lc0 threadripper AMD Ryzen Threadripper 7980X 64-Cores testing with a System76 Thelio Major (FA Z5 BIOS) and AMD Radeon Pro W7900 45GB on Pop 24.04 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2408113-PTS-LC0THREA38&gru&sor .
lc0 threadripper Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server OpenGL Compiler File-System Screen Resolution a b c d AMD Ryzen Threadripper 7980X 64-Cores @ 5.37GHz (64 Cores / 128 Threads) System76 Thelio Major (FA Z5 BIOS) AMD Device 14a4 4 x 32GB DDR5-4800MT/s Micron MTC20F1045S1RC48BA2 1000GB CT1000T700SSD5 + 257GB Flash Drive AMD Radeon Pro W7900 45GB AMD Device 14cc DELL P2415Q Aquantia AQC113C NBase-T/IEEE + Realtek RTL8125 2.5GbE + Intel Wi-Fi 6E Pop 24.04 6.9.3-76060903-generic (x86_64) COSMIC 0.1.0 Wayland 4.6 Mesa 24.0.9-0ubuntu0.1 (LLVM 17.0.6 DRM 3.57) GCC 13.2.0 ext4 1920x1080 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler Details - --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-13-uJ7kn6/gcc-13-13.2.0/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-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 - Scaling Governor: amd-pstate-epp powersave (EPP: balance_performance) - CPU Microcode: 0xa108105 Security Details - gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: 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; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected
lc0 threadripper simdjson: Kostya simdjson: TopTweet simdjson: LargeRand simdjson: PartialTweets simdjson: DistinctUserID lczero: Eigen xnnpack: FP32MobileNetV2 xnnpack: FP32MobileNetV3Large xnnpack: FP32MobileNetV3Small xnnpack: FP16MobileNetV2 xnnpack: FP16MobileNetV3Large xnnpack: FP16MobileNetV3Small xnnpack: QU8MobileNetV2 xnnpack: QU8MobileNetV3Large xnnpack: QU8MobileNetV3Small a b c d 5.74 10.38 1.75 9.46 10.25 124 4446 6435 4227 4113 5906 4113 3910 6216 4707 5.8 10.49 1.72 9.54 10.38 126 4485 6333 4189 4146 5939 4006 3874 6297 4815 5.68 10.42 1.73 9.59 10.05 135 4326 6369 4262 4161 6175 4106 4015 6216 4917 5.73 10.42 1.73 9.42 10.59 139 4232 6345 4245 4169 5936 4128 3857 6285 4750 OpenBenchmarking.org
simdjson Throughput Test: Kostya OpenBenchmarking.org GB/s, More Is Better simdjson 3.10 Throughput Test: Kostya b a d c 1.305 2.61 3.915 5.22 6.525 SE +/- 0.02, N = 3 5.80 5.74 5.73 5.68 1. (CXX) g++ options: -O3 -lrt
simdjson Throughput Test: TopTweet OpenBenchmarking.org GB/s, More Is Better simdjson 3.10 Throughput Test: TopTweet b d c a 3 6 9 12 15 SE +/- 0.06, N = 3 10.49 10.42 10.42 10.38 1. (CXX) g++ options: -O3 -lrt
simdjson Throughput Test: LargeRandom OpenBenchmarking.org GB/s, More Is Better simdjson 3.10 Throughput Test: LargeRandom a d c b 0.3938 0.7876 1.1814 1.5752 1.969 SE +/- 0.02, N = 3 1.75 1.73 1.73 1.72 1. (CXX) g++ options: -O3 -lrt
simdjson Throughput Test: PartialTweets OpenBenchmarking.org GB/s, More Is Better simdjson 3.10 Throughput Test: PartialTweets c b a d 3 6 9 12 15 SE +/- 0.06, N = 3 9.59 9.54 9.46 9.42 1. (CXX) g++ options: -O3 -lrt
simdjson Throughput Test: DistinctUserID OpenBenchmarking.org GB/s, More Is Better simdjson 3.10 Throughput Test: DistinctUserID d b a c 3 6 9 12 15 SE +/- 0.08, N = 3 10.59 10.38 10.25 10.05 1. (CXX) g++ options: -O3 -lrt
LeelaChessZero Backend: Eigen OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.31.1 Backend: Eigen d c b a 30 60 90 120 150 SE +/- 2.11, N = 9 139 135 126 124 1. (CXX) g++ options: -flto -pthread
XNNPACK Model: FP32MobileNetV2 OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: FP32MobileNetV2 d c a b 1000 2000 3000 4000 5000 SE +/- 55.33, N = 4 4232 4326 4446 4485 1. (CXX) g++ options: -O3 -lrt -lm
XNNPACK Model: FP32MobileNetV3Large OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: FP32MobileNetV3Large b d c a 1400 2800 4200 5600 7000 SE +/- 38.09, N = 4 6333 6345 6369 6435 1. (CXX) g++ options: -O3 -lrt -lm
XNNPACK Model: FP32MobileNetV3Small OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: FP32MobileNetV3Small b a d c 900 1800 2700 3600 4500 SE +/- 23.42, N = 4 4189 4227 4245 4262 1. (CXX) g++ options: -O3 -lrt -lm
XNNPACK Model: FP16MobileNetV2 OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: FP16MobileNetV2 a b c d 900 1800 2700 3600 4500 SE +/- 38.38, N = 4 4113 4146 4161 4169 1. (CXX) g++ options: -O3 -lrt -lm
XNNPACK Model: FP16MobileNetV3Large OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: FP16MobileNetV3Large a d b c 1300 2600 3900 5200 6500 SE +/- 94.62, N = 4 5906 5936 5939 6175 1. (CXX) g++ options: -O3 -lrt -lm
XNNPACK Model: FP16MobileNetV3Small OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: FP16MobileNetV3Small b c a d 900 1800 2700 3600 4500 SE +/- 46.16, N = 4 4006 4106 4113 4128 1. (CXX) g++ options: -O3 -lrt -lm
XNNPACK Model: QU8MobileNetV2 OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: QU8MobileNetV2 d b a c 900 1800 2700 3600 4500 SE +/- 51.74, N = 4 3857 3874 3910 4015 1. (CXX) g++ options: -O3 -lrt -lm
XNNPACK Model: QU8MobileNetV3Large OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: QU8MobileNetV3Large a c d b 1300 2600 3900 5200 6500 SE +/- 27.64, N = 4 6216 6216 6285 6297 1. (CXX) g++ options: -O3 -lrt -lm
XNNPACK Model: QU8MobileNetV3Small OpenBenchmarking.org us, Fewer Is Better XNNPACK 2cd86b Model: QU8MobileNetV3Small a d b c 1100 2200 3300 4400 5500 SE +/- 33.76, N = 4 4707 4750 4815 4917 1. (CXX) g++ options: -O3 -lrt -lm
Phoronix Test Suite v10.8.5