.66 |=================================================================== OSPRay 2.12 Benchmark: gravity_spheres_volume/dim_512/ao/real_time Items Per Second > Higher Is Better a . 4.93554 |================================================================= b . 4.98028 |================================================================== OSPRay 2.12 Benchmark: gravity_spheres_volume/dim_512/scivis/real_time Items Per Second > Higher Is Better a . 4.62468 |================================================================== b . 4.62434 |================================================================== OSPRay 2.12 Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time Items Per Second > Higher Is Better a . 7.67668 |================================================================== b . 7.70051 |================================================================== Palabos 2.3 Grid Size: 100 Mega Site Updates Per Second > Higher Is Better a . 121.93 |=================================================================== b . 122.23 |=================================================================== Palabos 2.3 Grid Size: 400 Mega Site Updates Per Second > Higher Is Better a . 139.30 |=================================================================== b . 140.08 |=================================================================== Palabos 2.3 Grid Size: 500 Mega Site Updates Per Second > Higher Is Better a . 143.85 |=================================================================== b . 144.06 |=================================================================== Palabos 2.3 Grid Size: 1000 Mega Site Updates Per Second > Higher Is Better Palabos 2.3 Grid Size: 4000 Mega Site Updates Per Second > Higher Is Better PETSc 3.19 Test: Streams MB/s > Higher Is Better a . 58312.10 |================================================================= b . 58276.79 |================================================================= QMCPACK 3.16 Input: Li2_STO_ae Total Execution Time - Seconds < Lower Is Better a . 136.22 |=================================================================== b . 132.76 |================================================================= QMCPACK 3.16 Input: simple-H2O Total Execution Time - Seconds < Lower Is Better a . 27.60 |==================================================================== b . 27.48 |==================================================================== QMCPACK 3.16 Input: FeCO6_b3lyp_gms Total Execution Time - Seconds < Lower Is Better a . 175.39 |=================================================================== b . 174.82 |=================================================================== QMCPACK 3.16 Input: FeCO6_b3lyp_gms Total Execution Time - Seconds < Lower Is Better a . 196.98 |=================================================================== b . 191.15 |================================================================= Remhos 1.0 Test: Sample Remap Example Seconds < Lower Is Better a . 23.54 |==================================================================== b . 23.65 |==================================================================== SQLite 3.41.2 Threads / Copies: 1 Seconds < Lower Is Better a . 106.01 |=================================================================== b . 105.00 |================================================================== SQLite 3.41.2 Threads / Copies: 2 Seconds < Lower Is Better a . 243.22 |=================================================================== b . 237.19 |================================================================= SQLite 3.41.2 Threads / Copies: 4 Seconds < Lower Is Better a . 266.58 |=================================================================== b . 262.62 |================================================================== SQLite 3.41.2 Threads / Copies: 8 Seconds < Lower Is Better a . 291.25 |=================================================================== b . 284.87 |================================================================== SQLite 3.41.2 Threads / Copies: 16 Seconds < Lower Is Better a . 373.82 |=================================================================== b . 374.33 |=================================================================== SQLite 3.41.2 Threads / Copies: 32 Seconds < Lower Is Better a . 505.42 |=================================================================== b . 502.77 |=================================================================== SQLite 3.41.2 Threads / Copies: 64 Seconds < Lower Is Better a . 681.45 |=================================================================== b . 680.81 |=================================================================== Stress-NG 0.15.10 Test: Hash Bogo Ops/s > Higher Is Better a . 7627578.66 |=============================================================== b . 7624159.24 |=============================================================== Stress-NG 0.15.10 Test: MMAP Bogo Ops/s > Higher Is Better a . 437.11 |=================================================================== b . 439.24 |=================================================================== Stress-NG 0.15.10 Test: NUMA Bogo Ops/s > Higher Is Better a . 752.30 |=================================================================== b . 741.66 |================================================================== Stress-NG 0.15.10 Test: Pipe Bogo Ops/s > Higher Is Better a . 18809740.35 |===================================================== b . 22201912.16 |============================================================== Stress-NG 0.15.10 Test: Poll Bogo Ops/s > Higher Is Better a . 4084623.29 |=============================================================== b . 4101817.96 |=============================================================== Stress-NG 0.15.10 Test: Zlib Bogo Ops/s > Higher Is Better a . 4517.78 |================================================================== b . 4518.88 |================================================================== Stress-NG 0.15.10 Test: Futex Bogo Ops/s > Higher Is Better a . 4610857.40 |=============================================================== b . 4644715.00 |=============================================================== Stress-NG 0.15.10 Test: MEMFD Bogo Ops/s > Higher Is Better a . 395.11 |=================================================================== b . 394.50 |=================================================================== Stress-NG 0.15.10 Test: Mutex Bogo Ops/s > Higher Is Better a . 18827346.28 |============================================================== b . 18816044.49 |============================================================== Stress-NG 0.15.10 Test: Atomic Bogo Ops/s > Higher Is Better a . 480.06 |=================================================================== b . 480.51 |=================================================================== Stress-NG 0.15.10 Test: Crypto Bogo Ops/s > Higher Is Better a . 78260.17 |================================================================= b . 78455.19 |================================================================= Stress-NG 0.15.10 Test: Malloc Bogo Ops/s > Higher Is Better a . 92853207.13 |============================================================== b . 92812375.37 |============================================================== Stress-NG 0.15.10 Test: Cloning Bogo Ops/s > Higher Is Better a . 3354.40 |================================================================== b . 3360.52 |================================================================== Stress-NG 0.15.10 Test: Forking Bogo Ops/s > Higher Is Better a . 51344.69 |================================================================= b . 51160.22 |================================================================= Stress-NG 0.15.10 Test: Pthread Bogo Ops/s > Higher Is Better a . 128353.64 |================================================================ b . 128387.57 |================================================================ Stress-NG 0.15.10 Test: AVL Tree Bogo Ops/s > Higher Is Better a . 283.41 |=================================================================== b . 282.42 |=================================================================== Stress-NG 0.15.10 Test: IO_uring Bogo Ops/s > Higher Is Better a . 439798.24 |================================================================ b . 440335.12 |================================================================ Stress-NG 0.15.10 Test: SENDFILE Bogo Ops/s > Higher Is Better a . 515575.47 |============================================================== b . 528847.31 |================================================================ Stress-NG 0.15.10 Test: CPU Cache Bogo Ops/s > Higher Is Better a . 1624118.54 |=============================================================== b . 1535034.64 |============================================================ Stress-NG 0.15.10 Test: CPU Stress Bogo Ops/s > Higher Is Better a . 82729.76 |================================================================= b . 82887.24 |================================================================= Stress-NG 0.15.10 Test: Semaphores Bogo Ops/s > Higher Is Better a . 66510329.66 |========================================================== b . 71041068.48 |============================================================== Stress-NG 0.15.10 Test: Matrix Math Bogo Ops/s > Higher Is Better a . 199178.68 |================================================================ b . 200423.60 |================================================================ Stress-NG 0.15.10 Test: Vector Math Bogo Ops/s > Higher Is Better a . 224417.23 |================================================================ b . 224460.71 |================================================================ Stress-NG 0.15.10 Test: Function Call Bogo Ops/s > Higher Is Better a . 24278.34 |================================================================= b . 24275.23 |================================================================= Stress-NG 0.15.10 Test: x86_64 RdRand Stress-NG 0.15.10 Test: Floating Point Bogo Ops/s > Higher Is Better a . 11201.44 |================================================================= b . 11221.55 |================================================================= Stress-NG 0.15.10 Test: Matrix 3D Math Bogo Ops/s > Higher Is Better a . 2806.09 |================================================================== b . 2795.85 |================================================================== Stress-NG 0.15.10 Test: Memory Copying Bogo Ops/s > Higher Is Better a . 10973.65 |================================================================= b . 10984.91 |================================================================= Stress-NG 0.15.10 Test: Vector Shuffle Bogo Ops/s > Higher Is Better a . 22825.44 |================================================================= b . 22200.86 |=============================================================== Stress-NG 0.15.10 Test: Socket Activity Bogo Ops/s > Higher Is Better a . 3072.80 |===================== b . 9580.27 |================================================================== Stress-NG 0.15.10 Test: Wide Vector Math Bogo Ops/s > Higher Is Better a . 1501239.29 |=============================================================== b . 1496970.66 |=============================================================== Stress-NG 0.15.10 Test: Context Switching Bogo Ops/s > Higher Is Better a . 11409509.77 |============================================================= b . 11620881.04 |============================================================== Stress-NG 0.15.10 Test: Fused Multiply-Add Bogo Ops/s > Higher Is Better a . 33507543.08 |============================================================== b . 33539318.97 |============================================================== Stress-NG 0.15.10 Test: Vector Floating Point Bogo Ops/s > Higher Is Better a . 94803.76 |================================================================ b . 95693.93 |================================================================= Stress-NG 0.15.10 Test: Glibc C String Functions Bogo Ops/s > Higher Is Better a . 33453867.32 |============================================================== b . 33092079.25 |============================================================= Stress-NG 0.15.10 Test: Glibc Qsort Data Sorting Bogo Ops/s > Higher Is Better a . 942.22 |=================================================================== b . 943.84 |=================================================================== Stress-NG 0.15.10 Test: System V Message Passing Bogo Ops/s > Higher Is Better a . 10692419.88 |============================================================== b . 10677047.79 |============================================================== SVT-AV1 1.6 Encoder Mode: Preset 4 - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 3.756 |==================================================================== b . 3.721 |=================================================================== SVT-AV1 1.6 Encoder Mode: Preset 8 - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 54.15 |=================================================================== b . 54.56 |==================================================================== SVT-AV1 1.6 Encoder Mode: Preset 12 - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 126.42 |================================================================== b . 127.68 |=================================================================== SVT-AV1 1.6 Encoder Mode: Preset 13 - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 127.14 |=================================================================== b . 127.68 |=================================================================== SVT-AV1 1.6 Encoder Mode: Preset 4 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 10.87 |==================================================================== b . 10.85 |==================================================================== SVT-AV1 1.6 Encoder Mode: Preset 8 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 85.31 |==================================================================== b . 85.50 |==================================================================== SVT-AV1 1.6 Encoder Mode: Preset 12 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 308.25 |=================================================================== b . 305.73 |================================================================== SVT-AV1 1.6 Encoder Mode: Preset 13 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 360.92 |================================================================== b . 364.28 |=================================================================== VVenC 1.8 Video Input: Bosphorus 4K - Video Preset: Fast Frames Per Second > Higher Is Better a . 5.440 |==================================================================== b . 5.387 |=================================================================== VVenC 1.8 Video Input: Bosphorus 4K - Video Preset: Faster Frames Per Second > Higher Is Better a . 10.93 |==================================================================== b . 10.89 |==================================================================== VVenC 1.8 Video Input: Bosphorus 1080p - Video Preset: Fast Frames Per Second > Higher Is Better a . 13.88 |==================================================================== b . 13.77 |=================================================================== VVenC 1.8 Video Input: Bosphorus 1080p - Video Preset: Faster Frames Per Second > Higher Is Better a . 24.88 |==================================================================== b . 24.80 |==================================================================== Whisper.cpp 1.4 Model: ggml-base.en - Input: 2016 State of the Union Seconds < Lower Is Better a . 156.48 |=================================================================== b . 151.05 |================================================================= Whisper.cpp 1.4 Model: ggml-small.en - Input: 2016 State of the Union Seconds < Lower Is Better a . 395.71 |=================================================================== b . 363.32 |============================================================== Whisper.cpp 1.4 Model: ggml-medium.en - Input: 2016 State of the Union Seconds < Lower Is Better a . 1018.28 |================================================================== b . 1003.11 |================================================================= Xonotic 0.8.6 Resolution: 1920 x 1080 - Effects Quality: Low Frames Per Second > Higher Is Better a . 671.42 |=================================================================== b . 669.86 |=================================================================== Xonotic 0.8.6 Resolution: 1920 x 1200 - Effects Quality: Low Frames Per Second > Higher Is Better a . 671.95 |=================================================================== b . 676.08 |=================================================================== Xonotic 0.8.6 Resolution: 2560 x 1440 - Effects Quality: Low Frames Per Second > Higher Is Better a . 673.17 |================================================================== b . 687.49 |=================================================================== Xonotic 0.8.6 Resolution: 3840 x 2160 - Effects Quality: Low Frames Per Second > Higher Is Better a . 670.04 |================================================================== b . 675.84 |=================================================================== Xonotic 0.8.6 Resolution: 1920 x 1080 - Effects Quality: High Frames Per Second > Higher Is Better a . 561.44 |=================================================================== b . 563.79 |=================================================================== Xonotic 0.8.6 Resolution: 1920 x 1200 - Effects Quality: High Frames Per Second > Higher Is Better a . 561.01 |================================================================== b . 567.94 |=================================================================== Xonotic 0.8.6 Resolution: 2560 x 1440 - Effects Quality: High Frames Per Second > Higher Is Better a . 560.97 |================================================================== b . 567.98 |=================================================================== Xonotic 0.8.6 Resolution: 3840 x 2160 - Effects Quality: High Frames Per Second > Higher Is Better a . 467.69 |=================================================================== b . 469.75 |=================================================================== Xonotic 0.8.6 Resolution: 1920 x 1080 - Effects Quality: Ultra Frames Per Second > Higher Is Better a . 518.67 |================================================================== b . 524.55 |=================================================================== Xonotic 0.8.6 Resolution: 1920 x 1200 - Effects Quality: Ultra Frames Per Second > Higher Is Better a . 521.50 |=================================================================== b . 521.20 |=================================================================== Xonotic 0.8.6 Resolution: 2560 x 1440 - Effects Quality: Ultra Frames Per Second > Higher Is Better a . 520.79 |================================================================== b . 527.21 |=================================================================== Xonotic 0.8.6 Resolution: 3840 x 2160 - Effects Quality: Ultra Frames Per Second > Higher Is Better a . 420.75 |=================================================================== b . 423.33 |=================================================================== Xonotic 0.8.6 Resolution: 1920 x 1080 - Effects Quality: Ultimate Frames Per Second > Higher Is Better a . 386.94 |=================================================================== b . 386.28 |=================================================================== Xonotic 0.8.6 Resolution: 1920 x 1200 - Effects Quality: Ultimate Frames Per Second > Higher Is Better a . 384.77 |=================================================================== b . 386.81 |=================================================================== Xonotic 0.8.6 Resolution: 2560 x 1440 - Effects Quality: Ultimate Frames Per Second > Higher Is Better a . 384.02 |=================================================================== b . 381.32 |=================================================================== Xonotic 0.8.6 Resolution: 3840 x 2160 - Effects Quality: Ultimate Frames Per Second > Higher Is Better a . 311.40 |=================================================================== b . 311.77 |=================================================================== Z3 Theorem Prover 4.12.1 SMT File: 1.smt2 Seconds < Lower Is Better a . 29.93 |==================================================================== b . 29.90 |==================================================================== Z3 Theorem Prover 4.12.1 SMT File: 2.smt2 Seconds < Lower Is Better a . 76.01 |==================================================================== b . 76.12 |====================================================================