AMD Ryzen Threadripper 3990X 64-Core testing with a Gigabyte TRX40 AORUS PRO WIFI (F6 BIOS) and AMD Radeon RX 5700 8GB on Pop 22.04 via the Phoronix Test Suite.
TR 7980X DIY Build Processor: AMD Ryzen Threadripper 7980X 64-Cores @ 8.21GHz (64 Cores / 128 Threads), Motherboard: ASUS Pro WS TRX50-SAGE WIFI (0607 BIOS), Chipset: AMD Device 14a4, Memory: 4 x 32 GB DRAM-4800MT/s F5-6400R3239G32GQ, Disk: 1000GB Western Digital WDS100T1X0E-00AFY0, Graphics: NAVI31 20GB (2025/1249MHz), Audio: Realtek ALC1220, Monitor: DELL U2723QE, Network: Aquantia Device 04c0 + Intel Device 125b + MEDIATEK Device 0616
OS: Pop 22.04, Kernel: 6.6.6-76060606-generic (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.4, OpenGL: 4.6 Mesa 23.3.2-1pop0~1704238321~22.04~36f1d0e (LLVM 15.0.7 DRM 3.54), Vulkan: 1.3.267, Compiler: GCC 11.4.0, File-System: ext4, Screen Resolution: 3840x2160
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa108105Python Notes: Python 3.10.12Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of Safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
TR 3990X System76 Thelio Processor: AMD Ryzen Threadripper 3990X 64-Core @ 2.90GHz (64 Cores / 128 Threads)Motherboard: Gigabyte TRX40 AORUS PRO WIFI (F6 BIOS)Chipset: AMD Starship/MatisseMemory: 4 x 32 GB DRAM-3000MT/s CMK64GX4M2D3000C16Disk: Samsung SSD 970 EVO Plus 500GBGraphics: AMD Radeon RX 5700 8GB (1750/875MHz)Audio: AMD Navi 10 HDMI AudioMonitor: DELL P2415QNetwork: Intel I211 + Intel Wi-Fi 6 AX200
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x830107aPython Notes: Python 3.10.12Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_rstack_overflow: Mitigation of Safe RET + 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 STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
TR Comparison Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server OpenGL Vulkan Compiler File-System Screen Resolution TR 7980X DIY Build TR 3990X System76 Thelio AMD Ryzen Threadripper 7980X 64-Cores @ 8.21GHz (64 Cores / 128 Threads) ASUS Pro WS TRX50-SAGE WIFI (0607 BIOS) AMD Device 14a4 4 x 32 GB DRAM-4800MT/s F5-6400R3239G32GQ 1000GB Western Digital WDS100T1X0E-00AFY0 NAVI31 20GB (2025/1249MHz) Realtek ALC1220 DELL U2723QE Aquantia Device 04c0 + Intel Device 125b + MEDIATEK Device 0616 Pop 22.04 6.6.6-76060606-generic (x86_64) GNOME Shell 42.5 X Server 1.21.1.4 4.6 Mesa 23.3.2-1pop0~1704238321~22.04~36f1d0e (LLVM 15.0.7 DRM 3.54) 1.3.267 GCC 11.4.0 ext4 3840x2160 AMD Ryzen Threadripper 3990X 64-Core @ 2.90GHz (64 Cores / 128 Threads) Gigabyte TRX40 AORUS PRO WIFI (F6 BIOS) AMD Starship/Matisse 4 x 32 GB DRAM-3000MT/s CMK64GX4M2D3000C16 Samsung SSD 970 EVO Plus 500GB AMD Radeon RX 5700 8GB (1750/875MHz) AMD Navi 10 HDMI Audio DELL P2415Q Intel I211 + Intel Wi-Fi 6 AX200 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v Processor Details - TR 7980X DIY Build: Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa108105 - TR 3990X System76 Thelio: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x830107a Python Details - Python 3.10.12 Security Details - TR 7980X DIY Build: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of Safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected - TR 3990X System76 Thelio: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_rstack_overflow: Mitigation of Safe RET + 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 STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
TR 7980X DIY Build vs. TR 3990X System76 Thelio Comparison Phoronix Test Suite Baseline +192.5% +192.5% +385% +385% +577.5% +577.5% 81.4% 80.2% 75.1% 72.1% 62.3% 61.9% 60.7% 59.1% 27.1% 11.5% 10.9% 4.9% CPU - 256 - ResNet-50 769.8% CPU - 16 - ResNet-50 503.1% Eigen 467% BLAS 411% W.P.D.F.I - CPU 344.3% F.D.F.I - CPU 337% F.D.F.I - CPU 335% W.P.D.F.I - CPU 334.7% C.S.9.P.Y.P - A.M.S 279% N.T.C.B.b.u.S.S.I - A.M.S 261.2% R.5.S.I - A.M.S 259.5% M.T.E.T.D.F - CPU 258% M.T.E.T.D.F - CPU 257.8% V.D.F.I - CPU 243.5% V.D.F.I - CPU 239.8% H.E.R.F.I - CPU 233% H.E.R.F.I - CPU 232.8% gravity_spheres_volume/dim_512/scivis/real_time 232% gravity_spheres_volume/dim_512/ao/real_time 222.6% B.L.N.Q.A.S.I - A.M.S 214.6% F.D.R.F.I - CPU 208.4% CPU - 1 - Efficientnet_v2_l 196.5% CPU - 1 - ResNet-50 195.3% CPU - 64 - ResNet-50 193.5% R.S.A.F.I - CPU 192.6% R.S.A.F.I - CPU 192% A.G.R.R.0.F.I - CPU 185% P.D.F - CPU 184.5% P.D.F - CPU 182.4% CPU - 64 - ResNet-152 181% CPU - 64 - Efficientnet_v2_l 179% P.V.B.D.F - CPU 178.7% P.V.B.D.F - CPU 175.6% CPU - 1 - ResNet-152 168.3% A.G.R.R.0.F.I - CPU 164.2% Compression Rating 154.8% Pathtracer ISPC - Asian Dragon 151.4% C.D.Y.C - A.M.S 151.1% C.D.Y.C.S.I - A.M.S 148.5% R.5.B - A.M.S 147.2% C.C.R.5.I - A.M.S 146.5% gravity_spheres_volume/dim_512/pathtracer/real_time 138% N.D.C.o.b.u.o.I - A.M.S 131.4% llama-2-13b.Q4_0.gguf 130.1% Chrysler Neon 1M 129.8% N.T.C.B.b.u.c - A.M.S 126.9% B.L.N.Q.A - A.M.S 122.4% d.M.M.S - Execution Time 120.7% N.T.C.D.m - A.M.S 120.6% llama-2-7b.Q4_0.gguf 119.9% llama-2-70b-chat.Q5_0.gguf 109.7% Preset 8 - Bosphorus 4K 107.7% C.P.D.T 106.4% d.S.M.S - Execution Time 101.1% Preset 4 - Bosphorus 4K 97.8% Pathtracer ISPC - Crown 95.1% 92.4% Barbershop - CPU-Only 85.2% N.T.C.D.m - A.M.S CPU 81.2% B.L.N.Q.A - A.M.S I.a.F.S.I.D.C 77.7% d.M.M.S - Mesh Time 76.6% R.O.R.S.I 76% N.T.C.B.b.u.c - A.M.S Classroom - CPU-Only 75.1% N.D.C.o.b.u.o.I - A.M.S Pabellon Barcelona - CPU-Only 67% Ninja 66.9% Time To Compile 66% BMW27 - CPU-Only 64.8% Time To Compile 64.5% Time To Compile 63.1% C.C.R.5.I - A.M.S R.5.B - A.M.S C.D.Y.C.S.I - A.M.S F.D.R.F.I - CPU 59.6% C.D.Y.C - A.M.S particle_volume/pathtracer/real_time 58.4% Fishy Cat - CPU-Only 57.8% Time To Compile 57.5% particle_volume/scivis/real_time 56% Preset 13 - Bosphorus 4K 55.5% CPU - Supercar 55.4% allmodconfig 54.9% Multi-Threaded 54.1% particle_volume/ao/real_time 53.8% CPU - Bedroom 53.3% defconfig 53.1% CPU - 1 - ResNet-50 52.4% Preset 12 - Bosphorus 4K 52.1% B.S.o.W 45.7% d.S.M.S - Mesh Time 45.1% D.R 45% Bumper Beam 37.8% ggml-medium.en - 2.S.o.t.U 36% ggml-small.en - 2.S.o.t.U 30.1% ggml-base.en - 2.S.o.t.U 29.2% B.L.N.Q.A.S.I - A.M.S R.5.S.I - A.M.S N.T.C.B.b.u.S.S.I - A.M.S C.S.9.P.Y.P - A.M.S TensorFlow TensorFlow LeelaChessZero LeelaChessZero OpenVINO OpenVINO OpenVINO OpenVINO Neural Magic DeepSparse Neural Magic DeepSparse Neural Magic DeepSparse OpenVINO OpenVINO OpenVINO OpenVINO OpenVINO OpenVINO OSPRay OSPRay Neural Magic DeepSparse OpenVINO PyTorch PyTorch PyTorch OpenVINO OpenVINO OpenVINO OpenVINO OpenVINO PyTorch PyTorch OpenVINO OpenVINO PyTorch OpenVINO 7-Zip Compression Embree Neural Magic DeepSparse Neural Magic DeepSparse Neural Magic DeepSparse Neural Magic DeepSparse OSPRay Neural Magic DeepSparse Llama.cpp OpenRadioss Neural Magic DeepSparse Neural Magic DeepSparse OpenFOAM Neural Magic DeepSparse Llama.cpp Llama.cpp SVT-AV1 OpenRadioss OpenFOAM SVT-AV1 Embree Numpy Benchmark Blender Neural Magic DeepSparse Chaos Group V-RAY Neural Magic DeepSparse OpenRadioss OpenFOAM OpenRadioss Neural Magic DeepSparse Blender Neural Magic DeepSparse Blender Timed LLVM Compilation Timed FFmpeg Compilation Blender Timed Godot Game Engine Compilation Timed Node.js Compilation Neural Magic DeepSparse Neural Magic DeepSparse Neural Magic DeepSparse OpenVINO Neural Magic DeepSparse OSPRay Blender Timed Gem5 Compilation OSPRay SVT-AV1 IndigoBench Timed Linux Kernel Compilation QuantLib OSPRay IndigoBench Timed Linux Kernel Compilation TensorFlow SVT-AV1 OpenRadioss OpenFOAM 7-Zip Compression OpenRadioss Whisper.cpp Whisper.cpp Whisper.cpp Neural Magic DeepSparse Neural Magic DeepSparse Neural Magic DeepSparse Neural Magic DeepSparse TR 7980X DIY Build TR 3990X System76 Thelio
TR Comparison openvino: Face Detection FP16-INT8 - CPU openvino: Face Detection FP16-INT8 - CPU openvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPU openvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPU openvino: Person Detection FP16 - CPU openvino: Person Detection FP16 - CPU openvino: Weld Porosity Detection FP16-INT8 - CPU openvino: Weld Porosity Detection FP16-INT8 - CPU openvino: Vehicle Detection FP16-INT8 - CPU openvino: Vehicle Detection FP16-INT8 - CPU openvino: Person Vehicle Bike Detection FP16 - CPU openvino: Person Vehicle Bike Detection FP16 - CPU openvino: Machine Translation EN To DE FP16 - CPU openvino: Machine Translation EN To DE FP16 - CPU openvino: Face Detection Retail FP16-INT8 - CPU openvino: Face Detection Retail FP16-INT8 - CPU openvino: Handwritten English Recognition FP16-INT8 - CPU openvino: Handwritten English Recognition FP16-INT8 - CPU openvino: Road Segmentation ADAS FP16-INT8 - CPU openvino: Road Segmentation ADAS FP16-INT8 - CPU tensorflow: CPU - 1 - ResNet-50 tensorflow: CPU - 16 - ResNet-50 tensorflow: CPU - 256 - ResNet-50 pytorch: CPU - 1 - ResNet-50 pytorch: CPU - 1 - ResNet-152 pytorch: CPU - 1 - Efficientnet_v2_l pytorch: CPU - 64 - ResNet-50 pytorch: CPU - 64 - ResNet-152 pytorch: CPU - 64 - Efficientnet_v2_l whisper-cpp: ggml-base.en - 2016 State of the Union whisper-cpp: ggml-small.en - 2016 State of the Union whisper-cpp: ggml-medium.en - 2016 State of the Union llama-cpp: llama-2-7b.Q4_0.gguf llama-cpp: llama-2-13b.Q4_0.gguf llama-cpp: llama-2-70b-chat.Q5_0.gguf lczero: BLAS lczero: Eigen deepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Stream deepsparse: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Asynchronous Multi-Stream deepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream deepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Stream deepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream deepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Stream deepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream deepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream deepsparse: CV Detection, YOLOv5s COCO - Asynchronous Multi-Stream deepsparse: CV Detection, YOLOv5s COCO - Asynchronous Multi-Stream deepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Stream deepsparse: CV Detection, YOLOv5s COCO, Sparse INT8 - Asynchronous Multi-Stream deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Stream deepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Stream deepsparse: CV Segmentation, 90% Pruned YOLACT Pruned - Asynchronous Multi-Stream deepsparse: ResNet-50, Baseline - Asynchronous Multi-Stream deepsparse: ResNet-50, Baseline - Asynchronous Multi-Stream deepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Stream deepsparse: ResNet-50, Sparse INT8 - Asynchronous Multi-Stream deepsparse: BERT-Large, NLP Question Answering - Asynchronous Multi-Stream deepsparse: BERT-Large, NLP Question Answering - Asynchronous Multi-Stream deepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Stream deepsparse: BERT-Large, NLP Question Answering, Sparse INT8 - Asynchronous Multi-Stream numpy: build-linux-kernel: defconfig build-linux-kernel: allmodconfig build-godot: Time To Compile build-ffmpeg: Time To Compile build-llvm: Ninja build-nodejs: Time To Compile build-gem5: Time To Compile blender: BMW27 - CPU-Only blender: Classroom - CPU-Only blender: Fishy Cat - CPU-Only blender: Pabellon Barcelona - CPU-Only blender: Barbershop - CPU-Only ospray: gravity_spheres_volume/dim_512/ao/real_time ospray: gravity_spheres_volume/dim_512/scivis/real_time ospray: gravity_spheres_volume/dim_512/pathtracer/real_time ospray: particle_volume/ao/real_time ospray: particle_volume/scivis/real_time ospray: particle_volume/pathtracer/real_time embree: Pathtracer ISPC - Asian Dragon embree: Pathtracer ISPC - Crown v-ray: CPU indigobench: CPU - Supercar indigobench: CPU - Bedroom compress-7zip: Compression Rating compress-7zip: Decompression Rating openfoam: drivaerFastback, Small Mesh Size - Mesh Time openfoam: drivaerFastback, Small Mesh Size - Execution Time openfoam: drivaerFastback, Medium Mesh Size - Mesh Time openfoam: drivaerFastback, Medium Mesh Size - Execution Time openradioss: Bird Strike on Windshield openradioss: Rubber O-Ring Seal Installation openradioss: Cell Phone Drop Test openradioss: Bumper Beam openradioss: INIVOL and Fluid Structure Interaction Drop Container openradioss: Chrysler Neon 1M quantlib: Multi-Threaded svt-av1: Preset 13 - Bosphorus 4K svt-av1: Preset 12 - Bosphorus 4K svt-av1: Preset 8 - Bosphorus 4K svt-av1: Preset 4 - Bosphorus 4K TR 7980X DIY Build TR 3990X System76 Thelio 78.74 405.37 112465.33 0.40 220.46 145.95 7702.02 8.09 4727.16 6.67 4215.15 7.48 429.11 74.42 13253.18 4.60 1711.04 37.31 1601.38 19.92 7.62 59.41 105.85 59.79 22.08 12.87 48.39 19.05 8.23 191.99942 445.32330 925.98065 30.78 18.43 3.04 557 533 905.4136 8.8140 190.4801 41.9629 280.0001 28.5428 22.2065 359.8588 130.5869 61.2123 133.5394 59.8761 22.2004 359.8583 41.9230 190.6468 280.0274 28.5402 2372.1682 3.3518 27.6721 288.5758 405.4849 19.7033 753.68 23.413 216.690 79.844 11.671 122.225 113.780 178.794 18.47 47.15 23.80 57.94 166.88 23.4257 22.6446 26.5062 22.5275 22.5832 239.319 103.9138 93.4593 88574 34.446 15.704 443349 529231 19.553856 36.85956 104.60125 617.2658 105.42 53.18 23.41 75.61 115.12 292.06 245683.2 194.101 198.740 90.450 9.246 18.02 1763.38 42563.23 1.14 77.50 412.09 1771.60 35.94 1391.09 22.91 1529.19 20.85 119.93 266.39 4296.88 7.34 514.20 124.25 548.40 58.28 5.00 9.85 12.17 20.25 8.23 4.34 16.49 6.78 2.95 247.99593 579.20672 1259.40500 14.00 8.01 1.45 109 94 1003.6586 31.8335 345.4413 92.5781 454.4448 70.3655 38.8841 816.6532 207.7083 153.6873 214.5970 148.8038 38.2162 832.6156 43.9894 722.5058 453.2638 70.5496 2645.3102 12.0505 49.8527 641.7697 515.5297 61.9787 391.71 35.857 335.736 131.357 19.377 204.001 185.538 281.538 30.43 82.54 37.56 96.76 309.06 7.26101 6.82030 11.1377 14.6492 14.4799 151.050 41.3331 47.9019 48871 22.162 10.242 173970 364981 28.370559 74.108433 184.71201 1362.3583 153.62 93.61 48.31 104.16 204.54 671.04 159470.1 124.791 130.631 43.545 4.675 OpenBenchmarking.org
TensorFlow This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.
Whisper.cpp Whisper.cpp is a port of OpenAI's Whisper model in C/C++. Whisper.cpp is developed by Georgi Gerganov for transcribing WAV audio files to text / speech recognition. Whisper.cpp supports ARM NEON, x86 AVX, and other advanced CPU features. Learn more via the OpenBenchmarking.org test page.
Llama.cpp Llama.cpp is a port of Facebook's LLaMA model in C/C++ developed by Georgi Gerganov. Llama.cpp allows the inference of LLaMA and other supported models in C/C++. For CPU inference Llama.cpp supports AVX2/AVX-512, ARM NEON, and other modern ISAs along with features like OpenBLAS usage. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Stream TR 7980X DIY Build TR 3990X System76 Thelio 7 14 21 28 35 SE +/- 0.0108, N = 3 SE +/- 0.0689, N = 3 8.8140 31.8335 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max TR 7980X DIY Build 545 4523 5508 TR 3990X System76 Thelio 2783 3163 4212 OpenBenchmarking.org Megahertz, More Is Better Neural Magic DeepSparse 1.6 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max TR 7980X DIY Build 41.7 256.6 316.6 TR 3990X System76 Thelio 54.4 228.2 289.8 OpenBenchmarking.org Watts, Fewer Is Better Neural Magic DeepSparse 1.6 CPU Power Consumption Monitor 80 160 240 320 400 1. TR 7980X DIY Build: Approximate power consumption of 11972 Joules per run. 2. TR 3990X System76 Thelio: Approximate power consumption of 11788 Joules per run. CPU Temp
Min Avg Max TR 7980X DIY Build 46.0 50.5 52.4 TR 3990X System76 Thelio 46.3 60.6 67.0 OpenBenchmarking.org Celsius, Fewer Is Better Neural Magic DeepSparse 1.6 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: NLP Text Classification, BERT base uncased SST2, Sparse INT8 - Scenario: Asynchronous Multi-Stream TR 7980X DIY Build TR 3990X System76 Thelio 7 14 21 28 35 Min: 8.8 / Avg: 8.81 / Max: 8.84 Min: 31.71 / Avg: 31.83 / Max: 31.94
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Stream TR 7980X DIY Build TR 3990X System76 Thelio 30 60 90 120 150 SE +/- 0.05, N = 3 SE +/- 0.87, N = 3 59.88 148.80 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max TR 7980X DIY Build 545 4417 4861 TR 3990X System76 Thelio 2669 2948 4208 OpenBenchmarking.org Megahertz, More Is Better Neural Magic DeepSparse 1.6 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1200 2400 3600 4800 6000 CPU Power Consumption
Min Avg Max TR 7980X DIY Build 53.9 250.0 288.6 TR 3990X System76 Thelio 54.4 250.6 290.9 OpenBenchmarking.org Watts, Fewer Is Better Neural Magic DeepSparse 1.6 CPU Power Consumption Monitor 70 140 210 280 350 1. TR 7980X DIY Build: Approximate power consumption of 10251 Joules per run. 2. TR 3990X System76 Thelio: Approximate power consumption of 10777 Joules per run. CPU Temp
Min Avg Max TR 7980X DIY Build 41.4 48.2 50.0 TR 3990X System76 Thelio 49.8 64.3 68.8 OpenBenchmarking.org Celsius, Fewer Is Better Neural Magic DeepSparse 1.6 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: CV Detection, YOLOv5s COCO, Sparse INT8 - Scenario: Asynchronous Multi-Stream TR 7980X DIY Build TR 3990X System76 Thelio 30 60 90 120 150 Min: 59.82 / Avg: 59.88 / Max: 59.98 Min: 147.7 / Avg: 148.8 / Max: 150.53
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream TR 7980X DIY Build TR 3990X System76 Thelio 14 28 42 56 70 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 19.70 61.98 CPU Peak Freq (Highest CPU Core Frequency
Min Avg Max TR 7980X DIY Build 545 4537 5585 TR 3990X System76 Thelio 2818 3144 4237 OpenBenchmarking.org Megahertz, More Is Better Neural Magic DeepSparse 1.6 CPU Peak Freq (Highest CPU Core Frequency) Monitor 1400 2800 4200 5600 7000 CPU Power Consumption
Min Avg Max TR 7980X DIY Build 79.3 234.5 281.7 TR 3990X System76 Thelio 54.0 231.0 290.0 OpenBenchmarking.org Watts, Fewer Is Better Neural Magic DeepSparse 1.6 CPU Power Consumption Monitor 70 140 210 280 350 1. TR 7980X DIY Build: Approximate power consumption of 10787 Joules per run. 2. TR 3990X System76 Thelio: Approximate power consumption of 11705 Joules per run. CPU Temp
Min Avg Max TR 7980X DIY Build 41.4 47.9 50.0 TR 3990X System76 Thelio 48.5 61.8 67.8 OpenBenchmarking.org Celsius, Fewer Is Better Neural Magic DeepSparse 1.6 CPU Temperature Monitor 20 40 60 80 100 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.6 Model: BERT-Large, NLP Question Answering, Sparse INT8 - Scenario: Asynchronous Multi-Stream TR 7980X DIY Build TR 3990X System76 Thelio 12 24 36 48 60 Min: 19.69 / Avg: 19.7 / Max: 19.71 Min: 61.92 / Avg: 61.98 / Max: 62.01
Timed Linux Kernel Compilation This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.
Timed Mesa Compilation This test profile times how long it takes to compile Mesa with Meson/Ninja. For minimizing build dependencies and avoid versioning conflicts, test this is just the core Mesa build without LLVM or the extra Gallium3D/Mesa drivers enabled. Learn more via the OpenBenchmarking.org test page.
Time To Compile
TR 7980X DIY Build: The test quit with a non-zero exit status. E: ninja: error: loading 'build.ninja': No such file or directory
TR 3990X System76 Thelio: The test quit with a non-zero exit status. E: ninja: error: loading 'build.ninja': No such file or directory
Timed Godot Game Engine Compilation This test times how long it takes to compile the Godot Game Engine. Godot is a popular, open-source, cross-platform 2D/3D game engine and is built using the SCons build system and targeting the X11 platform. Learn more via the OpenBenchmarking.org test page.
Timed Node.js Compilation This test profile times how long it takes to build/compile Node.js itself from source. Node.js is a JavaScript run-time built from the Chrome V8 JavaScript engine while itself is written in C/C++. Learn more via the OpenBenchmarking.org test page.
Timed Gem5 Compilation This test times how long it takes to compile Gem5. Gem5 is a simulator for computer system architecture research. Gem5 is widely used for computer architecture research within the industry, academia, and more. Learn more via the OpenBenchmarking.org test page.
Blender Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.
OSPRay Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Embree Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs (and GPUs via SYCL) and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.
Chaos Group V-RAY This is a test of Chaos Group's V-RAY benchmark. V-RAY is a commercial renderer that can integrate with various creator software products like SketchUp and 3ds Max. The V-RAY benchmark is standalone and supports CPU and NVIDIA CUDA/RTX based rendering. Learn more via the OpenBenchmarking.org test page.
OpenFOAM OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 10 Input: drivaerFastback, Small Mesh Size - Mesh Time TR 7980X DIY Build TR 3990X System76 Thelio 7 14 21 28 35 19.55 28.37 1. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm
OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 10 Input: drivaerFastback, Medium Mesh Size - Mesh Time TR 7980X DIY Build TR 3990X System76 Thelio 40 80 120 160 200 104.60 184.71 1. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm
OpenRadioss OpenRadioss is an open-source AGPL-licensed finite element solver for dynamic event analysis OpenRadioss is based on Altair Radioss and open-sourced in 2022. This open-source finite element solver is benchmarked with various example models available from https://www.openradioss.org/models/ and https://github.com/OpenRadioss/ModelExchange/tree/main/Examples. This test is currently using a reference OpenRadioss binary build offered via GitHub. Learn more via the OpenBenchmarking.org test page.
QuantLib QuantLib is an open-source library/framework around quantitative finance for modeling, trading and risk management scenarios. QuantLib is written in C++ with Boost and its built-in benchmark used reports the QuantLib Benchmark Index benchmark score. Learn more via the OpenBenchmarking.org test page.
SVT-AV1 This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.
Meta Performance Per Watts OpenBenchmarking.org Performance Per Watts, More Is Better Meta Performance Per Watts Performance Per Watts TR 7980X DIY Build TR 3990X System76 Thelio 20 40 60 80 100 99.29 41.70
CPU Peak Freq (Highest CPU Core Frequency) Monitor OpenBenchmarking.org Megahertz CPU Peak Freq (Highest CPU Core Frequency) Monitor Phoronix Test Suite System Monitoring TR 7980X DIY Build TR 3990X System76 Thelio 1000 2000 3000 4000 5000 Min: 545 / Avg: 4217.07 / Max: 5617 Min: 2196 / Avg: 3527.76 / Max: 4330
CPU Power Consumption Monitor OpenBenchmarking.org Watts CPU Power Consumption Monitor Phoronix Test Suite System Monitoring TR 7980X DIY Build TR 3990X System76 Thelio 70 140 210 280 350 Min: 39.88 / Avg: 305.94 / Max: 372.82 Min: 27.29 / Avg: 256.14 / Max: 305.51
CPU Temperature Monitor OpenBenchmarking.org Celsius CPU Temperature Monitor Phoronix Test Suite System Monitoring TR 7980X DIY Build TR 3990X System76 Thelio 20 40 60 80 100 Min: 36.88 / Avg: 64.84 / Max: 94.75 Min: 42.25 / Avg: 65.24 / Max: 89
TR 7980X DIY Build Processor: AMD Ryzen Threadripper 7980X 64-Cores @ 8.21GHz (64 Cores / 128 Threads), Motherboard: ASUS Pro WS TRX50-SAGE WIFI (0607 BIOS), Chipset: AMD Device 14a4, Memory: 4 x 32 GB DRAM-4800MT/s F5-6400R3239G32GQ, Disk: 1000GB Western Digital WDS100T1X0E-00AFY0, Graphics: NAVI31 20GB (2025/1249MHz), Audio: Realtek ALC1220, Monitor: DELL U2723QE, Network: Aquantia Device 04c0 + Intel Device 125b + MEDIATEK Device 0616
OS: Pop 22.04, Kernel: 6.6.6-76060606-generic (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.4, OpenGL: 4.6 Mesa 23.3.2-1pop0~1704238321~22.04~36f1d0e (LLVM 15.0.7 DRM 3.54), Vulkan: 1.3.267, Compiler: GCC 11.4.0, File-System: ext4, Screen Resolution: 3840x2160
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa108105Python Notes: Python 3.10.12Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_rstack_overflow: Mitigation of Safe RET + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS IBPB: conditional STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 12 January 2024 19:18 by user phoronix.
TR 3990X System76 Thelio Processor: AMD Ryzen Threadripper 3990X 64-Core @ 2.90GHz (64 Cores / 128 Threads), Motherboard: Gigabyte TRX40 AORUS PRO WIFI (F6 BIOS), Chipset: AMD Starship/Matisse, Memory: 4 x 32 GB DRAM-3000MT/s CMK64GX4M2D3000C16, Disk: Samsung SSD 970 EVO Plus 500GB, Graphics: AMD Radeon RX 5700 8GB (1750/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: DELL P2415Q, Network: Intel I211 + Intel Wi-Fi 6 AX200
OS: Pop 22.04, Kernel: 6.6.6-76060606-generic (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.4, OpenGL: 4.6 Mesa 23.3.2-1pop0~1704238321~22.04~36f1d0e (LLVM 15.0.7 DRM 3.54), Vulkan: 1.3.267, Compiler: GCC 11.4.0, File-System: ext4, Screen Resolution: 3840x2160
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-XeT9lY/gcc-11-11.4.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vProcessor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0x830107aPython Notes: Python 3.10.12Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Mitigation of untrained return thunk; SMT enabled with STIBP protection + spec_rstack_overflow: Mitigation of Safe RET + 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 STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 13 January 2024 10:23 by user phoronix.