df2888 Intel Xeon E E-2488 testing with a Supermicro Super Server X13SCL-F v0123456789 (1.1 BIOS) and llvmpipe on Ubuntu 22.04 via the Phoronix Test Suite. a: Processor: Intel Xeon E E-2488 @ 3.20GHz (8 Cores / 16 Threads), Motherboard: Supermicro Super Server X13SCL-F v0123456789 (1.1 BIOS), Chipset: Intel Device 7a27, Memory: 2 x 32GB DRAM-4400MT/s Micron MTC20C2085S1EC48BA1, Disk: 960GB Micron_7450_MTFDKBA960TFR + 3201GB Micron_7450_MTFDKCC3T2TFS, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Intel I210 OS: Ubuntu 22.04, Kernel: 6.2.0-26-generic (x86_64), Desktop: GNOME Shell 42.9, Display Server: X Server 1.21.1.4, Vulkan: 1.3.238, Compiler: GCC 11.4.0, File-System: ext4, Screen Resolution: 1920x1080 b: Processor: Intel Xeon E E-2488 @ 3.20GHz (8 Cores / 16 Threads), Motherboard: Supermicro Super Server X13SCL-F v0123456789 (1.1 BIOS), Chipset: Intel Device 7a27, Memory: 2 x 32GB DRAM-4400MT/s Micron MTC20C2085S1EC48BA1, Disk: 960GB Micron_7450_MTFDKBA960TFR + 3201GB Micron_7450_MTFDKCC3T2TFS, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Intel I210 OS: Ubuntu 22.04, Kernel: 6.2.0-26-generic (x86_64), Desktop: GNOME Shell 42.9, Display Server: X Server 1.21.1.4, Vulkan: 1.3.238, Compiler: GCC 11.4.0, File-System: ext4, Screen Resolution: 1920x1080 c: Processor: Intel Xeon E E-2488 @ 3.20GHz (8 Cores / 16 Threads), Motherboard: Supermicro Super Server X13SCL-F v0123456789 (1.1 BIOS), Chipset: Intel Device 7a27, Memory: 2 x 32GB DRAM-4400MT/s Micron MTC20C2085S1EC48BA1, Disk: 960GB Micron_7450_MTFDKBA960TFR + 3201GB Micron_7450_MTFDKCC3T2TFS, Graphics: llvmpipe, Monitor: VE228, Network: 2 x Intel I210 OS: Ubuntu 22.04, Kernel: 6.2.0-26-generic (x86_64), Desktop: GNOME Shell 42.9, Display Server: X Server 1.21.1.4, Vulkan: 1.3.238, Compiler: GCC 11.4.0, File-System: ext4, Screen Resolution: 1920x1080 OpenVINO 2024.0 Model: Road Segmentation ADAS FP16 - Device: CPU ms < Lower Is Better a . 44.80 |================================================================== b . 39.65 |=========================================================== c . 45.87 |==================================================================== OpenVINO 2024.0 Model: Road Segmentation ADAS FP16 - Device: CPU FPS > Higher Is Better a . 89.40 |=========================================================== b . 100.83 |=================================================================== c . 87.16 |========================================================== PostgreSQL 16 Scaling Factor: 1 - Clients: 800 - Mode: Read Only TPS > Higher Is Better a . 965813 |=========================================================== b . 1078377 |================================================================== c . 1057878 |================================================================= PostgreSQL 16 Scaling Factor: 1 - Clients: 800 - Mode: Read Only - Average Latency ms < Lower Is Better a . 0.828 |==================================================================== b . 0.742 |============================================================= c . 0.756 |============================================================== SVT-VP9 0.3 Tuning: Visual Quality Optimized - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 59.67 |============================================================= b . 65.11 |=================================================================== c . 66.15 |==================================================================== SVT-AV1 2.0 Encoder Mode: Preset 12 - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 105.78 |============================================================= b . 117.03 |=================================================================== c . 110.57 |=============================================================== PostgreSQL 16 Scaling Factor: 1000 - Clients: 800 - Mode: Read Write TPS > Higher Is Better a . 27541 |================================================================= b . 25978 |============================================================== c . 28685 |==================================================================== PostgreSQL 16 Scaling Factor: 1000 - Clients: 800 - Mode: Read Write - Average Latency ms < Lower Is Better a . 29.11 |================================================================ b . 30.80 |==================================================================== c . 27.89 |============================================================== uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Super Fast Frames Per Second > Higher Is Better a . 98.26 |============================================================= b . 108.39 |=================================================================== c . 104.61 |================================================================= SVT-AV1 2.0 Encoder Mode: Preset 8 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 136.09 |=================================================================== b . 135.71 |=================================================================== c . 124.19 |============================================================= PostgreSQL 16 Scaling Factor: 1000 - Clients: 1000 - Mode: Read Write TPS > Higher Is Better a . 25189 |==================================================================== b . 24732 |=================================================================== c . 23058 |============================================================== PostgreSQL 16 Scaling Factor: 1000 - Clients: 1000 - Mode: Read Write - Average Latency ms < Lower Is Better a . 39.73 |============================================================== b . 40.43 |=============================================================== c . 43.37 |==================================================================== PostgreSQL 16 Scaling Factor: 1 - Clients: 800 - Mode: Read Write TPS > Higher Is Better a . 2347 |=================================================================== b . 2236 |================================================================ c . 2413 |===================================================================== PostgreSQL 16 Scaling Factor: 1 - Clients: 800 - Mode: Read Write - Average Latency ms < Lower Is Better a . 341.05 |================================================================ b . 357.81 |=================================================================== c . 331.59 |============================================================== uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Very Fast Frames Per Second > Higher Is Better a . 91.58 |=============================================================== b . 98.64 |==================================================================== c . 96.72 |=================================================================== x265 3.4 Video Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 84.85 |=============================================================== b . 91.03 |==================================================================== c . 89.72 |=================================================================== SVT-VP9 0.3 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 83.76 |================================================================ b . 89.47 |==================================================================== c . 88.88 |==================================================================== PostgreSQL 16 Scaling Factor: 100 - Clients: 800 - Mode: Read Only - Average Latency ms < Lower Is Better a . 0.952 |================================================================ b . 0.948 |================================================================ c . 1.009 |==================================================================== Neural Magic DeepSparse 1.7 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 446.59 |=================================================================== b . 419.60 |=============================================================== c . 441.69 |================================================================== Neural Magic DeepSparse 1.7 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 8.9566 |=============================================================== b . 9.5325 |=================================================================== c . 9.0559 |================================================================ PostgreSQL 16 Scaling Factor: 100 - Clients: 800 - Mode: Read Only TPS > Higher Is Better a . 840809 |=================================================================== b . 844015 |=================================================================== c . 793236 |=============================================================== SVT-AV1 2.0 Encoder Mode: Preset 13 - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 106.95 |=============================================================== b . 109.01 |================================================================ c . 113.63 |=================================================================== PostgreSQL 16 Scaling Factor: 1 - Clients: 1000 - Mode: Read Write TPS > Higher Is Better a . 1743 |================================================================= b . 1846 |===================================================================== c . 1815 |==================================================================== PostgreSQL 16 Scaling Factor: 1 - Clients: 1000 - Mode: Read Only - Average Latency ms < Lower Is Better a . 1.164 |==================================================================== b . 1.108 |================================================================ c . 1.171 |==================================================================== PostgreSQL 16 Scaling Factor: 1 - Clients: 1000 - Mode: Read Only TPS > Higher Is Better a . 859514 |================================================================ b . 902210 |=================================================================== c . 853917 |=============================================================== SVT-HEVC 1.5.0 Tuning: 7 - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 48.38 |================================================================= b . 50.98 |==================================================================== c . 49.07 |================================================================= PostgreSQL 16 Scaling Factor: 1000 - Clients: 800 - Mode: Read Only TPS > Higher Is Better a . 619738 |=================================================================== b . 588199 |================================================================ c . 594778 |================================================================ PostgreSQL 16 Scaling Factor: 1000 - Clients: 800 - Mode: Read Only - Average Latency ms < Lower Is Better a . 1.292 |================================================================= b . 1.360 |==================================================================== c . 1.345 |=================================================================== SPECFEM3D 4.0 Model: Layered Halfspace Seconds < Lower Is Better a . 150.98 |=================================================================== b . 151.95 |=================================================================== c . 144.47 |================================================================ uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Ultra Fast Frames Per Second > Higher Is Better a . 120.88 |================================================================ b . 126.91 |=================================================================== c . 122.28 |================================================================= Timed FFmpeg Compilation 6.1 Time To Compile Seconds < Lower Is Better a . 42.19 |==================================================================== b . 40.22 |================================================================= c . 40.38 |================================================================= OpenVINO 2024.0 Model: Person Detection FP32 - Device: CPU FPS > Higher Is Better a . 28.53 |================================================================= b . 29.06 |================================================================== c . 29.87 |==================================================================== OpenVINO 2024.0 Model: Person Detection FP32 - Device: CPU ms < Lower Is Better a . 140.16 |=================================================================== b . 137.62 |================================================================== c . 133.88 |================================================================ Neural Magic DeepSparse 1.7 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 109.83 |================================================================== b . 106.20 |================================================================ c . 111.06 |=================================================================== Neural Magic DeepSparse 1.7 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 9.1063 |================================================================= b . 9.4160 |=================================================================== c . 9.0038 |================================================================ Neural Magic DeepSparse 1.7 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 111.06 |=================================================================== b . 106.58 |================================================================ c . 111.31 |=================================================================== Neural Magic DeepSparse 1.7 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 9.0036 |================================================================ b . 9.3819 |=================================================================== c . 8.9835 |================================================================ 7-Zip Compression 22.01 Test: Decompression Rating MIPS > Higher Is Better a . 65180 |================================================================= b . 67934 |==================================================================== c . 67334 |=================================================================== PostgreSQL 16 Scaling Factor: 1000 - Clients: 1000 - Mode: Read Only TPS > Higher Is Better a . 599514 |=================================================================== b . 581001 |================================================================= c . 576586 |================================================================ PostgreSQL 16 Scaling Factor: 1000 - Clients: 1000 - Mode: Read Only - Average Latency ms < Lower Is Better a . 1.668 |================================================================= b . 1.721 |=================================================================== c . 1.734 |==================================================================== Neural Magic DeepSparse 1.7 Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 802.07 |================================================================= b . 813.73 |================================================================== c . 831.54 |=================================================================== Neural Magic DeepSparse 1.7 Model: ResNet-50, Sparse INT8 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 4.9727 |=================================================================== b . 4.9012 |================================================================== c . 4.7965 |================================================================= Neural Magic DeepSparse 1.7 Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 19.23 |==================================================================== b . 18.55 |================================================================== c . 18.63 |================================================================== Neural Magic DeepSparse 1.7 Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 51.98 |================================================================== b . 53.88 |==================================================================== c . 53.67 |==================================================================== Neural Magic DeepSparse 1.7 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 109.27 |================================================================= b . 112.98 |=================================================================== c . 109.11 |================================================================= Neural Magic DeepSparse 1.7 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 36.60 |==================================================================== b . 35.39 |================================================================== c . 36.65 |==================================================================== Neural Magic DeepSparse 1.7 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 26.49 |==================================================================== b . 25.59 |================================================================== c . 26.36 |==================================================================== Neural Magic DeepSparse 1.7 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 150.90 |================================================================= b . 156.23 |=================================================================== c . 151.64 |================================================================= SVT-HEVC 1.5.0 Tuning: 10 - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 103.05 |================================================================== b . 101.69 |================================================================= c . 105.17 |=================================================================== ClickHouse 22.12.3.5 100M Rows Hits Dataset, First Run / Cold Cache Queries Per Minute, Geo Mean > Higher Is Better a . 226.10 |=================================================================== b . 226.19 |=================================================================== c . 218.80 |================================================================= x264 2022-02-22 Video Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 31.00 |=================================================================== b . 30.62 |================================================================== c . 31.63 |==================================================================== PostgreSQL 16 Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average Latency ms < Lower Is Better a . 1.411 |=================================================================== b . 1.386 |================================================================== c . 1.431 |==================================================================== PostgreSQL 16 Scaling Factor: 100 - Clients: 1000 - Mode: Read Only TPS > Higher Is Better a . 708887 |================================================================== b . 721517 |=================================================================== c . 699050 |================================================================= SVT-AV1 2.0 Encoder Mode: Preset 4 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 16.23 |==================================================================== b . 15.72 |================================================================== c . 15.82 |================================================================== uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Super Fast Frames Per Second > Higher Is Better a . 20.31 |================================================================== b . 20.90 |==================================================================== c . 20.26 |================================================================== PostgreSQL 16 Scaling Factor: 100 - Clients: 800 - Mode: Read Write TPS > Higher Is Better a . 34974 |==================================================================== b . 34383 |=================================================================== c . 33976 |================================================================== PostgreSQL 16 Scaling Factor: 100 - Clients: 800 - Mode: Read Write - Average Latency ms < Lower Is Better a . 22.88 |================================================================== b . 23.27 |=================================================================== c . 23.55 |==================================================================== OpenVINO 2024.0 Model: Face Detection Retail FP16 - Device: CPU ms < Lower Is Better a . 9.88 |=================================================================== b . 10.06 |==================================================================== c . 9.78 |================================================================== OpenVINO 2024.0 Model: Face Detection Retail FP16 - Device: CPU FPS > Higher Is Better a . 808.93 |================================================================== b . 794.56 |================================================================= c . 817.16 |=================================================================== Neural Magic DeepSparse 1.7 Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 109.34 |================================================================= b . 112.38 |=================================================================== c . 110.66 |================================================================== Neural Magic DeepSparse 1.7 Model: ResNet-50, Baseline - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 36.57 |==================================================================== b . 35.58 |================================================================== c . 36.14 |=================================================================== SVT-VP9 0.3 Tuning: VMAF Optimized - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 281.08 |=================================================================== b . 274.27 |================================================================= c . 274.48 |================================================================= Neural Magic DeepSparse 1.7 Model: ResNet-50, Baseline - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 10.27 |==================================================================== b . 10.02 |================================================================== c . 10.05 |=================================================================== Neural Magic DeepSparse 1.7 Model: ResNet-50, Baseline - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 97.34 |================================================================== b . 99.74 |==================================================================== c . 99.49 |==================================================================== ClickHouse 22.12.3.5 100M Rows Hits Dataset, Second Run Queries Per Minute, Geo Mean > Higher Is Better a . 234.71 |================================================================= b . 240.38 |=================================================================== c . 237.98 |================================================================== VVenC 1.11 Video Input: Bosphorus 1080p - Video Preset: Faster Frames Per Second > Higher Is Better a . 40.20 |=================================================================== b . 40.75 |==================================================================== c . 39.80 |================================================================== OpenVINO 2024.0 Model: Person Detection FP16 - Device: CPU ms < Lower Is Better a . 137.85 |=================================================================== b . 135.48 |================================================================= c . 138.64 |=================================================================== NAMD 3.0b6 Input: ATPase with 327,506 Atoms ns/day > Higher Is Better a . 0.67208 |================================================================== b . 0.66733 |================================================================== c . 0.65683 |================================================================= OpenVINO 2024.0 Model: Person Detection FP16 - Device: CPU FPS > Higher Is Better a . 29.01 |=================================================================== b . 29.50 |==================================================================== c . 28.85 |=================================================================== Neural Magic DeepSparse 1.7 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 10.27 |==================================================================== b . 10.06 |=================================================================== c . 10.06 |=================================================================== Neural Magic DeepSparse 1.7 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 97.32 |=================================================================== b . 99.33 |==================================================================== c . 99.39 |==================================================================== Neural Magic DeepSparse 1.7 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 51.32 |==================================================================== b . 50.26 |=================================================================== c . 50.38 |=================================================================== Neural Magic DeepSparse 1.7 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 77.94 |=================================================================== b . 79.58 |==================================================================== c . 79.38 |==================================================================== ClickHouse 22.12.3.5 100M Rows Hits Dataset, Third Run Queries Per Minute, Geo Mean > Higher Is Better a . 243.23 |=================================================================== b . 239.55 |================================================================== c . 238.31 |================================================================== Neural Magic DeepSparse 1.7 Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 334.71 |================================================================== b . 340.82 |=================================================================== c . 334.20 |================================================================== Neural Magic DeepSparse 1.7 Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 11.94 |==================================================================== b . 11.72 |=================================================================== c . 11.96 |==================================================================== SPECFEM3D 4.0 Model: Tomographic Model Seconds < Lower Is Better a . 56.66 |==================================================================== b . 56.61 |==================================================================== c . 55.57 |=================================================================== SVT-AV1 2.0 Encoder Mode: Preset 8 - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 35.47 |=================================================================== b . 36.02 |==================================================================== c . 36.16 |==================================================================== Neural Magic DeepSparse 1.7 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 95.44 |==================================================================== b . 93.85 |=================================================================== c . 93.66 |=================================================================== Neural Magic DeepSparse 1.7 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 14.83 |==================================================================== b . 14.55 |=================================================================== c . 14.56 |=================================================================== Neural Magic DeepSparse 1.7 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 10.48 |=================================================================== b . 10.65 |==================================================================== c . 10.68 |==================================================================== Neural Magic DeepSparse 1.7 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 362.27 |=================================================================== b . 355.53 |================================================================== c . 356.21 |================================================================== Neural Magic DeepSparse 1.7 Model: CV Segmentation, 90% Pruned YOLACT Pruned - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 11.04 |=================================================================== b . 11.25 |==================================================================== c . 11.23 |==================================================================== Neural Magic DeepSparse 1.7 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 67.43 |=================================================================== b . 68.71 |==================================================================== c . 68.67 |==================================================================== Neural Magic DeepSparse 1.7 Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 1.4926 |=================================================================== b . 1.4802 |================================================================== c . 1.4665 |================================================================== Neural Magic DeepSparse 1.7 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 9.8776 |=================================================================== b . 9.7105 |================================================================== c . 9.7892 |================================================================== Neural Magic DeepSparse 1.7 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 101.16 |================================================================== b . 102.89 |=================================================================== c . 102.06 |================================================================== Neural Magic DeepSparse 1.7 Model: ResNet-50, Sparse INT8 - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 667.99 |================================================================== b . 673.15 |================================================================== c . 679.28 |=================================================================== SPECFEM3D 4.0 Model: Mount St. Helens Seconds < Lower Is Better a . 56.75 |==================================================================== b . 57.07 |==================================================================== c . 56.14 |=================================================================== SVT-VP9 0.3 Tuning: VMAF Optimized - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 81.76 |==================================================================== b . 80.44 |=================================================================== c . 80.61 |=================================================================== NAMD 3.0b6 Input: STMV with 1,066,628 Atoms ns/day > Higher Is Better a . 0.20937 |================================================================== b . 0.20927 |================================================================== c . 0.20601 |================================================================= Neural Magic DeepSparse 1.7 Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 3.8825 |=================================================================== b . 3.8228 |================================================================== c . 3.8275 |================================================================== Neural Magic DeepSparse 1.7 Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 257.33 |================================================================== b . 261.27 |=================================================================== c . 260.96 |=================================================================== Timed Linux Kernel Compilation 6.8 Build: defconfig Seconds < Lower Is Better a . 91.85 |==================================================================== b . 90.49 |=================================================================== c . 90.51 |=================================================================== OpenVINO 2024.0 Model: Weld Porosity Detection FP16-INT8 - Device: CPU ms < Lower Is Better a . 7.68 |===================================================================== b . 7.71 |===================================================================== c . 7.60 |==================================================================== OpenVINO 2024.0 Model: Road Segmentation ADAS FP16-INT8 - Device: CPU FPS > Higher Is Better a . 210.19 |================================================================== b . 213.22 |=================================================================== c . 211.43 |================================================================== OpenVINO 2024.0 Model: Road Segmentation ADAS FP16-INT8 - Device: CPU ms < Lower Is Better a . 38.04 |==================================================================== b . 37.50 |=================================================================== c . 37.82 |==================================================================== Neural Magic DeepSparse 1.7 Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 72.63 |==================================================================== b . 71.70 |=================================================================== c . 72.73 |==================================================================== Neural Magic DeepSparse 1.7 Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 55.06 |=================================================================== b . 55.78 |==================================================================== c . 54.99 |=================================================================== OpenVINO 2024.0 Model: Weld Porosity Detection FP16-INT8 - Device: CPU FPS > Higher Is Better a . 1040.68 |================================================================= b . 1036.69 |================================================================= c . 1051.22 |================================================================== uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Slow Frames Per Second > Higher Is Better a . 32.90 |==================================================================== b . 32.49 |=================================================================== c . 32.94 |==================================================================== GROMACS 2024 Implementation: MPI CPU - Input: water_GMX50_bare Ns Per Day > Higher Is Better a . 1.387 |=================================================================== b . 1.406 |==================================================================== c . 1.403 |==================================================================== 7-Zip Compression 22.01 Test: Compression Rating MIPS > Higher Is Better a . 95609 |=================================================================== b . 96616 |==================================================================== c . 96892 |==================================================================== Neural Magic DeepSparse 1.7 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 445.64 |=================================================================== b . 442.90 |=================================================================== c . 439.84 |================================================================== Neural Magic DeepSparse 1.7 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 8.9756 |================================================================== b . 9.0311 |=================================================================== c . 9.0939 |=================================================================== BRL-CAD 7.38.2 VGR Performance Metric VGR Performance Metric > Higher Is Better a . 189384 |=================================================================== b . 187148 |================================================================== c . 188833 |=================================================================== OpenVINO 2024.0 Model: Face Detection FP16-INT8 - Device: CPU ms < Lower Is Better a . 371.82 |================================================================== b . 374.64 |=================================================================== c . 370.27 |================================================================== uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Ultra Fast Frames Per Second > Higher Is Better a . 24.89 |==================================================================== b . 24.72 |==================================================================== c . 24.60 |=================================================================== SVT-HEVC 1.5.0 Tuning: 1 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 9.54 |==================================================================== b . 9.59 |===================================================================== c . 9.65 |===================================================================== OpenVINO 2024.0 Model: Face Detection FP16-INT8 - Device: CPU FPS > Higher Is Better a . 10.74 |==================================================================== b . 10.67 |=================================================================== c . 10.79 |==================================================================== uvg266 0.4.1 Video Input: Bosphorus 1080p - Video Preset: Medium Frames Per Second > Higher Is Better a . 37.15 |==================================================================== b . 36.75 |=================================================================== c . 36.90 |==================================================================== OpenVINO 2024.0 Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU ms < Lower Is Better a . 0.93 |==================================================================== b . 0.93 |==================================================================== c . 0.94 |===================================================================== OpenVINO 2024.0 Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU FPS > Higher Is Better a . 21990.19 |================================================================= b . 21792.13 |================================================================ c . 21757.77 |================================================================ PostgreSQL 16 Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average Latency ms < Lower Is Better a . 33.40 |==================================================================== b . 33.51 |==================================================================== c . 33.16 |=================================================================== PostgreSQL 16 Scaling Factor: 100 - Clients: 1000 - Mode: Read Write TPS > Higher Is Better a . 29945 |==================================================================== b . 29843 |=================================================================== c . 30160 |==================================================================== Timed LLVM Compilation 16.0 Build System: Unix Makefiles Seconds < Lower Is Better a . 646.62 |=================================================================== b . 639.85 |================================================================== c . 643.71 |=================================================================== Timed LLVM Compilation 16.0 Build System: Ninja Seconds < Lower Is Better a . 636.00 |=================================================================== b . 629.47 |================================================================== c . 630.04 |================================================================== OpenFOAM 10 Input: drivaerFastback, Small Mesh Size - Mesh Time Seconds < Lower Is Better a . 30.33 |==================================================================== b . 30.02 |=================================================================== c . 30.04 |=================================================================== Timed Godot Game Engine Compilation 4.0 Time To Compile Seconds < Lower Is Better a . 307.12 |=================================================================== b . 306.66 |=================================================================== c . 304.58 |================================================================== OpenVINO 2024.0 Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU FPS > Higher Is Better a . 8563.76 |================================================================== b . 8538.59 |================================================================== c . 8495.04 |================================================================= VVenC 1.11 Video Input: Bosphorus 4K - Video Preset: Faster Frames Per Second > Higher Is Better a . 11.60 |=================================================================== b . 11.70 |==================================================================== c . 11.65 |==================================================================== SPECFEM3D 4.0 Model: Water-layered Halfspace Seconds < Lower Is Better a . 139.21 |=================================================================== b . 139.33 |=================================================================== c . 138.25 |================================================================== x265 3.4 Video Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 19.48 |=================================================================== b . 19.50 |==================================================================== c . 19.63 |==================================================================== VVenC 1.11 Video Input: Bosphorus 1080p - Video Preset: Fast Frames Per Second > Higher Is Better a . 15.45 |=================================================================== b . 15.57 |==================================================================== c . 15.49 |==================================================================== OpenVINO 2024.0 Model: Face Detection FP16 - Device: CPU FPS > Higher Is Better a . 2.68 |===================================================================== b . 2.66 |==================================================================== c . 2.66 |==================================================================== OpenVINO 2024.0 Model: Vehicle Detection FP16 - Device: CPU ms < Lower Is Better a . 16.60 |==================================================================== b . 16.69 |==================================================================== c . 16.57 |==================================================================== SVT-AV1 2.0 Encoder Mode: Preset 4 - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 4.354 |==================================================================== b . 4.336 |==================================================================== c . 4.367 |==================================================================== OpenVINO 2024.0 Model: Vehicle Detection FP16 - Device: CPU FPS > Higher Is Better a . 240.78 |=================================================================== b . 239.55 |=================================================================== c . 241.22 |=================================================================== Neural Magic DeepSparse 1.7 Model: Llama2 Chat 7b Quantized - Scenario: Asynchronous Multi-Stream items/sec > Higher Is Better a . 5.0711 |=================================================================== b . 5.1063 |=================================================================== c . 5.0995 |=================================================================== Neural Magic DeepSparse 1.7 Model: Llama2 Chat 7b Quantized - Scenario: Asynchronous Multi-Stream ms/batch < Lower Is Better a . 781.12 |=================================================================== b . 775.76 |=================================================================== c . 776.83 |=================================================================== OpenVINO 2024.0 Model: Noise Suppression Poconet-Like FP16 - Device: CPU ms < Lower Is Better a . 15.19 |==================================================================== b . 15.09 |==================================================================== c . 15.11 |==================================================================== OpenVINO 2024.0 Model: Machine Translation EN To DE FP16 - Device: CPU FPS > Higher Is Better a . 33.68 |==================================================================== b . 33.70 |==================================================================== c . 33.48 |==================================================================== Blender 4.0 Blend File: BMW27 - Compute: CPU-Only Seconds < Lower Is Better a . 123.52 |=================================================================== b . 122.72 |=================================================================== c . 122.81 |=================================================================== OpenVINO 2024.0 Model: Machine Translation EN To DE FP16 - Device: CPU ms < Lower Is Better a . 118.74 |=================================================================== b . 118.68 |=================================================================== c . 119.45 |=================================================================== OpenVINO 2024.0 Model: Noise Suppression Poconet-Like FP16 - Device: CPU FPS > Higher Is Better a . 524.22 |=================================================================== b . 527.54 |=================================================================== c . 527.20 |=================================================================== SVT-VP9 0.3 Tuning: Visual Quality Optimized - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 224.63 |=================================================================== b . 225.96 |=================================================================== c . 224.72 |=================================================================== Blender 4.0 Blend File: Fishy Cat - Compute: CPU-Only Seconds < Lower Is Better a . 165.95 |=================================================================== b . 165.79 |=================================================================== c . 165.05 |=================================================================== OpenVINO 2024.0 Model: Vehicle Detection FP16-INT8 - Device: CPU FPS > Higher Is Better a . 497.27 |=================================================================== b . 499.46 |=================================================================== c . 499.92 |=================================================================== OpenVINO 2024.0 Model: Weld Porosity Detection FP16 - Device: CPU ms < Lower Is Better a . 29.22 |==================================================================== b . 29.35 |==================================================================== c . 29.37 |==================================================================== OpenVINO 2024.0 Model: Weld Porosity Detection FP16 - Device: CPU FPS > Higher Is Better a . 273.62 |=================================================================== b . 272.46 |=================================================================== c . 272.25 |=================================================================== OpenVINO 2024.0 Model: Vehicle Detection FP16-INT8 - Device: CPU ms < Lower Is Better a . 16.08 |==================================================================== b . 16.01 |==================================================================== c . 16.00 |==================================================================== OpenVINO 2024.0 Model: Face Detection FP16 - Device: CPU ms < Lower Is Better a . 1495.10 |================================================================== b . 1502.55 |================================================================== c . 1499.57 |================================================================== OpenVINO 2024.0 Model: Person Vehicle Bike Detection FP16 - Device: CPU ms < Lower Is Better a . 10.38 |==================================================================== b . 10.42 |==================================================================== c . 10.43 |==================================================================== OpenVINO 2024.0 Model: Handwritten English Recognition FP16-INT8 - Device: CPU ms < Lower Is Better a . 43.92 |==================================================================== b . 43.88 |==================================================================== c . 44.09 |==================================================================== OpenVINO 2024.0 Model: Handwritten English Recognition FP16-INT8 - Device: CPU FPS > Higher Is Better a . 182.10 |=================================================================== b . 182.21 |=================================================================== c . 181.39 |=================================================================== uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Medium Frames Per Second > Higher Is Better a . 6.68 |===================================================================== b . 6.68 |===================================================================== c . 6.71 |===================================================================== OpenVINO 2024.0 Model: Person Vehicle Bike Detection FP16 - Device: CPU FPS > Higher Is Better a . 384.90 |=================================================================== b . 383.45 |=================================================================== c . 383.23 |=================================================================== SVT-HEVC 1.5.0 Tuning: 1 - Input: Bosphorus 4K Frames Per Second > Higher Is Better a . 2.38 |===================================================================== b . 2.37 |===================================================================== c . 2.37 |===================================================================== uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Very Fast Frames Per Second > Higher Is Better a . 19.31 |==================================================================== b . 19.23 |==================================================================== c . 19.27 |==================================================================== OpenVINO 2024.0 Model: Handwritten English Recognition FP16 - Device: CPU ms < Lower Is Better a . 52.04 |==================================================================== b . 51.83 |==================================================================== c . 51.99 |==================================================================== OpenVINO 2024.0 Model: Handwritten English Recognition FP16 - Device: CPU FPS > Higher Is Better a . 153.66 |=================================================================== b . 154.28 |=================================================================== c . 153.81 |=================================================================== OpenVINO 2024.0 Model: Person Re-Identification Retail FP16 - Device: CPU ms < Lower Is Better a . 18.18 |==================================================================== b . 18.17 |==================================================================== c . 18.24 |==================================================================== OpenVINO 2024.0 Model: Person Re-Identification Retail FP16 - Device: CPU FPS > Higher Is Better a . 439.76 |=================================================================== b . 439.93 |=================================================================== c . 438.29 |=================================================================== Blender 4.0 Blend File: Classroom - Compute: CPU-Only Seconds < Lower Is Better a . 336.83 |=================================================================== b . 335.62 |=================================================================== c . 336.08 |=================================================================== Timed Node.js Compilation 19.8.1 Time To Compile Seconds < Lower Is Better a . 523.99 |=================================================================== b . 522.11 |=================================================================== c . 523.42 |=================================================================== OpenFOAM 10 Input: drivaerFastback, Medium Mesh Size - Mesh Time Seconds < Lower Is Better a . 221.91 |=================================================================== b . 222.69 |=================================================================== c . 222.19 |=================================================================== SVT-VP9 0.3 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 291.27 |=================================================================== b . 291.65 |=================================================================== c . 292.25 |=================================================================== OpenVINO 2024.0 Model: Face Detection Retail FP16-INT8 - Device: CPU FPS > Higher Is Better a . 1634.31 |================================================================== b . 1639.46 |================================================================== c . 1636.64 |================================================================== Blender 4.0 Blend File: Barbershop - Compute: CPU-Only Seconds < Lower Is Better a . 1297.10 |================================================================== b . 1298.17 |================================================================== c . 1294.72 |================================================================== Timed Linux Kernel Compilation 6.8 Build: allmodconfig Seconds < Lower Is Better a . 1311.87 |================================================================== b . 1308.54 |================================================================== c . 1308.93 |================================================================== SVT-HEVC 1.5.0 Tuning: 10 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 381.27 |=================================================================== b . 381.92 |=================================================================== c . 382.17 |=================================================================== Timed GCC Compilation 13.2 Time To Compile Seconds < Lower Is Better a . 791.27 |=================================================================== b . 790.87 |=================================================================== c . 789.59 |=================================================================== OpenVINO 2024.0 Model: Face Detection Retail FP16-INT8 - Device: CPU ms < Lower Is Better a . 4.89 |===================================================================== b . 4.88 |===================================================================== c . 4.88 |===================================================================== uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Slow Frames Per Second > Higher Is Better a . 5.97 |===================================================================== b . 5.98 |===================================================================== c . 5.98 |===================================================================== OpenFOAM 10 Input: drivaerFastback, Small Mesh Size - Execution Time Seconds < Lower Is Better a . 226.48 |=================================================================== b . 226.11 |=================================================================== c . 226.42 |=================================================================== VVenC 1.11 Video Input: Bosphorus 4K - Video Preset: Fast Frames Per Second > Higher Is Better a . 4.954 |==================================================================== b . 4.946 |==================================================================== c . 4.952 |==================================================================== Neural Magic DeepSparse 1.7 Model: Llama2 Chat 7b Quantized - Scenario: Synchronous Single-Stream ms/batch < Lower Is Better a . 124.62 |=================================================================== b . 124.47 |=================================================================== c . 124.55 |=================================================================== Neural Magic DeepSparse 1.7 Model: Llama2 Chat 7b Quantized - Scenario: Synchronous Single-Stream items/sec > Higher Is Better a . 8.0234 |=================================================================== b . 8.0329 |=================================================================== c . 8.0276 |=================================================================== SVT-AV1 2.0 Encoder Mode: Preset 12 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 532.48 |=================================================================== b . 531.91 |=================================================================== c . 532.51 |=================================================================== Blender 4.0 Blend File: Pabellon Barcelona - Compute: CPU-Only Seconds < Lower Is Better a . 414.01 |=================================================================== b . 413.66 |=================================================================== c . 413.87 |=================================================================== SVT-AV1 2.0 Encoder Mode: Preset 13 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 644.28 |=================================================================== b . 644.76 |=================================================================== c . 644.70 |=================================================================== OpenFOAM 10 Input: drivaerFastback, Medium Mesh Size - Execution Time Seconds < Lower Is Better a . 2182.84 |================================================================== b . 2184.34 |================================================================== c . 2183.10 |================================================================== OpenVINO 2024.0 Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU ms < Lower Is Better a . 0.36 |===================================================================== b . 0.36 |===================================================================== c . 0.36 |===================================================================== PostgreSQL 16 Scaling Factor: 1 - Clients: 1000 - Mode: Read Write - Average Latency ms < Lower Is Better a . 575.61 |=================================================================== b . 541.63 |=============================================================== c . 551.01 |================================================================ x264 2022-02-22 Video Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 155.70 |================================================================== b . 157.78 |=================================================================== c . 152.15 |================================================================= SVT-HEVC 1.5.0 Tuning: 7 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better a . 171.89 |============================================================== b . 166.20 |============================================================ c . 185.24 |=================================================================== SPECFEM3D 4.0 Model: Homogeneous Halfspace Seconds < Lower Is Better a . 72.92 |==================================================================== b . 72.26 |=================================================================== c . 71.57 |===================================================================