tensorflow ryzen zen 4 Tests for a future article. AMD Ryzen 9 7950X3D 16-Core testing with a ASUS ROG CROSSHAIR X670E HERO (9927 BIOS) and AMD Radeon RX 7900 XTX on Ubuntu 23.04 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2303294-PTS-TENSORFL34&sor&grr .
tensorflow ryzen zen 4 Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server OpenGL OpenCL Compiler File-System Screen Resolution 7600x a 7600x b 7950x3d a 7950x3d b AMD Ryzen 5 7600X 6-Core @ 4.70GHz (6 Cores / 12 Threads) ASUS ROG CROSSHAIR X670E HERO (9927 BIOS) AMD Device 14d8 32GB Western Digital WD_BLACK SN850X 1000GB + 2000GB AMD Radeon RX 7900 XTX (2304/1249MHz) AMD Device ab30 ASUS MG28U Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411 Ubuntu 23.04 6.2.8-060208-generic (x86_64) GNOME Shell 44.0 X Server 1.21.1.7 + Wayland 4.6 Mesa 23.1.0-devel (git-de8b14f 2023-03-24 lunar-oibaf-ppa) (LLVM 15.0.7 DRM 3.49) OpenCL 2.1 AMD-APP (3513.0) GCC 12.2.0 ext4 3840x2160 AMD Ryzen 9 7950X3D 16-Core @ 4.20GHz (16 Cores / 32 Threads) OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Processor Details - Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa601203 Python Details - Python 3.11.2 Security Details - itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
tensorflow ryzen zen 4 blender: Barbershop - CPU-Only tensorflow: CPU - 256 - ResNet-50 tensorflow: CPU - 512 - GoogLeNet tensorflow: CPU - 256 - GoogLeNet tensorflow: CPU - 512 - AlexNet tensorflow: CPU - 64 - ResNet-50 blender: Pabellon Barcelona - CPU-Only blender: Classroom - CPU-Only tensorflow: CPU - 256 - AlexNet tensorflow: CPU - 32 - ResNet-50 tensorflow: CPU - 64 - GoogLeNet blender: Fishy Cat - CPU-Only blender: BMW27 - CPU-Only tensorflow: CPU - 16 - ResNet-50 tensorflow: CPU - 32 - GoogLeNet tensorflow: CPU - 64 - AlexNet tensorflow: CPU - 32 - AlexNet tensorflow: CPU - 16 - GoogLeNet tensorflow: CPU - 16 - AlexNet tensorflow: CPU - 512 - ResNet-50 7600x a 7600x b 7950x3d a 7950x3d b 1202.85 27.53 79.77 79.54 188.7 27.63 411.84 333.15 186.23 27.81 80.05 163.19 127.87 27.86 80.73 166.15 145.54 82.08 114.42 1202.67 27.54 79.77 79.61 190.30 27.66 410.66 333.68 186.31 27.80 79.92 162.92 128.26 27.85 80.76 164.76 145.45 81.95 114.73 480.88 43.85 136.9 137.95 430.57 44.16 167.38 136 419.28 44.34 145.1 66.56 52.1 43.14 149.5 340.13 259.68 146.69 176 480.34 43.9 136.97 137.75 430.58 44.24 166.53 135.8 418.85 44.32 145.19 66.16 51.96 43.24 149.66 339.62 258.45 146.43 176.37 OpenBenchmarking.org
Blender Blend File: Barbershop - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.5 Blend File: Barbershop - Compute: CPU-Only 7950x3d b 7950x3d a 7600x b 7600x a 300 600 900 1200 1500 480.34 480.88 1202.67 1202.85
TensorFlow Device: CPU - Batch Size: 256 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 256 - Model: ResNet-50 7950x3d b 7950x3d a 7600x b 7600x a 10 20 30 40 50 43.90 43.85 27.54 27.53
TensorFlow Device: CPU - Batch Size: 512 - Model: GoogLeNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 512 - Model: GoogLeNet 7950x3d b 7950x3d a 7600x b 7600x a 30 60 90 120 150 136.97 136.90 79.77 79.77
TensorFlow Device: CPU - Batch Size: 256 - Model: GoogLeNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 256 - Model: GoogLeNet 7950x3d a 7950x3d b 7600x b 7600x a 30 60 90 120 150 SE +/- 0.05, N = 3 137.95 137.75 79.61 79.54
TensorFlow Device: CPU - Batch Size: 512 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 512 - Model: AlexNet 7950x3d b 7950x3d a 7600x b 7600x a 90 180 270 360 450 SE +/- 1.32, N = 3 430.58 430.57 190.30 188.70
TensorFlow Device: CPU - Batch Size: 64 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 64 - Model: ResNet-50 7950x3d b 7950x3d a 7600x b 7600x a 10 20 30 40 50 SE +/- 0.00, N = 3 44.24 44.16 27.66 27.63
Blender Blend File: Pabellon Barcelona - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.5 Blend File: Pabellon Barcelona - Compute: CPU-Only 7950x3d b 7950x3d a 7600x b 7600x a 90 180 270 360 450 166.53 167.38 410.66 411.84
Blender Blend File: Classroom - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.5 Blend File: Classroom - Compute: CPU-Only 7950x3d b 7950x3d a 7600x a 7600x b 70 140 210 280 350 135.80 136.00 333.15 333.68
TensorFlow Device: CPU - Batch Size: 256 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 256 - Model: AlexNet 7950x3d a 7950x3d b 7600x b 7600x a 90 180 270 360 450 SE +/- 0.42, N = 3 419.28 418.85 186.31 186.23
TensorFlow Device: CPU - Batch Size: 32 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 32 - Model: ResNet-50 7950x3d a 7950x3d b 7600x a 7600x b 10 20 30 40 50 SE +/- 0.01, N = 3 44.34 44.32 27.81 27.80
TensorFlow Device: CPU - Batch Size: 64 - Model: GoogLeNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 64 - Model: GoogLeNet 7950x3d b 7950x3d a 7600x a 7600x b 30 60 90 120 150 SE +/- 0.04, N = 3 145.19 145.10 80.05 79.92
Blender Blend File: Fishy Cat - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.5 Blend File: Fishy Cat - Compute: CPU-Only 7950x3d b 7950x3d a 7600x b 7600x a 40 80 120 160 200 66.16 66.56 162.92 163.19
Blender Blend File: BMW27 - Compute: CPU-Only OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.5 Blend File: BMW27 - Compute: CPU-Only 7950x3d b 7950x3d a 7600x a 7600x b 30 60 90 120 150 51.96 52.10 127.87 128.26
TensorFlow Device: CPU - Batch Size: 16 - Model: ResNet-50 OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 16 - Model: ResNet-50 7950x3d b 7950x3d a 7600x a 7600x b 10 20 30 40 50 SE +/- 0.01, N = 3 43.24 43.14 27.86 27.85
TensorFlow Device: CPU - Batch Size: 32 - Model: GoogLeNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 32 - Model: GoogLeNet 7950x3d b 7950x3d a 7600x b 7600x a 30 60 90 120 150 SE +/- 0.02, N = 3 149.66 149.50 80.76 80.73
TensorFlow Device: CPU - Batch Size: 64 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 64 - Model: AlexNet 7950x3d a 7950x3d b 7600x a 7600x b 70 140 210 280 350 SE +/- 0.14, N = 3 340.13 339.62 166.15 164.76
TensorFlow Device: CPU - Batch Size: 32 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 32 - Model: AlexNet 7950x3d a 7950x3d b 7600x a 7600x b 60 120 180 240 300 SE +/- 0.21, N = 3 259.68 258.45 145.54 145.45
TensorFlow Device: CPU - Batch Size: 16 - Model: GoogLeNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 16 - Model: GoogLeNet 7950x3d a 7950x3d b 7600x a 7600x b 30 60 90 120 150 SE +/- 0.08, N = 3 146.69 146.43 82.08 81.95
TensorFlow Device: CPU - Batch Size: 16 - Model: AlexNet OpenBenchmarking.org images/sec, More Is Better TensorFlow 2.12 Device: CPU - Batch Size: 16 - Model: AlexNet 7950x3d b 7950x3d a 7600x b 7600x a 40 80 120 160 200 SE +/- 0.04, N = 3 176.37 176.00 114.73 114.42
Phoronix Test Suite v10.8.5