qaq Intel Core i7-4790K testing with a Apple Mac-42FD25EABCABB274 (IM151.88Z.0207.B01.1411201235 BIOS) and AMD Radeon R7 370 R9 270X/370X 2GB on Fedora 32 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/2009247-AS-QAQ88563900 .
qaq Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Display Driver OpenGL OpenCL Vulkan Compiler File-System Screen Resolution qaq Intel Core i7-4790K @ 4.40GHz (4 Cores / 8 Threads) Apple Mac-42FD25EABCABB274 (IM151.88Z.0207.B01.1411201235 BIOS) Intel 4th Gen Core DRAM 16384MB 1000GB APPLE HDD ST1000 + 121GB APPLE SSD SD0128 + 2000GB Basic AMD Radeon R7 370 R9 270X/370X 2GB Cirrus Logic CS4206 iMac Broadcom NetXtreme BCM57766 PCIe + Broadcom BCM4360 802.11ac Fedora 32 5.8.4-200.fc32.x86_64 (x86_64) KDE Plasma 5.18.5 X Server 1.20.8 amdgpu 19.1.0 4.6 Mesa 20.1.6 (LLVM 10.0.0) OpenCL 1.1 Mesa 20.1.6 + OpenCL 1.2 pocl 1.5 RelWithDebInfo LLVM 10.0.0 RELOC SLEEF DISTRO POCL_DEBUG + OpenCL 2.0 beignet 1.3 1.2.131 Clang 10.0.0 + LLVM 10.0.1 ext4 3840x2160 OpenBenchmarking.org - radeon.si_support=0 amdgpu.si_support=1 - Scaling Governor: intel_cpufreq ondemand - SELinux + itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Mitigation of Microcode + tsx_async_abort: Not affected
qaq ncnn: CPU - squeezenet_int8 ncnn: CPU - mobilenet_v3 ncnn: CPU - squeezenet ncnn: CPU - mnasnet ncnn: CPU - blazeface ncnn: CPU - googlenet_int8 ncnn: CPU - vgg16_int8 ncnn: CPU - resnet18_int8 ncnn: CPU - alexnet ncnn: CPU - resnet50_int8 ncnn: CPU - mobilenetv2_yolov3 qaq 22.81 7.31 4.90 7.11 2.01 69.70 249.71 40.42 23.89 141.69 31.91 OpenBenchmarking.org
NCNN Target: CPU - Model: squeezenet_int8 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: squeezenet_int8 qaq 5 10 15 20 25 SE +/- 0.08, N = 6 22.81 MIN: 22.14 / MAX: 28.78 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: mobilenet_v3 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: mobilenet_v3 qaq 2 4 6 8 10 SE +/- 0.07, N = 6 7.31 MIN: 7.01 / MAX: 14.2 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: squeezenet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: squeezenet qaq 1.1025 2.205 3.3075 4.41 5.5125 SE +/- 0.01, N = 6 4.90 MIN: 4.76 / MAX: 6.86 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: mnasnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: mnasnet qaq 2 4 6 8 10 SE +/- 0.04, N = 6 7.11 MIN: 6.84 / MAX: 15.67 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: blazeface OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: blazeface qaq 0.4523 0.9046 1.3569 1.8092 2.2615 SE +/- 0.05, N = 6 2.01 MIN: 1.89 / MAX: 49.74 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: googlenet_int8 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: googlenet_int8 qaq 16 32 48 64 80 SE +/- 0.22, N = 6 69.70 MIN: 68.19 / MAX: 122.15 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: vgg16_int8 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: vgg16_int8 qaq 50 100 150 200 250 SE +/- 0.70, N = 6 249.71 MIN: 244.16 / MAX: 383.9 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: resnet18_int8 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: resnet18_int8 qaq 9 18 27 36 45 SE +/- 0.12, N = 6 40.42 MIN: 39.57 / MAX: 68.46 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: alexnet OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: alexnet qaq 6 12 18 24 30 SE +/- 0.05, N = 6 23.89 MIN: 23.31 / MAX: 107.1 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: resnet50_int8 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: resnet50_int8 qaq 30 60 90 120 150 SE +/- 0.29, N = 6 141.69 MIN: 139.15 / MAX: 312.19 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
NCNN Target: CPU - Model: mobilenetv2_yolov3 OpenBenchmarking.org ms, Fewer Is Better NCNN 20200916 Target: CPU - Model: mobilenetv2_yolov3 qaq 7 14 21 28 35 SE +/- 0.05, N = 6 31.91 MIN: 31.18 / MAX: 38.27 1. (CXX) g++ options: -O2 -rdynamic -lgomp -lpthread
Phoronix Test Suite v10.8.4