AMD 5th Gen EPYC 9005 Turin vs. NVIDIA GH200 Performance Benchmarks AMD 5th Gen EPYC "Turin" Linux benchmarks by Michael Larabel for a future article on launch day of EPYC 9005 series Linux performance. EPYC 9965: Processor: AMD EPYC 9965 192-Core @ 2.25GHz (192 Cores / 384 Threads), Motherboard: AMD VOLCANO (RVOT1000D BIOS), Chipset: AMD Device 153a, Memory: 12 x 64GB DDR5-6000MT/s Samsung M321R8GA0PB1-CCPKC, Disk: 2 x 1920GB KIOXIA KCD8XPUG1T92, Graphics: ASPEED, Network: Broadcom NetXtreme BCM5720 PCIe OS: Ubuntu 24.04, Kernel: 6.10.0-phx (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200 EPYC 9575F: Processor: AMD EPYC 9575F 64-Core @ 3.30GHz (64 Cores / 128 Threads), Motherboard: AMD VOLCANO (RVOT1000D BIOS), Chipset: AMD Device 153a, Memory: 12 x 64GB DDR5-6000MT/s Samsung M321R8GA0PB1-CCPKC, Disk: 2 x 1920GB KIOXIA KCD8XPUG1T92, Graphics: ASPEED, Network: Broadcom NetXtreme BCM5720 PCIe OS: Ubuntu 24.04, Kernel: 6.10.0-phx (x86_64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1200 NVIDIA GH200 144G HBM3e: Processor: ARMv8 Neoverse-V2 @ 3.47GHz (72 Cores), Motherboard: Pegatron JIMBO P4352 (00022432 BIOS), Memory: 1 x 480GB LPDDR5-6400MT/s NVIDIA 699-2G530-0236-RC1, Disk: 1000GB CT1000T700SSD3, Graphics: NVIDIA GH200 144G HBM3e 143GB, Network: 2 x Intel X550 OS: Ubuntu 24.04, Kernel: 6.8.0-47-generic-64k (aarch64), Display Driver: NVIDIA, OpenCL: OpenCL 3.0 CUDA 12.6.65, Compiler: GCC 13.2.0 + CUDA 11.8, File-System: ext4, Screen Resolution: 1920x1200 CPU Power Consumption Monitor Phoronix Test Suite System Monitoring Watts EPYC 9965 ............... MIN: 19 AVG: 275 MAX: 462 EPYC 9575F .............. MIN: 19 AVG: 313 MAX: 403 NVIDIA GH200 144G HBM3e . MIN: 38 AVG: 170 MAX: 300 miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM2 Billion Interactions/s > Higher Is Better EPYC 9965 ............... 428.05 |============================================= EPYC 9575F .............. 244.07 |========================== NVIDIA GH200 144G HBM3e . 54.66 |====== miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM2 Billion Interactions/s Per Watt > Higher Is Better EPYC 9965 ............... 1.448 |============================================== EPYC 9575F .............. 0.763 |======================== NVIDIA GH200 144G HBM3e . 0.255 |======== OpenSSL 3.3 Algorithm: SHA256 byte/s > Higher Is Better EPYC 9965 ............... 223613640290 |======================================= EPYC 9575F .............. 113637325283 |==================== NVIDIA GH200 144G HBM3e . 86724440007 |=============== OpenSSL 3.3 Algorithm: SHA512 byte/s > Higher Is Better EPYC 9965 ............... 87509206173 |======================================== EPYC 9575F .............. 45568081887 |===================== NVIDIA GH200 144G HBM3e . 49490676478 |======================= OpenSSL 3.3 Algorithm: AES-128-GCM byte/s > Higher Is Better EPYC 9965 ............... 2512760030913 |====================================== EPYC 9575F .............. 1289299058233 |=================== NVIDIA GH200 144G HBM3e . 609122521270 |========= OpenSSL 3.3 Algorithm: SHA256 byte/s Per Watt > Higher Is Better EPYC 9965 ............... 591912038.17 |======================================= EPYC 9575F .............. 295380807.41 |=================== NVIDIA GH200 144G HBM3e . 440489220.25 |============================= OpenSSL 3.3 Algorithm: AES-256-GCM byte/s > Higher Is Better EPYC 9965 ............... 2283280593023 |====================================== EPYC 9575F .............. 1193699244990 |==================== NVIDIA GH200 144G HBM3e . 515680506310 |========= OpenSSL 3.3 Algorithm: AES-128-GCM byte/s Per Watt > Higher Is Better EPYC 9965 ............... 6477196397.29 |====================================== EPYC 9575F .............. 3601051103.71 |===================== NVIDIA GH200 144G HBM3e . 2139935369.45 |============= OpenSSL 3.3 Algorithm: ChaCha20 byte/s > Higher Is Better EPYC 9965 ............... 1551003421237 |====================================== EPYC 9575F .............. 733060286053 |================== NVIDIA GH200 144G HBM3e . 149255169543 |==== OpenSSL 3.3 Algorithm: AES-256-GCM byte/s Per Watt > Higher Is Better EPYC 9965 ............... 5989469249.57 |====================================== EPYC 9575F .............. 3325699800.96 |===================== NVIDIA GH200 144G HBM3e . 1809281985.06 |=========== OpenSSL 3.3 Algorithm: ChaCha20-Poly1305 byte/s > Higher Is Better EPYC 9965 ............... 1053141452663 |====================================== EPYC 9575F .............. 499260152310 |================== NVIDIA GH200 144G HBM3e . 108427679423 |==== OpenSSL 3.3 Algorithm: SHA512 byte/s Per Watt > Higher Is Better EPYC 9965 ............... 235679194.64 |===================================== EPYC 9575F .............. 118475697.80 |=================== NVIDIA GH200 144G HBM3e . 248755218.22 |======================================= OpenSSL 3.3 Algorithm: ChaCha20 byte/s Per Watt > Higher Is Better EPYC 9965 ............... 3660529028.05 |====================================== EPYC 9575F .............. 1908650939.88 |==================== NVIDIA GH200 144G HBM3e . 768573496.68 |======== OpenSSL 3.3 Algorithm: ChaCha20-Poly1305 byte/s Per Watt > Higher Is Better EPYC 9965 ............... 2506574135.73 |====================================== EPYC 9575F .............. 1303373595.94 |==================== NVIDIA GH200 144G HBM3e . 535429561.19 |======== miniFE 2.2 Problem Size: Small CG Mflops > Higher Is Better EPYC 9965 ............... 65294.5 |========================================== EPYC 9575F .............. 69114.4 |============================================ NVIDIA GH200 144G HBM3e . 46786.3 |============================== miniFE 2.2 Problem Size: Small CG Mflops Per Watt > Higher Is Better EPYC 9965 ............... 414.89 |========================================= EPYC 9575F .............. 372.32 |===================================== NVIDIA GH200 144G HBM3e . 453.14 |============================================= Algebraic Multi-Grid Benchmark 1.2 Figure Of Merit > Higher Is Better EPYC 9965 ............... 2976345000 |====================================== EPYC 9575F .............. 3197611250 |========================================= NVIDIA GH200 144G HBM3e . 2256454667 |============================= Algebraic Multi-Grid Benchmark 1.2 Figure Of Merit Per Watt > Higher Is Better EPYC 9965 ............... 10018207.52 |========================== EPYC 9575F .............. 10869334.83 |============================ NVIDIA GH200 144G HBM3e . 15435851.48 |======================================== OpenVINO 2024.0 Model: Face Detection FP16-INT8 - Device: CPU FPS > Higher Is Better EPYC 9965 ............... 231.14 |============================================= EPYC 9575F .............. 141.43 |============================ NVIDIA GH200 144G HBM3e . 6.56 |= OpenVINO 2024.0 Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU FPS > Higher Is Better EPYC 9965 ............... 208220.83 |========================================== EPYC 9575F .............. 173795.43 |=================================== NVIDIA GH200 144G HBM3e . 4896.13 |= FFmpeg 7.0 Encoder: libx265 - Scenario: Platform FPS Per Watt > Higher Is Better EPYC 9965 .. 0.508 |=========================================================== EPYC 9575F . 0.492 |========================================================= FFmpeg 7.0 Encoder: libx265 - Scenario: Platform FPS > Higher Is Better EPYC 9965 .. 60.11 |============================================= EPYC 9575F . 79.53 |=========================================================== OpenVINO 2024.0 Model: Person Detection FP16 - Device: CPU FPS > Higher Is Better EPYC 9965 ............... 640.09 |============================================ EPYC 9575F .............. 650.80 |============================================= NVIDIA GH200 144G HBM3e . 32.71 |== OpenVINO 2024.0 Model: Weld Porosity Detection FP16-INT8 - Device: CPU FPS > Higher Is Better EPYC 9965 ............... 22159.64 |=========================================== EPYC 9575F .............. 13533.20 |========================== NVIDIA GH200 144G HBM3e . 465.82 |= OpenVINO 2024.0 Model: Vehicle Detection FP16-INT8 - Device: CPU FPS > Higher Is Better EPYC 9965 ............... 12591.54 |=========================================== EPYC 9575F .............. 7427.27 |========================= NVIDIA GH200 144G HBM3e . 153.47 |= OpenVINO 2024.0 Model: Person Vehicle Bike Detection FP16 - Device: CPU FPS > Higher Is Better EPYC 9965 ............... 11033.61 |=========================================== EPYC 9575F .............. 5717.19 |====================== NVIDIA GH200 144G HBM3e . 547.01 |== OpenVINO 2024.0 Model: Machine Translation EN To DE FP16 - Device: CPU FPS > Higher Is Better EPYC 9965 ............... 1227.23 |============================================ EPYC 9575F .............. 761.69 |=========================== NVIDIA GH200 144G HBM3e . 110.98 |==== OpenVINO 2024.0 Model: Face Detection Retail FP16-INT8 - Device: CPU FPS > Higher Is Better EPYC 9965 ............... 37263.74 |=========================================== EPYC 9575F .............. 21512.07 |========================= NVIDIA GH200 144G HBM3e . 519.02 |= OpenVINO 2024.0 Model: Road Segmentation ADAS FP16-INT8 - Device: CPU FPS > Higher Is Better EPYC 9965 ............... 3497.17 |============================================ EPYC 9575F .............. 2198.48 |============================ NVIDIA GH200 144G HBM3e . 51.07 |= OpenVINO 2024.0 Model: Noise Suppression Poconet-Like FP16 - Device: CPU FPS > Higher Is Better EPYC 9965 ............... 10251.07 |=========================================== EPYC 9575F .............. 7401.31 |=============================== NVIDIA GH200 144G HBM3e . 199.48 |= Embree 4.3 Binary: Pathtracer ISPC - Model: Asian Dragon Frames Per Second Per Watt > Higher Is Better EPYC 9965 .. 1.494 |=========================================================== EPYC 9575F . 0.573 |======================= Embree 4.3 Binary: Pathtracer ISPC - Model: Crown Frames Per Second Per Watt > Higher Is Better EPYC 9965 .. 1.026 |=========================================================== EPYC 9575F . 0.429 |========================= Embree 4.3 Binary: Pathtracer ISPC - Model: Asian Dragon Frames Per Second > Higher Is Better EPYC 9965 .. 267.70 |========================================================== EPYC 9575F . 137.41 |============================== Embree 4.3 Binary: Pathtracer ISPC - Model: Asian Dragon Obj Frames Per Second > Higher Is Better EPYC 9965 .. 230.97 |========================================================== EPYC 9575F . 117.90 |============================== Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Slow Frames Per Second Per Watt > Higher Is Better EPYC 9965 .. 0.282 |=========================================================== EPYC 9575F . 0.141 |============================== uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Very Fast Frames Per Second Per Watt > Higher Is Better EPYC 9965 .. 0.341 |=========================================================== EPYC 9575F . 0.319 |======================================================= Embree 4.3 Binary: Pathtracer ISPC - Model: Crown Frames Per Second > Higher Is Better EPYC 9965 .. 210.40 |========================================================== EPYC 9575F . 111.29 |=============================== SVT-AV1 2.0 Encoder Mode: Preset 13 - Input: Bosphorus 4K Frames Per Second > Higher Is Better EPYC 9965 .. 244.58 |================================================== EPYC 9575F . 283.70 |========================================================== SVT-AV1 2.0 Encoder Mode: Preset 12 - Input: Bosphorus 4K Frames Per Second > Higher Is Better EPYC 9965 .. 248.70 |================================================= EPYC 9575F . 294.41 |========================================================== Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Medium Frames Per Second Per Watt > Higher Is Better EPYC 9965 .. 0.282 |=========================================================== EPYC 9575F . 0.146 |=============================== uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Slow Frames Per Second Per Watt > Higher Is Better EPYC 9965 .. 0.183 |=========================================================== EPYC 9575F . 0.090 |============================= uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Medium Frames Per Second Per Watt > Higher Is Better EPYC 9965 .. 0.203 |=========================================================== EPYC 9575F . 0.104 |============================== SVT-AV1 2.0 Encoder Mode: Preset 8 - Input: Bosphorus 4K Frames Per Second > Higher Is Better EPYC 9965 .. 104.64 |================================================== EPYC 9575F . 121.05 |========================================================== SVT-AV1 2.0 Encoder Mode: Preset 4 - Input: Bosphorus 4K Frames Per Second > Higher Is Better EPYC 9965 .. 10.02 |=============================================== EPYC 9575F . 12.46 |=========================================================== Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Slow Frames Per Second > Higher Is Better EPYC 9965 .. 68.61 |=========================================================== EPYC 9575F . 40.32 |=================================== Embree 4.3 Binary: Pathtracer ISPC - Model: Asian Dragon Obj Frames Per Second Per Watt > Higher Is Better EPYC 9965 .. 1.635 |=========================================================== EPYC 9575F . 0.580 |===================== SVT-AV1 2.0 Encoder Mode: Preset 13 - Input: Bosphorus 4K Frames Per Second Per Watt > Higher Is Better EPYC 9965 .. 2.445 |========================================================== EPYC 9575F . 2.478 |=========================================================== SVT-AV1 2.0 Encoder Mode: Preset 12 - Input: Bosphorus 4K Frames Per Second Per Watt > Higher Is Better EPYC 9965 .. 2.524 |========================================================== EPYC 9575F . 2.583 |=========================================================== Kvazaar 2.2 Video Input: Bosphorus 4K - Video Preset: Medium Frames Per Second > Higher Is Better EPYC 9965 .. 68.82 |=========================================================== EPYC 9575F . 41.30 |=================================== uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Slow Frames Per Second > Higher Is Better EPYC 9965 .. 50.96 |=========================================================== EPYC 9575F . 28.82 |================================= uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Medium Frames Per Second > Higher Is Better EPYC 9965 .. 55.69 |=========================================================== EPYC 9575F . 32.60 |=================================== SVT-AV1 2.0 Encoder Mode: Preset 8 - Input: Bosphorus 4K Frames Per Second Per Watt > Higher Is Better EPYC 9965 .. 0.825 |=========================================================== EPYC 9575F . 0.790 |======================================================== SVT-AV1 2.0 Encoder Mode: Preset 4 - Input: Bosphorus 4K Frames Per Second Per Watt > Higher Is Better EPYC 9965 .. 0.074 |=========================================================== EPYC 9575F . 0.066 |===================================================== uvg266 0.4.1 Video Input: Bosphorus 4K - Video Preset: Very Fast Frames Per Second > Higher Is Better EPYC 9965 .. 70.48 |================================================== EPYC 9575F . 82.89 |=========================================================== miniBUDE 20210901 Implementation: OpenMP - Input Deck: BM2 GFInst/s > Higher Is Better EPYC 9965 ............... 10701.18 |=========================================== EPYC 9575F .............. 6101.79 |========================= NVIDIA GH200 144G HBM3e . 1366.40 |===== High Performance Conjugate Gradient 3.1 X Y Z: 144 144 144 - RT: 60 GFLOP/s > Higher Is Better EPYC 9965 ............... 60.70 |============================================ EPYC 9575F .............. 63.90 |============================================== NVIDIA GH200 144G HBM3e . 42.32 |============================== High Performance Conjugate Gradient 3.1 X Y Z: 144 144 144 - RT: 60 GFLOP/s Per Watt > Higher Is Better EPYC 9965 ............... 0.176 |================================== EPYC 9575F .............. 0.163 |=============================== NVIDIA GH200 144G HBM3e . 0.239 |============================================== Xmrig 6.21 Variant: GhostRider - Hash Count: 1M H/s > Higher Is Better EPYC 9965 ............... 15447.6 |============================================ EPYC 9575F .............. 11601.9 |================================= NVIDIA GH200 144G HBM3e . 4015.5 |=========== Xmrig 6.21 Variant: GhostRider - Hash Count: 1M H/s Per Watt > Higher Is Better EPYC 9965 ............... 48.04 |============================================== EPYC 9575F .............. 30.52 |============================= NVIDIA GH200 144G HBM3e . 28.69 |=========================== Intel Open Image Denoise 2.2 Run: RTLightmap.hdr.4096x4096 - Device: CPU-Only Images / Sec > Higher Is Better EPYC 9965 .. 3.22 |============================================================ EPYC 9575F . 2.40 |============================================= Intel Open Image Denoise 2.2 Run: RT.hdr_alb_nrm.3840x2160 - Device: CPU-Only Images / Sec Per Watt > Higher Is Better EPYC 9965 .. 0.034 |=========================================================== EPYC 9575F . 0.021 |==================================== Intel Open Image Denoise 2.2 Run: RT.ldr_alb_nrm.3840x2160 - Device: CPU-Only Images / Sec Per Watt > Higher Is Better EPYC 9965 .. 0.034 |=========================================================== EPYC 9575F . 0.021 |==================================== Intel Open Image Denoise 2.2 Run: RTLightmap.hdr.4096x4096 - Device: CPU-Only Images / Sec Per Watt > Higher Is Better EPYC 9965 .. 0.013 |=========================================================== EPYC 9575F . 0.008 |==================================== Intel Open Image Denoise 2.2 Run: RT.hdr_alb_nrm.3840x2160 - Device: CPU-Only Images / Sec > Higher Is Better EPYC 9965 .. 6.73 |============================================================ EPYC 9575F . 5.04 |============================================= Intel Open Image Denoise 2.2 Run: RT.ldr_alb_nrm.3840x2160 - Device: CPU-Only Images / Sec > Higher Is Better EPYC 9965 .. 6.74 |============================================================ EPYC 9575F . 5.06 |============================================= TensorFlow 2.16.1 Device: CPU - Batch Size: 256 - Model: ResNet-50 images/sec Per Watt > Higher Is Better EPYC 9965 .. 0.702 |====================================================== EPYC 9575F . 0.773 |=========================================================== TensorFlow 2.16.1 Device: CPU - Batch Size: 256 - Model: AlexNet images/sec Per Watt > Higher Is Better EPYC 9965 .. 10.778 |========================================================== EPYC 9575F . 8.736 |=============================================== TensorFlow 2.16.1 Device: CPU - Batch Size: 256 - Model: ResNet-50 images/sec > Higher Is Better EPYC 9965 ............... 201.12 |======================================= EPYC 9575F .............. 234.68 |============================================= TensorFlow 2.16.1 Device: CPU - Batch Size: 512 - Model: ResNet-50 images/sec Per Watt > Higher Is Better EPYC 9965 .. 0.819 |=========================================================== EPYC 9575F . 0.772 |======================================================== TensorFlow 2.16.1 Device: CPU - Batch Size: 256 - Model: AlexNet images/sec > Higher Is Better EPYC 9965 ............... 2085.02 |============================================ EPYC 9575F .............. 2050.63 |=========================================== TensorFlow 2.16.1 Device: CPU - Batch Size: 512 - Model: ResNet-50 images/sec > Higher Is Better EPYC 9965 ............... 247.22 |============================================ EPYC 9575F .............. 254.02 |============================================= OpenVKL 2.0.0 Benchmark: vklBenchmarkCPU ISPC Items / Sec > Higher Is Better EPYC 9965 .. 4241 |============================================================ EPYC 9575F . 2388 |================================== OpenVKL 2.0.0 Benchmark: vklBenchmarkCPU ISPC Items / Sec Per Watt > Higher Is Better EPYC 9965 .. 12.597 |========================================================== EPYC 9575F . 6.722 |=============================== OSPRay 3.1 Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time Items Per Second > Higher Is Better EPYC 9965 .. 64.80 |=========================================================== EPYC 9575F . 35.65 |================================ OSPRay 3.1 Benchmark: particle_volume/ao/real_time Items Per Second > Higher Is Better EPYC 9965 .. 61.69 |=========================================================== EPYC 9575F . 36.40 |=================================== OSPRay 3.1 Benchmark: particle_volume/scivis/real_time Items Per Second > Higher Is Better EPYC 9965 .. 61.77 |=========================================================== EPYC 9575F . 36.45 |=================================== OSPRay 3.1 Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time Items Per Second Per Watt > Higher Is Better EPYC 9965 .. 0.183 |=========================================================== EPYC 9575F . 0.097 |=============================== OSPRay 3.1 Benchmark: particle_volume/ao/real_time Items Per Second Per Watt > Higher Is Better EPYC 9965 .. 0.170 |=========================================================== EPYC 9575F . 0.095 |================================= OSPRay 3.1 Benchmark: particle_volume/scivis/real_time Items Per Second Per Watt > Higher Is Better EPYC 9965 .. 0.190 |=========================================================== EPYC 9575F . 0.105 |================================= GraphicsMagick 1.3.43 Operation: Noise-Gaussian Iterations Per Minute > Higher Is Better EPYC 9965 ............... 296 |==================================== EPYC 9575F .............. 291 |=================================== NVIDIA GH200 144G HBM3e . 397 |================================================ GraphicsMagick 1.3.43 Operation: Enhanced Iterations Per Minute > Higher Is Better EPYC 9965 ............... 495 |================================================ EPYC 9575F .............. 336 |================================= NVIDIA GH200 144G HBM3e . 369 |==================================== GraphicsMagick 1.3.43 Operation: Noise-Gaussian Iterations Per Minute Per Watt > Higher Is Better EPYC 9965 ............... 1.779 |================================== EPYC 9575F .............. 1.295 |========================= NVIDIA GH200 144G HBM3e . 2.385 |============================================== GraphicsMagick 1.3.43 Operation: Enhanced Iterations Per Minute Per Watt > Higher Is Better EPYC 9965 ............... 1.866 |============================================== EPYC 9575F .............. 0.986 |======================== NVIDIA GH200 144G HBM3e . 1.542 |====================================== GraphicsMagick 1.3.43 Operation: Sharpen Iterations Per Minute Per Watt > Higher Is Better EPYC 9965 ............... 1.536 |============================================ EPYC 9575F .............. 0.904 |========================== NVIDIA GH200 144G HBM3e . 1.600 |============================================== GraphicsMagick 1.3.43 Operation: Sharpen Iterations Per Minute > Higher Is Better EPYC 9965 ............... 445 |================================================ EPYC 9575F .............. 301 |================================ NVIDIA GH200 144G HBM3e . 428 |============================================== Coremark 1.0 CoreMark Size 666 - Iterations Per Second Iterations/Sec > Higher Is Better EPYC 9965 ............... 4247085.81 |========================================= EPYC 9575F .............. 3884226.13 |===================================== NVIDIA GH200 144G HBM3e . 2396679.00 |======================= Coremark 1.0 CoreMark Size 666 - Iterations Per Second Iterations/Sec Per Watt > Higher Is Better EPYC 9965 ............... 16273.75 |=================================== EPYC 9575F .............. 12356.08 |========================== NVIDIA GH200 144G HBM3e . 20070.84 |=========================================== LuxCoreRender 2.6 Scene: DLSC - Acceleration: CPU M samples/sec > Higher Is Better EPYC 9965 .. 24.02 |=========================================================== EPYC 9575F . 15.36 |====================================== LuxCoreRender 2.6 Scene: DLSC - Acceleration: CPU M samples/sec Per Watt > Higher Is Better EPYC 9965 .. 0.066 |=========================================================== EPYC 9575F . 0.041 |===================================== LuxCoreRender 2.6 Scene: Orange Juice - Acceleration: CPU M samples/sec > Higher Is Better EPYC 9965 .. 35.22 |=========================================================== EPYC 9575F . 22.26 |===================================== LuxCoreRender 2.6 Scene: Orange Juice - Acceleration: CPU M samples/sec Per Watt > Higher Is Better EPYC 9965 .. 0.098 |=========================================================== EPYC 9575F . 0.061 |===================================== SecureMark 1.0.4 Benchmark: SecureMark-TLS marks > Higher Is Better EPYC 9965 ............... 365533 |================================= EPYC 9575F .............. 491451 |============================================= NVIDIA GH200 144G HBM3e . 266899 |======================== SecureMark 1.0.4 Benchmark: SecureMark-TLS marks Per Watt > Higher Is Better EPYC 9965 ............... 4523.07 |=============================== EPYC 9575F .............. 5300.98 |==================================== NVIDIA GH200 144G HBM3e . 6404.02 |============================================ Zstd Compression 1.5.4 Compression Level: 19 - Decompression Speed MB/s Per Watt > Higher Is Better EPYC 9965 ............... 16.54 |============================ EPYC 9575F .............. 18.19 |=============================== NVIDIA GH200 144G HBM3e . 27.05 |============================================== Zstd Compression 1.5.4 Compression Level: 19, Long Mode - Decompression Speed MB/s Per Watt > Higher Is Better EPYC 9965 ............... 16.95 |========================== EPYC 9575F .............. 19.41 |============================== NVIDIA GH200 144G HBM3e . 29.48 |============================================== Zstd Compression 1.5.4 Compression Level: 19 - Compression Speed MB/s > Higher Is Better EPYC 9965 ............... 19.4 |==================================== EPYC 9575F .............. 25.5 |=============================================== NVIDIA GH200 144G HBM3e . 15.2 |============================ Zstd Compression 1.5.4 Compression Level: 19 - Decompression Speed MB/s > Higher Is Better EPYC 9965 ............... 1520.5 |================================= EPYC 9575F .............. 2081.2 |============================================= NVIDIA GH200 144G HBM3e . 1266.6 |=========================== Zstd Compression 1.5.4 Compression Level: 19, Long Mode - Compression Speed MB/s > Higher Is Better EPYC 9965 ............... 9.29 |==================================== EPYC 9575F .............. 11.80 |============================================== NVIDIA GH200 144G HBM3e . 8.64 |================================== Zstd Compression 1.5.4 Compression Level: 19, Long Mode - Decompression Speed MB/s > Higher Is Better EPYC 9965 ............... 1469.0 |================================== EPYC 9575F .............. 1970.8 |============================================= NVIDIA GH200 144G HBM3e . 1305.9 |============================== srsRAN Project 23.10.1-20240325 Test: PUSCH Processor Benchmark, Throughput Total Mbps Per Watt > Higher Is Better EPYC 9965 ............... 28.62 |============================================== EPYC 9575F .............. 17.93 |============================= NVIDIA GH200 144G HBM3e . 16.72 |=========================== srsRAN Project 23.10.1-20240325 Test: PUSCH Processor Benchmark, Throughput Thread Mbps Per Watt > Higher Is Better EPYC 9965 ............... 1.751 |============================ EPYC 9575F .............. 2.863 |============================================== NVIDIA GH200 144G HBM3e . 1.797 |============================= srsRAN Project 23.10.1-20240325 Test: PUSCH Processor Benchmark, Throughput Total Mbps > Higher Is Better EPYC 9965 ............... 8940.9 |============================================= EPYC 9575F .............. 4804.7 |======================== NVIDIA GH200 144G HBM3e . 1848.2 |========= srsRAN Project 23.10.1-20240325 Test: PUSCH Processor Benchmark, Throughput Thread Mbps > Higher Is Better EPYC 9965 ............... 129.6 |=========================== EPYC 9575F .............. 224.5 |============================================== NVIDIA GH200 144G HBM3e . 75.0 |=============== QuantLib 1.32 Configuration: Multi-Threaded MFLOPS > Higher Is Better EPYC 9965 ............... 566481.4 |=========================================== EPYC 9575F .............. 297665.1 |======================= NVIDIA GH200 144G HBM3e . 261141.5 |==================== QuantLib 1.32 Configuration: Multi-Threaded MFLOPS Per Watt > Higher Is Better EPYC 9965 ............... 1606.22 |============================================ EPYC 9575F .............. 897.44 |========================= NVIDIA GH200 144G HBM3e . 1209.80 |================================= 7-Zip Compression 22.01 Test: Compression Rating MIPS > Higher Is Better EPYC 9965 ............... 1007150 |============================================ EPYC 9575F .............. 655241 |============================= NVIDIA GH200 144G HBM3e . 436600 |=================== 7-Zip Compression 22.01 Test: Decompression Rating MIPS > Higher Is Better EPYC 9965 ............... 1001858 |============================================ EPYC 9575F .............. 527037 |======================= NVIDIA GH200 144G HBM3e . 424643 |=================== 7-Zip Compression 22.01 Test: Decompression Rating MIPS Per Watt > Higher Is Better EPYC 9965 ............... 3214.79 |============================================ EPYC 9575F .............. 1634.38 |====================== NVIDIA GH200 144G HBM3e . 2612.52 |==================================== WebP Image Encode 1.2.4 Encode Settings: Quality 100, Lossless, Highest Compression MP/s Per Watt > Higher Is Better EPYC 9965 ............... 0.007 |=========================== EPYC 9575F .............. 0.009 |=================================== NVIDIA GH200 144G HBM3e . 0.012 |============================================== WebP Image Encode 1.2.4 Encode Settings: Quality 100, Lossless, Highest Compression MP/s > Higher Is Better EPYC 9965 ............... 0.61 |=================================== EPYC 9575F .............. 0.83 |=============================================== NVIDIA GH200 144G HBM3e . 0.54 |=============================== ASKAP 1.0 Test: tConvolve MPI - Gridding Mpix/sec Per Watt > Higher Is Better EPYC 9965 ............... 144.36 |============================================ EPYC 9575F .............. 146.79 |============================================= NVIDIA GH200 144G HBM3e . 72.88 |====================== ASKAP 1.0 Test: tConvolve MPI - Degridding Mpix/sec > Higher Is Better EPYC 9965 ............... 46390.5 |============================================ EPYC 9575F .............. 43732.6 |========================================= NVIDIA GH200 144G HBM3e . 15401.6 |=============== ASKAP 1.0 Test: tConvolve MPI - Gridding Mpix/sec > Higher Is Better EPYC 9965 ............... 44845.2 |========================================= EPYC 9575F .............. 48448.0 |============================================ NVIDIA GH200 144G HBM3e . 10933.8 |========== ASTC Encoder 4.7 Preset: Thorough MT/s > Higher Is Better EPYC 9965 ............... 128.76 |============================================= EPYC 9575F .............. 72.38 |========================= NVIDIA GH200 144G HBM3e . 46.41 |================ ASTC Encoder 4.7 Preset: Thorough MT/s Per Watt > Higher Is Better EPYC 9965 ............... 0.808 |============================================== EPYC 9575F .............. 0.329 |=================== NVIDIA GH200 144G HBM3e . 0.286 |================ ASTC Encoder 4.7 Preset: Very Thorough MT/s Per Watt > Higher Is Better EPYC 9965 ............... 0.075 |============================================== EPYC 9575F .............. 0.032 |==================== NVIDIA GH200 144G HBM3e . 0.030 |================== ASTC Encoder 4.7 Preset: Very Thorough MT/s > Higher Is Better EPYC 9965 ............... 18.7713 |============================================ EPYC 9575F .............. 10.1197 |======================== NVIDIA GH200 144G HBM3e . 6.5885 |=============== ASTC Encoder 4.7 Preset: Exhaustive MT/s > Higher Is Better EPYC 9965 ............... 11.5312 |============================================ EPYC 9575F .............. 6.2035 |======================== NVIDIA GH200 144G HBM3e . 4.0660 |================ ASTC Encoder 4.7 Preset: Exhaustive MT/s Per Watt > Higher Is Better EPYC 9965 ............... 0.046 |============================================== EPYC 9575F .............. 0.020 |==================== NVIDIA GH200 144G HBM3e . 0.018 |================== Stockfish 16.1 Chess Benchmark Nodes Per Second > Higher Is Better EPYC 9965 ............... 362007178 |========================================== EPYC 9575F .............. 182135967 |===================== NVIDIA GH200 144G HBM3e . 98146858 |=========== Stockfish 16.1 Chess Benchmark Nodes Per Second Per Watt > Higher Is Better EPYC 9965 ............... 931347.77 |========================================== EPYC 9575F .............. 490782.75 |====================== NVIDIA GH200 144G HBM3e . 375426.89 |================= GROMACS 2024 Implementation: MPI CPU - Input: water_GMX50_bare Ns Per Day Per Watt > Higher Is Better EPYC 9965 ............... 0.080 |============================================== EPYC 9575F .............. 0.046 |========================== NVIDIA GH200 144G HBM3e . 0.028 |================ GROMACS 2024 Implementation: MPI CPU - Input: water_GMX50_bare Ns Per Day > Higher Is Better EPYC 9965 ............... 23.142 |============================================= EPYC 9575F .............. 14.651 |============================ NVIDIA GH200 144G HBM3e . 6.002 |============ LAMMPS Molecular Dynamics Simulator 23Jun2022 Model: Rhodopsin Protein ns/day Per Watt > Higher Is Better EPYC 9965 ............... 0.422 |==================== EPYC 9575F .............. 0.462 |====================== NVIDIA GH200 144G HBM3e . 0.967 |============================================== LAMMPS Molecular Dynamics Simulator 23Jun2022 Model: 20k Atoms ns/day Per Watt > Higher Is Better EPYC 9965 ............... 0.168 |==================================== EPYC 9575F .............. 0.141 |=============================== NVIDIA GH200 144G HBM3e . 0.212 |============================================== NAMD 3.0b6 Input: ATPase with 327,506 Atoms ns/day Per Watt > Higher Is Better EPYC 9965 .. 0.179 |=========================================================== EPYC 9575F . 0.056 |================== LAMMPS Molecular Dynamics Simulator 23Jun2022 Model: Rhodopsin Protein ns/day > Higher Is Better EPYC 9965 ............... 55.02 |========================================== EPYC 9575F .............. 42.85 |================================= NVIDIA GH200 144G HBM3e . 59.84 |============================================== LAMMPS Molecular Dynamics Simulator 23Jun2022 Model: 20k Atoms ns/day > Higher Is Better EPYC 9965 ............... 66.77 |============================================== EPYC 9575F .............. 54.79 |====================================== NVIDIA GH200 144G HBM3e . 58.14 |======================================== NAMD 3.0b6 Input: ATPase with 327,506 Atoms ns/day > Higher Is Better EPYC 9965 .. 19.90 |=========================================================== EPYC 9575F . 13.34 |======================================== NAMD 3.0b6 Input: STMV with 1,066,628 Atoms ns/day Per Watt > Higher Is Better EPYC 9965 .. 0.037 |=========================================================== EPYC 9575F . 0.014 |====================== NAMD 3.0b6 Input: STMV with 1,066,628 Atoms ns/day > Higher Is Better EPYC 9965 .. 5.03185 |========================================================= EPYC 9575F . 3.80313 |=========================================== Speedb 2.7 Test: Random Read Op/s > Higher Is Better EPYC 9965 ............... 969890739 |========================================== EPYC 9575F .............. 526671792 |======================= NVIDIA GH200 144G HBM3e . 569021111 |========================= RocksDB 9.0 Test: Random Read Op/s Per Watt > Higher Is Better EPYC 9965 ............... 2704939.66 |========================================= EPYC 9575F .............. 1462898.91 |====================== NVIDIA GH200 144G HBM3e . 2103254.00 |================================ Speedb 2.7 Test: Random Read Op/s Per Watt > Higher Is Better EPYC 9965 ............... 2712772.81 |========================================= EPYC 9575F .............. 1520207.33 |======================= NVIDIA GH200 144G HBM3e . 2132224.78 |================================ Speedb 2.7 Test: Read While Writing Op/s Per Watt > Higher Is Better EPYC 9965 ............... 76045.35 |=========================================== EPYC 9575F .............. 28692.82 |================ NVIDIA GH200 144G HBM3e . 56303.84 |================================ RocksDB 9.0 Test: Random Read Op/s > Higher Is Better EPYC 9965 ............... 958354064 |========================================== EPYC 9575F .............. 505733525 |====================== NVIDIA GH200 144G HBM3e . 568339463 |========================= Speedb 2.7 Test: Read While Writing Op/s > Higher Is Better EPYC 9965 ............... 23940015 |=========================================== EPYC 9575F .............. 10783281 |=================== NVIDIA GH200 144G HBM3e . 10241042 |================== Memcached 1.6.19 Set To Get Ratio: 1:100 Ops/sec Per Watt > Higher Is Better EPYC 9965 ............... 35232.59 |====================================== EPYC 9575F .............. 39869.27 |=========================================== NVIDIA GH200 144G HBM3e . 27521.07 |============================== Memcached 1.6.19 Set To Get Ratio: 1:100 Ops/sec > Higher Is Better EPYC 9965 ............... 9290017.46 |=========================== EPYC 9575F .............. 13808558.42 |======================================== NVIDIA GH200 144G HBM3e . 3049705.54 |========= Meta Performance Per Watts Performance Per Watts Performance Per Watts > Higher Is Better NVIDIA GH200 144G HBM3e . 227401.82 |========================================== ClickHouse 22.12.3.5 100M Rows Hits Dataset, Third Run Queries Per Minute, Geo Mean Per Watt > Higher Is Better EPYC 9965 ............... 4.202 |======================== EPYC 9575F .............. 4.181 |======================== NVIDIA GH200 144G HBM3e . 7.980 |============================================== ClickHouse 22.12.3.5 100M Rows Hits Dataset, First Run / Cold Cache Queries Per Minute, Geo Mean > Higher Is Better EPYC 9965 ............... 576.76 |================================= EPYC 9575F .............. 779.31 |============================================= NVIDIA GH200 144G HBM3e . 581.59 |================================== ClickHouse 22.12.3.5 100M Rows Hits Dataset, Second Run Queries Per Minute, Geo Mean > Higher Is Better EPYC 9965 ............... 587.17 |================================= EPYC 9575F .............. 809.27 |============================================= NVIDIA GH200 144G HBM3e . 594.01 |================================= ClickHouse 22.12.3.5 100M Rows Hits Dataset, Third Run Queries Per Minute, Geo Mean > Higher Is Better EPYC 9965 ............... 598.32 |================================= EPYC 9575F .............. 809.87 |============================================= NVIDIA GH200 144G HBM3e . 596.49 |================================= John The Ripper 2023.03.14 Test: Blowfish Real C/S Per Watt > Higher Is Better EPYC 9965 ............... 1128.50 |============================================ EPYC 9575F .............. 562.77 |====================== NVIDIA GH200 144G HBM3e . 508.72 |==================== John The Ripper 2023.03.14 Test: bcrypt Real C/S Per Watt > Higher Is Better EPYC 9965 ............... 1127.52 |============================================ EPYC 9575F .............. 561.71 |====================== NVIDIA GH200 144G HBM3e . 517.44 |==================== John The Ripper 2023.03.14 Test: WPA PSK Real C/S Per Watt > Higher Is Better EPYC 9965 ............... 4459.57 |============================================ EPYC 9575F .............. 2539.28 |========================= NVIDIA GH200 144G HBM3e . 525.73 |===== John The Ripper 2023.03.14 Test: Blowfish Real C/S > Higher Is Better EPYC 9965 ............... 386082 |============================================= EPYC 9575F .............. 198954 |======================= NVIDIA GH200 144G HBM3e . 74756 |========= John The Ripper 2023.03.14 Test: bcrypt Real C/S > Higher Is Better EPYC 9965 ............... 386474 |============================================= EPYC 9575F .............. 199078 |======================= NVIDIA GH200 144G HBM3e . 74749 |========= John The Ripper 2023.03.14 Test: WPA PSK Real C/S > Higher Is Better EPYC 9965 ............... 1774667 |============================================ EPYC 9575F .............. 857702 |===================== NVIDIA GH200 144G HBM3e . 76130 |== Liquid-DSP 1.6 Threads: 1 - Buffer Length: 256 - Filter Length: 32 samples/s > Higher Is Better EPYC 9965 ............... 39896000 |============================ EPYC 9575F .............. 53801667 |====================================== NVIDIA GH200 144G HBM3e . 60790333 |=========================================== Liquid-DSP 1.6 Threads: 1 - Buffer Length: 256 - Filter Length: 512 samples/s > Higher Is Better EPYC 9965 ............... 20527750 |=============================== EPYC 9575F .............. 28274667 |=========================================== NVIDIA GH200 144G HBM3e . 3458733 |===== Liquid-DSP 1.6 Threads: 128 - Buffer Length: 256 - Filter Length: 512 samples/s > Higher Is Better EPYC 9965 ............... 2281700000 |========================================= EPYC 9575F .............. 1822933333 |================================= NVIDIA GH200 144G HBM3e . 248263333 |==== Liquid-DSP 1.6 Threads: 256 - Buffer Length: 256 - Filter Length: 32 samples/s > Higher Is Better EPYC 9965 ............... 8064266667 |========================================= EPYC 9575F .............. 5265866667 |=========================== NVIDIA GH200 144G HBM3e . 4502400000 |======================= Liquid-DSP 1.6 Threads: 1 - Buffer Length: 256 - Filter Length: 32 samples/s Per Watt > Higher Is Better EPYC 9965 ............... 525714.32 |================ EPYC 9575F .............. 637828.74 |=================== NVIDIA GH200 144G HBM3e . 1381459.61 |========================================= Liquid-DSP 1.6 Threads: 256 - Buffer Length: 256 - Filter Length: 57 samples/s > Higher Is Better EPYC 9965 ............... 8185800000 |========================================= EPYC 9575F .............. 5442600000 |=========================== NVIDIA GH200 144G HBM3e . 2669733333 |============= Liquid-DSP 1.6 Threads: 1 - Buffer Length: 256 - Filter Length: 512 samples/s Per Watt > Higher Is Better EPYC 9965 ............... 266440.14 |================================== EPYC 9575F .............. 325399.16 |========================================== NVIDIA GH200 144G HBM3e . 82973.28 |=========== Liquid-DSP 1.6 Threads: 128 - Buffer Length: 256 - Filter Length: 512 samples/s Per Watt > Higher Is Better EPYC 9965 ............... 6644407.92 |========================================= EPYC 9575F .............. 5541653.37 |================================== NVIDIA GH200 144G HBM3e . 1396900.82 |========= Liquid-DSP 1.6 Threads: 256 - Buffer Length: 256 - Filter Length: 32 samples/s Per Watt > Higher Is Better EPYC 9965 ............... 23123930.33 |======================================== EPYC 9575F .............. 14781256.88 |========================== NVIDIA GH200 144G HBM3e . 17535441.66 |============================== Liquid-DSP 1.6 Threads: 256 - Buffer Length: 256 - Filter Length: 57 samples/s Per Watt > Higher Is Better EPYC 9965 ............... 23573051.76 |======================================== EPYC 9575F .............. 15772371.31 |=========================== NVIDIA GH200 144G HBM3e . 11492978.86 |==================== Liquid-DSP 1.6 Threads: 256 - Buffer Length: 256 - Filter Length: 512 samples/s Per Watt > Higher Is Better EPYC 9965 ............... 8815828.44 |========================================= EPYC 9575F .............. 5494414.93 |========================== NVIDIA GH200 144G HBM3e . 1400334.37 |======= Liquid-DSP 1.6 Threads: 256 - Buffer Length: 256 - Filter Length: 512 samples/s > Higher Is Better EPYC 9965 ............... 3181633333 |========================================= EPYC 9575F .............. 1817933333 |======================= NVIDIA GH200 144G HBM3e . 253630000 |=== Numpy Benchmark Score > Higher Is Better EPYC 9965 ............... 723.81 |================================== EPYC 9575F .............. 962.28 |============================================= NVIDIA GH200 144G HBM3e . 699.30 |================================= Numpy Benchmark Score Per Watt > Higher Is Better EPYC 9965 ............... 8.873 |========================= EPYC 9575F .............. 10.291 |============================ NVIDIA GH200 144G HBM3e . 16.277 |============================================= OpenSSL 3.3 Algorithm: RSA4096 sign/s > Higher Is Better EPYC 9965 ............... 85251.0 |============================================ EPYC 9575F .............. 45479.7 |======================= NVIDIA GH200 144G HBM3e . 15332.5 |======== NAS Parallel Benchmarks 3.4 Test / Class: LU.C Total Mop/s Per Watt > Higher Is Better EPYC 9965 ............... 1309.07 |============================================ EPYC 9575F .............. 1157.05 |======================================= NVIDIA GH200 144G HBM3e . 204.41 |======= NAS Parallel Benchmarks 3.4 Test / Class: SP.C Total Mop/s Per Watt > Higher Is Better EPYC 9965 ............... 566.79 |============================================= EPYC 9575F .............. 518.54 |========================================= NVIDIA GH200 144G HBM3e . 105.21 |======== NAS Parallel Benchmarks 3.4 Test / Class: MG.C Total Mop/s Per Watt > Higher Is Better EPYC 9965 ............... 791.79 |============================ EPYC 9575F .............. 1247.11 |============================================ NVIDIA GH200 144G HBM3e . 529.28 |=================== NAS Parallel Benchmarks 3.4 Test / Class: LU.C Total Mop/s > Higher Is Better EPYC 9965 ............... 313412.10 |========================================== EPYC 9575F .............. 303092.49 |========================================= NVIDIA GH200 144G HBM3e . 44257.15 |====== NAS Parallel Benchmarks 3.4 Test / Class: IS.D Total Mop/s Per Watt > Higher Is Better EPYC 9965 ............... 40.83 |============================================== EPYC 9575F .............. 30.31 |================================== NVIDIA GH200 144G HBM3e . 14.06 |================ NAS Parallel Benchmarks 3.4 Test / Class: CG.C Total Mop/s Per Watt > Higher Is Better EPYC 9965 ............... 280.56 |================================= EPYC 9575F .............. 381.14 |============================================= NVIDIA GH200 144G HBM3e . 175.40 |===================== NAS Parallel Benchmarks 3.4 Test / Class: SP.C Total Mop/s > Higher Is Better EPYC 9965 ............... 152747.87 |========================================== EPYC 9575F .............. 148567.25 |========================================= NVIDIA GH200 144G HBM3e . 24153.15 |======= NAS Parallel Benchmarks 3.4 Test / Class: IS.D Total Mop/s > Higher Is Better EPYC 9965 ............... 7503.01 |============================================ EPYC 9575F .............. 7043.18 |========================================= NVIDIA GH200 144G HBM3e . 1931.21 |=========== NAS Parallel Benchmarks 3.4 Test / Class: MG.C Total Mop/s > Higher Is Better EPYC 9965 ............... 112588.99 |============================== EPYC 9575F .............. 159641.78 |========================================== NVIDIA GH200 144G HBM3e . 63701.22 |================= NAS Parallel Benchmarks 3.4 Test / Class: CG.C Total Mop/s > Higher Is Better EPYC 9965 ............... 51319.46 |================================== EPYC 9575F .............. 65470.23 |=========================================== NVIDIA GH200 144G HBM3e . 27521.87 |================== OpenSSL 3.3 Algorithm: RSA4096 verify/s Per Watt > Higher Is Better EPYC 9965 ............... 8963.31 |============================================ EPYC 9575F .............. 4395.32 |====================== NVIDIA GH200 144G HBM3e . 5640.20 |============================ OpenSSL 3.3 Algorithm: RSA4096 verify/s > Higher Is Better EPYC 9965 ............... 3351713.3 |========================================== EPYC 9575F .............. 1673091.2 |===================== NVIDIA GH200 144G HBM3e . 1055054.5 |============= BRL-CAD 7.38.2 VGR Performance Metric VGR Performance Metric > Higher Is Better EPYC 9965 ............... 8348390 |============================================ EPYC 9575F .............. 1406490 |======= BRL-CAD 7.38.2 VGR Performance Metric VGR Performance Metric Per Watt > Higher Is Better EPYC 9965 .. 83197.80 |======================================================== EPYC 9575F . 3698.57 |== Pennant 1.0.1 Test: leblancbig Hydro Cycle Time - Seconds < Lower Is Better EPYC 9965 ............... 2.379513 |======================= EPYC 9575F .............. 3.238866 |=============================== NVIDIA GH200 144G HBM3e . 4.473779 |=========================================== Pennant 1.0.1 Test: sedovbig Hydro Cycle Time - Seconds < Lower Is Better EPYC 9965 ............... 3.806965 |========================= EPYC 9575F .............. 6.221108 |========================================== NVIDIA GH200 144G HBM3e . 6.430304 |=========================================== PyBench 2018-02-16 Total For Average Test Times Milliseconds < Lower Is Better EPYC 9965 ............... 646 |=========================================== EPYC 9575F .............. 482 |================================ NVIDIA GH200 144G HBM3e . 724 |================================================ OSPRay Studio 1.0 Camera: 1 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU ms < Lower Is Better EPYC 9965 .. 535 |================================ EPYC 9575F . 1015 |============================================================ OSPRay Studio 1.0 Camera: 1 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPU ms < Lower Is Better EPYC 9965 .. 8530 |=============================== EPYC 9575F . 16174 |=========================================================== OSPRay Studio 1.0 Camera: 1 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU ms < Lower Is Better EPYC 9965 .. 17091 |============================= EPYC 9575F . 35274 |=========================================================== OSPRay Studio 1.0 Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer - Acceleration: CPU ms < Lower Is Better EPYC 9965 .. 632 |================================ EPYC 9575F . 1203 |============================================================ OSPRay Studio 1.0 Camera: 3 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer - Acceleration: CPU ms < Lower Is Better EPYC 9965 .. 10067 |=============================== EPYC 9575F . 19196 |=========================================================== OSPRay Studio 1.0 Camera: 3 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer - Acceleration: CPU ms < Lower Is Better EPYC 9965 .. 20149 |============================= EPYC 9575F . 41324 |=========================================================== OpenVINO 2024.0 Model: Face Detection FP16-INT8 - Device: CPU ms < Lower Is Better EPYC 9965 ............... 414.32 |======= EPYC 9575F .............. 225.86 |==== NVIDIA GH200 144G HBM3e . 2708.94 |============================================ OpenVINO 2024.0 Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU ms < Lower Is Better EPYC 9965 ............... 0.73 |========= EPYC 9575F .............. 0.25 |=== NVIDIA GH200 144G HBM3e . 3.67 |=============================================== OpenVINO 2024.0 Model: Person Detection FP16 - Device: CPU ms < Lower Is Better EPYC 9965 ............... 149.81 |============ EPYC 9575F .............. 49.09 |==== NVIDIA GH200 144G HBM3e . 549.25 |============================================= OpenVINO 2024.0 Model: Weld Porosity Detection FP16-INT8 - Device: CPU ms < Lower Is Better EPYC 9965 ............... 8.38 |========== EPYC 9575F .............. 4.60 |===== NVIDIA GH200 144G HBM3e . 38.63 |============================================== OpenVINO 2024.0 Model: Vehicle Detection FP16-INT8 - Device: CPU ms < Lower Is Better EPYC 9965 ............... 7.52 |=== EPYC 9575F .............. 4.24 |== NVIDIA GH200 144G HBM3e . 117.25 |============================================= OpenVINO 2024.0 Model: Person Vehicle Bike Detection FP16 - Device: CPU ms < Lower Is Better EPYC 9965 ............... 8.63 |============ EPYC 9575F .............. 5.52 |======== NVIDIA GH200 144G HBM3e . 32.89 |============================================== OpenVINO 2024.0 Model: Machine Translation EN To DE FP16 - Device: CPU ms < Lower Is Better EPYC 9965 ............... 78.07 |====================== EPYC 9575F .............. 41.94 |============ NVIDIA GH200 144G HBM3e . 161.98 |============================================= OpenVINO 2024.0 Model: Face Detection Retail FP16-INT8 - Device: CPU ms < Lower Is Better EPYC 9965 ............... 4.97 |======= EPYC 9575F .............. 2.85 |==== NVIDIA GH200 144G HBM3e . 34.66 |============================================== OpenVINO 2024.0 Model: Road Segmentation ADAS FP16-INT8 - Device: CPU ms < Lower Is Better EPYC 9965 ............... 27.34 |=== EPYC 9575F .............. 14.41 |== NVIDIA GH200 144G HBM3e . 351.90 |============================================= OpenVINO 2024.0 Model: Noise Suppression Poconet-Like FP16 - Device: CPU ms < Lower Is Better EPYC 9965 ............... 10.88 |====== EPYC 9575F .............. 8.38 |==== NVIDIA GH200 144G HBM3e . 90.20 |============================================== Timed Linux Kernel Compilation 6.8 Build: allmodconfig Seconds < Lower Is Better EPYC 9965 ............... 248.29 |============================================= EPYC 9575F .............. 203.80 |===================================== NVIDIA GH200 144G HBM3e . 246.66 |============================================= Timed Godot Game Engine Compilation 4.0 Time To Compile Seconds < Lower Is Better EPYC 9965 ............... 91.05 |================================= EPYC 9575F .............. 83.44 |============================== NVIDIA GH200 144G HBM3e . 124.00 |============================================= Timed Node.js Compilation 21.7.2 Time To Compile Seconds < Lower Is Better EPYC 9965 ............... 117.82 |========================== EPYC 9575F .............. 133.86 |============================== NVIDIA GH200 144G HBM3e . 202.94 |============================================= Timed Gem5 Compilation 23.0.1 Time To Compile Seconds < Lower Is Better EPYC 9965 ............... 140.59 |====================================== EPYC 9575F .............. 130.80 |=================================== NVIDIA GH200 144G HBM3e . 166.10 |============================================= Timed LLVM Compilation 16.0 Build System: Ninja Seconds < Lower Is Better EPYC 9965 ............... 99.43 |========================== EPYC 9575F .............. 110.53 |============================= NVIDIA GH200 144G HBM3e . 168.97 |============================================= RELION 4.0.1 Test: Basic - Device: CPU Seconds < Lower Is Better EPYC 9965 .. 169.70 |============================================ EPYC 9575F . 223.81 |========================================================== CloverLeaf 1.3 Input: clover_bm64_short Seconds < Lower Is Better EPYC 9965 ............... 24.11 |======================================= EPYC 9575F .............. 21.12 |================================== NVIDIA GH200 144G HBM3e . 28.41 |============================================== GPAW 23.6 Input: Carbon Nanotube Seconds < Lower Is Better EPYC 9965 ............... 27.56 |========================================== EPYC 9575F .............. 29.87 |============================================== NWChem 7.0.2 Input: C240 Buckyball Seconds < Lower Is Better EPYC 9965 ............... 1594.0 |============================================= EPYC 9575F .............. 1225.7 |=================================== NVIDIA GH200 144G HBM3e . 1214.7 |================================== Xcompact3d Incompact3d 2021-03-11 Input: X3D-benchmarking input.i3d Seconds < Lower Is Better EPYC 9965 ............... 196.32 |===================================== EPYC 9575F .............. 217.38 |========================================= NVIDIA GH200 144G HBM3e . 240.29 |============================================= OpenFOAM 10 Input: drivaerFastback, Small Mesh Size - Execution Time Seconds < Lower Is Better EPYC 9965 ............... 25.34 |=============================== EPYC 9575F .............. 25.60 |================================ NVIDIA GH200 144G HBM3e . 37.37 |============================================== OpenFOAM 10 Input: drivaerFastback, Medium Mesh Size - Mesh Time Seconds < Lower Is Better EPYC 9965 ............... 160.68 |============================================= EPYC 9575F .............. 95.09 |=========================== NVIDIA GH200 144G HBM3e . 83.98 |======================== OpenFOAM 10 Input: drivaerFastback, Medium Mesh Size - Execution Time Seconds < Lower Is Better EPYC 9965 ............... 176.40 |====================== EPYC 9575F .............. 233.12 |============================= NVIDIA GH200 144G HBM3e . 362.27 |============================================= OpenRadioss 2023.09.15 Model: Cell Phone Drop Test Seconds < Lower Is Better EPYC 9965 .. 22.03 |=========================================================== EPYC 9575F . 17.64 |=============================================== OpenRadioss 2023.09.15 Model: INIVOL and Fluid Structure Interaction Drop Container Seconds < Lower Is Better EPYC 9965 .. 77.76 |=================================================== EPYC 9575F . 89.66 |=========================================================== OpenRadioss 2023.09.15 Model: Chrysler Neon 1M Seconds < Lower Is Better EPYC 9965 .. 92.04 |================================================= EPYC 9575F . 109.72 |========================================================== Blender 4.1 Blend File: BMW27 - Compute: CPU-Only Seconds < Lower Is Better EPYC 9965 .. 8.20 |================================= EPYC 9575F . 14.63 |=========================================================== Blender 4.1 Blend File: Classroom - Compute: CPU-Only Seconds < Lower Is Better EPYC 9965 .. 20.31 |=============================== EPYC 9575F . 38.21 |=========================================================== Blender 4.1 Blend File: Fishy Cat - Compute: CPU-Only Seconds < Lower Is Better EPYC 9965 .. 11.02 |================================== EPYC 9575F . 19.12 |=========================================================== Blender 4.1 Blend File: Pabellon Barcelona - Compute: CPU-Only Seconds < Lower Is Better EPYC 9965 .. 25.67 |================================= EPYC 9575F . 45.75 |=========================================================== Blender 4.1 Blend File: Barbershop - Compute: CPU-Only Seconds < Lower Is Better EPYC 9965 .. 74.24 |=============================== EPYC 9575F . 138.36 |========================================================== Blender 4.1 Blend File: Junkshop - Compute: CPU-Only Seconds < Lower Is Better EPYC 9965 .. 11.26 |================================= EPYC 9575F . 19.84 |=========================================================== Parallel BZIP2 Compression 1.1.13 FreeBSD-13.0-RELEASE-amd64-memstick.img Compression Seconds < Lower Is Better EPYC 9965 ............... 0.891238 |======================= EPYC 9575F .............. 1.263408 |================================ NVIDIA GH200 144G HBM3e . 1.682797 |=========================================== libavif avifenc 1.0 Encoder Speed: 0 Seconds < Lower Is Better EPYC 9965 ............... 53.27 |====================== EPYC 9575F .............. 40.85 |================= NVIDIA GH200 144G HBM3e . 108.21 |============================================= Helsing 1.0-beta Digit Range: 14 digit Seconds < Lower Is Better EPYC 9965 ............... 39.77 |============================== EPYC 9575F .............. 61.83 |============================================== NVIDIA GH200 144G HBM3e . 56.97 |========================================== Primesieve 12.1 Length: 1e12 Seconds < Lower Is Better EPYC 9965 ............... 1.396 |======================== EPYC 9575F .............. 2.123 |===================================== NVIDIA GH200 144G HBM3e . 2.667 |============================================== Primesieve 12.1 Length: 1e13 Seconds < Lower Is Better EPYC 9965 ............... 15.98 |======================= EPYC 9575F .............. 25.40 |===================================== NVIDIA GH200 144G HBM3e . 31.46 |============================================== OpenFOAM 10 Input: drivaerFastback, Small Mesh Size - Mesh Time Seconds < Lower Is Better NVIDIA GH200 144G HBM3e . 16.74 |============================================== QMCPACK 3.17.1 Input: Li2_STO_ae Total Execution Time - Seconds < Lower Is Better EPYC 9965 ............... 94.59 |============================================== EPYC 9575F .............. 72.93 |=================================== NVIDIA GH200 144G HBM3e . 69.61 |================================== Timed Linux Kernel Compilation 6.8 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 308 MAX: 388 EPYC 9575F .............. MIN: 39 AVG: 352 MAX: 401 NVIDIA GH200 144G HBM3e . MIN: 39 AVG: 230 MAX: 271 Timed Godot Game Engine Compilation 4.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 39 AVG: 177 MAX: 410 EPYC 9575F .............. MIN: 39 AVG: 249 MAX: 401 NVIDIA GH200 144G HBM3e . MIN: 39 AVG: 141 MAX: 273 Timed Node.js Compilation 21.7.2 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 258 MAX: 407 EPYC 9575F .............. MIN: 39 AVG: 314 MAX: 401 NVIDIA GH200 144G HBM3e . MIN: 39 AVG: 181 MAX: 279 Timed Gem5 Compilation 23.0.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 164 MAX: 402 EPYC 9575F .............. MIN: 38 AVG: 212 MAX: 401 NVIDIA GH200 144G HBM3e . MIN: 39 AVG: 119 MAX: 239 Timed LLVM Compilation 16.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 254 MAX: 404 EPYC 9575F .............. MIN: 38 AVG: 313 MAX: 401 NVIDIA GH200 144G HBM3e . MIN: 39 AVG: 167 MAX: 240 OpenSSL 3.3 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 374 MAX: 397 EPYC 9575F .............. MIN: 39 AVG: 381 MAX: 401 NVIDIA GH200 144G HBM3e . MIN: 40 AVG: 187 MAX: 196 OpenSSL 3.3 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 378 MAX: 402 EPYC 9575F .............. MIN: 39 AVG: 385 MAX: 401 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 197 MAX: 273 OpenSSL 3.3 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 371 MAX: 392 EPYC 9575F .............. MIN: 39 AVG: 385 MAX: 401 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 199 MAX: 243 OpenSSL 3.3 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 388 MAX: 420 EPYC 9575F .............. MIN: 38 AVG: 358 MAX: 378 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 285 MAX: 293 OpenSSL 3.3 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 381 MAX: 413 EPYC 9575F .............. MIN: 38 AVG: 359 MAX: 383 NVIDIA GH200 144G HBM3e . MIN: 44 AVG: 285 MAX: 294 OpenSSL 3.3 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 424 MAX: 462 EPYC 9575F .............. MIN: 39 AVG: 384 MAX: 401 NVIDIA GH200 144G HBM3e . MIN: 43 AVG: 194 MAX: 217 OpenSSL 3.3 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 39 AVG: 420 MAX: 459 EPYC 9575F .............. MIN: 38 AVG: 383 MAX: 401 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 203 MAX: 225 John The Ripper 2023.03.14 CPU Power Consumption Monitor Watts < Lower Is Better Xeon 6980P .............. MIN: 51 AVG: 445 MAX: 540 EPYC 9575F .............. MIN: 38 AVG: 354 MAX: 400 EPYC 9965 ............... MIN: 38 AVG: 343 MAX: 386 Xeon 6980P - DDR5-6400 .. MIN: 50 AVG: 447 MAX: 540 EPYC 9965 ............... MIN: 38 AVG: 342 MAX: 387 EPYC 9575F .............. MIN: 38 AVG: 354 MAX: 400 NVIDIA GH200 144G HBM3e . MIN: 42 AVG: 147 MAX: 165 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 144 MAX: 165 John The Ripper 2023.03.14 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 398 MAX: 450 EPYC 9575F .............. MIN: 39 AVG: 338 MAX: 381 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 145 MAX: 162 ClickHouse 22.12.3.5 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 142 MAX: 305 EPYC 9575F .............. MIN: 38 AVG: 194 MAX: 368 NVIDIA GH200 144G HBM3e . MIN: 40 AVG: 75 MAX: 283 Memcached 1.6.19 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 264 MAX: 306 EPYC 9575F .............. MIN: 38 AVG: 346 MAX: 402 NVIDIA GH200 144G HBM3e . MIN: 40 AVG: 111 MAX: 125 RocksDB 9.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 354 MAX: 380 EPYC 9575F .............. MIN: 38 AVG: 346 MAX: 371 NVIDIA GH200 144G HBM3e . MIN: 40 AVG: 270 MAX: 291 Speedb 2.7 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 358 MAX: 384 EPYC 9575F .............. MIN: 38 AVG: 346 MAX: 371 NVIDIA GH200 144G HBM3e . MIN: 42 AVG: 267 MAX: 287 Speedb 2.7 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 39 AVG: 315 MAX: 417 EPYC 9575F .............. MIN: 38 AVG: 376 MAX: 401 NVIDIA GH200 144G HBM3e . MIN: 43 AVG: 182 MAX: 216 SecureMark 1.0.4 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 81 MAX: 83 EPYC 9575F .............. MIN: 38 AVG: 93 MAX: 97 NVIDIA GH200 144G HBM3e . MIN: 39 AVG: 42 MAX: 44 Coremark 1.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 261 MAX: 347 EPYC 9575F .............. MIN: 38 AVG: 314 MAX: 400 NVIDIA GH200 144G HBM3e . MIN: 39 AVG: 119 MAX: 160 Algebraic Multi-Grid Benchmark 1.2 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 297 MAX: 370 EPYC 9575F .............. MIN: 38 AVG: 294 MAX: 401 NVIDIA GH200 144G HBM3e . MIN: 40 AVG: 146 MAX: 260 RELION 4.0.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 376 MAX: 395 EPYC 9575F . MIN: 38 AVG: 389 MAX: 400 miniBUDE 20210901 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 296 MAX: 389 EPYC 9575F .............. MIN: 38 AVG: 320 MAX: 383 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 214 MAX: 245 LAMMPS Molecular Dynamics Simulator 23Jun2022 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 20 AVG: 130 MAX: 331 EPYC 9575F .............. MIN: 38 AVG: 93 MAX: 302 NVIDIA GH200 144G HBM3e . MIN: 40 AVG: 62 MAX: 280 LAMMPS Molecular Dynamics Simulator 23Jun2022 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 397 MAX: 423 EPYC 9575F .............. MIN: 38 AVG: 389 MAX: 400 NVIDIA GH200 144G HBM3e . MIN: 40 AVG: 275 MAX: 292 miniFE 2.2 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 157 MAX: 341 EPYC 9575F .............. MIN: 38 AVG: 186 MAX: 400 NVIDIA GH200 144G HBM3e . MIN: 42 AVG: 103 MAX: 197 ASKAP 1.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 311 MAX: 397 EPYC 9575F .............. MIN: 38 AVG: 330 MAX: 400 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 150 MAX: 266 NAMD 3.0b6 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 111 MAX: 383 EPYC 9575F . MIN: 38 AVG: 236 MAX: 398 NAMD 3.0b6 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 135 MAX: 374 EPYC 9575F . MIN: 38 AVG: 276 MAX: 398 GROMACS 2024 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 289 MAX: 417 EPYC 9575F .............. MIN: 38 AVG: 315 MAX: 400 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 212 MAX: 289 CloverLeaf 1.3 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 286 MAX: 340 EPYC 9575F .............. MIN: 38 AVG: 332 MAX: 399 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 142 MAX: 183 QuantLib 1.32 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 353 MAX: 388 EPYC 9575F .............. MIN: 39 AVG: 332 MAX: 399 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 216 MAX: 290 QMCPACK 3.17.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 360 MAX: 391 EPYC 9575F .............. MIN: 39 AVG: 376 MAX: 400 NVIDIA GH200 144G HBM3e . MIN: 42 AVG: 268 MAX: 298 GPAW 23.6 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 325 MAX: 417 EPYC 9575F . MIN: 38 AVG: 352 MAX: 400 High Performance Conjugate Gradient 3.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 345 MAX: 388 EPYC 9575F .............. MIN: 38 AVG: 391 MAX: 400 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 177 MAX: 219 Pennant 1.0.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 172 MAX: 384 EPYC 9575F .............. MIN: 38 AVG: 196 MAX: 398 NVIDIA GH200 144G HBM3e . MIN: 42 AVG: 140 MAX: 266 Pennant 1.0.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 197 MAX: 381 EPYC 9575F .............. MIN: 38 AVG: 248 MAX: 400 NVIDIA GH200 144G HBM3e . MIN: 42 AVG: 151 MAX: 253 NAS Parallel Benchmarks 3.4 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 239 MAX: 392 EPYC 9575F .............. MIN: 39 AVG: 262 MAX: 400 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 217 MAX: 255 NAS Parallel Benchmarks 3.4 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 269 MAX: 413 EPYC 9575F .............. MIN: 38 AVG: 287 MAX: 400 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 230 MAX: 257 NAS Parallel Benchmarks 3.4 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 184 MAX: 364 EPYC 9575F .............. MIN: 38 AVG: 232 MAX: 400 NVIDIA GH200 144G HBM3e . MIN: 42 AVG: 137 MAX: 236 NAS Parallel Benchmarks 3.4 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 142 MAX: 374 EPYC 9575F .............. MIN: 38 AVG: 128 MAX: 396 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 120 MAX: 274 NAS Parallel Benchmarks 3.4 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 183 MAX: 400 EPYC 9575F .............. MIN: 38 AVG: 172 MAX: 400 NVIDIA GH200 144G HBM3e . MIN: 40 AVG: 157 MAX: 271 NWChem 7.0.2 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 388 MAX: 410 EPYC 9575F .............. MIN: 38 AVG: 393 MAX: 400 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 227 MAX: 281 Xcompact3d Incompact3d 2021-03-11 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 358 MAX: 390 EPYC 9575F .............. MIN: 38 AVG: 390 MAX: 400 NVIDIA GH200 144G HBM3e . MIN: 42 AVG: 173 MAX: 253 BRL-CAD 7.38.2 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 100 MAX: 250 EPYC 9575F . MIN: 38 AVG: 380 MAX: 401 OpenFOAM 10 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 322 MAX: 406 EPYC 9575F .............. MIN: 39 AVG: 346 MAX: 400 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 200 MAX: 288 OpenFOAM 10 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 378 MAX: 409 EPYC 9575F .............. MIN: 38 AVG: 388 MAX: 400 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 201 MAX: 292 OpenRadioss 2023.09.15 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 39 AVG: 232 MAX: 389 EPYC 9575F . MIN: 39 AVG: 285 MAX: 399 OpenRadioss 2023.09.15 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 39 AVG: 344 MAX: 407 EPYC 9575F . MIN: 39 AVG: 370 MAX: 400 OpenRadioss 2023.09.15 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 301 MAX: 408 EPYC 9575F . MIN: 39 AVG: 350 MAX: 400 Blender 4.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 252 MAX: 384 EPYC 9575F . MIN: 38 AVG: 299 MAX: 389 Blender 4.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 322 MAX: 390 EPYC 9575F . MIN: 38 AVG: 350 MAX: 389 Blender 4.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 271 MAX: 393 EPYC 9575F . MIN: 38 AVG: 322 MAX: 398 Blender 4.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 320 MAX: 381 EPYC 9575F . MIN: 38 AVG: 356 MAX: 394 Blender 4.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 366 MAX: 403 EPYC 9575F . MIN: 38 AVG: 379 MAX: 400 Blender 4.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 278 MAX: 410 EPYC 9575F . MIN: 38 AVG: 324 MAX: 401 LuxCoreRender 2.6 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 366 MAX: 403 EPYC 9575F . MIN: 38 AVG: 375 MAX: 401 LuxCoreRender 2.6 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 361 MAX: 413 EPYC 9575F . MIN: 38 AVG: 367 MAX: 401 OSPRay 3.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 355 MAX: 404 EPYC 9575F . MIN: 39 AVG: 367 MAX: 401 OSPRay 3.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 363 MAX: 379 EPYC 9575F . MIN: 39 AVG: 382 MAX: 396 OSPRay 3.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 326 MAX: 379 EPYC 9575F . MIN: 38 AVG: 347 MAX: 394 OSPRay Studio 1.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 312 MAX: 391 EPYC 9575F . MIN: 39 AVG: 338 MAX: 401 OSPRay Studio 1.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 335 MAX: 392 EPYC 9575F . MIN: 39 AVG: 334 MAX: 401 OSPRay Studio 1.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 316 MAX: 392 EPYC 9575F . MIN: 38 AVG: 308 MAX: 401 OSPRay Studio 1.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 313 MAX: 391 EPYC 9575F . MIN: 38 AVG: 340 MAX: 401 OSPRay Studio 1.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 302 MAX: 391 EPYC 9575F . MIN: 38 AVG: 342 MAX: 401 OSPRay Studio 1.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 324 MAX: 393 EPYC 9575F . MIN: 38 AVG: 319 MAX: 401 Embree 4.3 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 179 MAX: 398 EPYC 9575F . MIN: 38 AVG: 240 MAX: 401 Embree 4.3 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 141 MAX: 395 EPYC 9575F . MIN: 38 AVG: 203 MAX: 400 Embree 4.3 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 205 MAX: 399 EPYC 9575F . MIN: 38 AVG: 260 MAX: 401 Intel Open Image Denoise 2.2 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 200 MAX: 360 EPYC 9575F . MIN: 38 AVG: 240 MAX: 389 Intel Open Image Denoise 2.2 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 200 MAX: 356 EPYC 9575F . MIN: 38 AVG: 238 MAX: 390 Intel Open Image Denoise 2.2 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 251 MAX: 355 EPYC 9575F . MIN: 38 AVG: 292 MAX: 383 OpenVKL 2.0.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 337 MAX: 409 EPYC 9575F . MIN: 38 AVG: 355 MAX: 401 7-Zip Compression 22.01 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 39 AVG: 312 MAX: 408 EPYC 9575F .............. MIN: 38 AVG: 322 MAX: 401 NVIDIA GH200 144G HBM3e . MIN: 42 AVG: 163 MAX: 246 Parallel BZIP2 Compression 1.1.13 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 81 MAX: 240 EPYC 9575F .............. MIN: 38 AVG: 104 MAX: 361 NVIDIA GH200 144G HBM3e . MIN: 40 AVG: 69 MAX: 208 Zstd Compression 1.5.4 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 92 MAX: 113 EPYC 9575F .............. MIN: 38 AVG: 114 MAX: 156 NVIDIA GH200 144G HBM3e . MIN: 39 AVG: 47 MAX: 55 Zstd Compression 1.5.4 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 87 MAX: 101 EPYC 9575F .............. MIN: 38 AVG: 102 MAX: 129 NVIDIA GH200 144G HBM3e . MIN: 39 AVG: 44 MAX: 49 PyBench 2018-02-16 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 74 MAX: 84 EPYC 9575F .............. MIN: 38 AVG: 81 MAX: 98 NVIDIA GH200 144G HBM3e . MIN: 39 AVG: 42 MAX: 43 Numpy Benchmark CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 82 MAX: 90 EPYC 9575F .............. MIN: 38 AVG: 94 MAX: 101 NVIDIA GH200 144G HBM3e . MIN: 39 AVG: 43 MAX: 48 SVT-AV1 2.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 100 MAX: 236 EPYC 9575F . MIN: 38 AVG: 115 MAX: 298 SVT-AV1 2.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 99 MAX: 241 EPYC 9575F . MIN: 38 AVG: 114 MAX: 323 SVT-AV1 2.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 127 MAX: 238 EPYC 9575F . MIN: 38 AVG: 153 MAX: 299 SVT-AV1 2.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 135 MAX: 221 EPYC 9575F . MIN: 38 AVG: 187 MAX: 334 Kvazaar 2.2 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 243 MAX: 354 EPYC 9575F . MIN: 38 AVG: 286 MAX: 368 Kvazaar 2.2 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 244 MAX: 356 EPYC 9575F . MIN: 38 AVG: 284 MAX: 373 uvg266 0.4.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 279 MAX: 376 EPYC 9575F . MIN: 38 AVG: 320 MAX: 385 uvg266 0.4.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 39 AVG: 274 MAX: 375 EPYC 9575F . MIN: 39 AVG: 313 MAX: 383 uvg266 0.4.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 39 AVG: 207 MAX: 303 EPYC 9575F . MIN: 38 AVG: 260 MAX: 393 FFmpeg 7.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 118 MAX: 149 EPYC 9575F . MIN: 38 AVG: 162 MAX: 226 WebP Image Encode 1.2.4 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 83 MAX: 94 EPYC 9575F .............. MIN: 38 AVG: 95 MAX: 114 NVIDIA GH200 144G HBM3e . MIN: 39 AVG: 44 MAX: 46 libavif avifenc 1.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 148 MAX: 312 EPYC 9575F .............. MIN: 38 AVG: 201 MAX: 400 NVIDIA GH200 144G HBM3e . MIN: 39 AVG: 69 MAX: 237 ASTC Encoder 4.7 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 159 MAX: 364 EPYC 9575F .............. MIN: 38 AVG: 220 MAX: 386 NVIDIA GH200 144G HBM3e . MIN: 39 AVG: 162 MAX: 263 ASTC Encoder 4.7 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 250 MAX: 375 EPYC 9575F .............. MIN: 38 AVG: 314 MAX: 390 NVIDIA GH200 144G HBM3e . MIN: 40 AVG: 219 MAX: 261 ASTC Encoder 4.7 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 253 MAX: 378 EPYC 9575F .............. MIN: 38 AVG: 312 MAX: 391 NVIDIA GH200 144G HBM3e . MIN: 42 AVG: 223 MAX: 267 GraphicsMagick 1.3.43 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 166 MAX: 181 EPYC 9575F .............. MIN: 38 AVG: 225 MAX: 244 NVIDIA GH200 144G HBM3e . MIN: 42 AVG: 166 MAX: 283 GraphicsMagick 1.3.43 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 265 MAX: 285 EPYC 9575F .............. MIN: 38 AVG: 341 MAX: 368 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 239 MAX: 263 GraphicsMagick 1.3.43 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 290 MAX: 312 EPYC 9575F .............. MIN: 39 AVG: 333 MAX: 356 NVIDIA GH200 144G HBM3e . MIN: 43 AVG: 268 MAX: 300 Liquid-DSP 1.6 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 76 MAX: 83 EPYC 9575F .............. MIN: 38 AVG: 84 MAX: 91 NVIDIA GH200 144G HBM3e . MIN: 40 AVG: 44 MAX: 46 Liquid-DSP 1.6 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 77 MAX: 82 EPYC 9575F .............. MIN: 38 AVG: 87 MAX: 94 NVIDIA GH200 144G HBM3e . MIN: 39 AVG: 42 MAX: 43 Liquid-DSP 1.6 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 39 AVG: 343 MAX: 397 EPYC 9575F .............. MIN: 38 AVG: 329 MAX: 377 NVIDIA GH200 144G HBM3e . MIN: 39 AVG: 178 MAX: 203 Liquid-DSP 1.6 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 349 MAX: 397 EPYC 9575F .............. MIN: 38 AVG: 356 MAX: 401 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 257 MAX: 292 Liquid-DSP 1.6 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 39 AVG: 347 MAX: 400 EPYC 9575F .............. MIN: 38 AVG: 345 MAX: 391 NVIDIA GH200 144G HBM3e . MIN: 42 AVG: 232 MAX: 270 Liquid-DSP 1.6 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 39 AVG: 361 MAX: 417 EPYC 9575F .............. MIN: 38 AVG: 331 MAX: 372 NVIDIA GH200 144G HBM3e . MIN: 42 AVG: 181 MAX: 205 srsRAN Project 23.10.1-20240325 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 39 AVG: 312 MAX: 368 EPYC 9575F .............. MIN: 39 AVG: 268 MAX: 362 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 111 MAX: 134 srsRAN Project 23.10.1-20240325 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 74 MAX: 84 EPYC 9575F .............. MIN: 38 AVG: 78 MAX: 95 NVIDIA GH200 144G HBM3e . MIN: 39 AVG: 42 MAX: 43 TensorFlow 2.16.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 286 MAX: 308 EPYC 9575F . MIN: 38 AVG: 304 MAX: 323 TensorFlow 2.16.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 193 MAX: 260 EPYC 9575F . MIN: 38 AVG: 235 MAX: 310 TensorFlow 2.16.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 .. MIN: 38 AVG: 302 MAX: 330 EPYC 9575F . MIN: 38 AVG: 329 MAX: 347 OpenVINO 2024.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 337 MAX: 378 EPYC 9575F .............. MIN: 38 AVG: 348 MAX: 378 NVIDIA GH200 144G HBM3e . MIN: 39 AVG: 161 MAX: 212 OpenVINO 2024.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 39 AVG: 328 MAX: 358 EPYC 9575F .............. MIN: 39 AVG: 308 MAX: 331 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 118 MAX: 127 OpenVINO 2024.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 39 AVG: 358 MAX: 414 EPYC 9575F .............. MIN: 39 AVG: 372 MAX: 401 NVIDIA GH200 144G HBM3e . MIN: 40 AVG: 97 MAX: 136 OpenVINO 2024.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 39 AVG: 333 MAX: 359 EPYC 9575F .............. MIN: 39 AVG: 331 MAX: 355 NVIDIA GH200 144G HBM3e . MIN: 40 AVG: 122 MAX: 132 OpenVINO 2024.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 39 AVG: 373 MAX: 404 EPYC 9575F .............. MIN: 38 AVG: 333 MAX: 357 NVIDIA GH200 144G HBM3e . MIN: 40 AVG: 154 MAX: 182 OpenVINO 2024.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 39 AVG: 350 MAX: 387 EPYC 9575F .............. MIN: 39 AVG: 298 MAX: 322 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 149 MAX: 162 OpenVINO 2024.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 39 AVG: 364 MAX: 410 EPYC 9575F .............. MIN: 38 AVG: 372 MAX: 401 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 160 MAX: 179 OpenVINO 2024.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 39 AVG: 346 MAX: 374 EPYC 9575F .............. MIN: 38 AVG: 335 MAX: 361 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 177 MAX: 194 OpenVINO 2024.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 39 AVG: 334 MAX: 368 EPYC 9575F .............. MIN: 38 AVG: 317 MAX: 342 NVIDIA GH200 144G HBM3e . MIN: 42 AVG: 148 MAX: 176 OpenVINO 2024.0 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 39 AVG: 360 MAX: 403 EPYC 9575F .............. MIN: 38 AVG: 345 MAX: 376 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 187 MAX: 213 Xmrig 6.21 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 322 MAX: 345 EPYC 9575F .............. MIN: 38 AVG: 380 MAX: 401 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 140 MAX: 147 Helsing 1.0-beta CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 39 AVG: 343 MAX: 387 EPYC 9575F .............. MIN: 38 AVG: 347 MAX: 373 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 267 MAX: 291 Stockfish 16.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 39 AVG: 389 MAX: 421 EPYC 9575F .............. MIN: 39 AVG: 371 MAX: 401 NVIDIA GH200 144G HBM3e . MIN: 43 AVG: 261 MAX: 291 Primesieve 12.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 119 MAX: 379 EPYC 9575F .............. MIN: 38 AVG: 146 MAX: 376 NVIDIA GH200 144G HBM3e . MIN: 42 AVG: 111 MAX: 254 Primesieve 12.1 CPU Power Consumption Monitor Watts < Lower Is Better EPYC 9965 ............... MIN: 38 AVG: 302 MAX: 387 EPYC 9575F .............. MIN: 38 AVG: 321 MAX: 393 NVIDIA GH200 144G HBM3e . MIN: 41 AVG: 233 MAX: 272