nvidia-gpu-compute-oring-agx-32g ARMv8 Cortex-A78E testing with a NVIDIA Jetson AGX Orin Developer Kit (36.3.0-gcid-36191598 BIOS) and Orin on Ubuntu 22.04 via the Phoronix Test Suite. baseline: Processor: ARMv8 Cortex-A78E @ 2.20GHz (12 Cores), Motherboard: NVIDIA Jetson AGX Orin Developer Kit (36.3.0-gcid-36191598 BIOS), Memory: 30GB, Disk: 1000GB Samsung SSD 960 EVO 1TB + 64GB G1M15M, Graphics: Orin, Network: Realtek RTL8822CE 802.11ac PCIe OS: Ubuntu 22.04, Kernel: 5.15.136-tegra (aarch64), Desktop: GNOME Shell 42.9, Display Server: X Server 1.21.1.4, Display Driver: NVIDIA, Vulkan: 1.3.251, Compiler: GCC 11.4.0 + CUDA 12.2, File-System: ext4, Screen Resolution: 6582x1234 Betsy GPU Compressor 1.1 Beta Codec: ETC1 - Quality: Highest Seconds < Lower Is Better Betsy GPU Compressor 1.1 Beta Codec: ETC2 RGB - Quality: Highest Seconds < Lower Is Better cl-mem 2017-01-13 Benchmark: Copy cl-mem 2017-01-13 Benchmark: Read cl-mem 2017-01-13 Benchmark: Write FAHBench 2.3.2 Ns Per Day > Higher Is Better Libplacebo 6.338.2 FPS > Higher Is Better NCNN 20230517 Target: Vulkan GPU - Model: mobilenet ms < Lower Is Better baseline . 16.10 |============================================================= NCNN 20230517 Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2 ms < Lower Is Better baseline . 3.23 |============================================================== NCNN 20230517 Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3 ms < Lower Is Better baseline . 3.10 |============================================================== NCNN 20230517 Target: Vulkan GPU - Model: shufflenet-v2 ms < Lower Is Better baseline . 2.75 |============================================================== NCNN 20230517 Target: Vulkan GPU - Model: mnasnet ms < Lower Is Better baseline . 2.99 |============================================================== NCNN 20230517 Target: Vulkan GPU - Model: efficientnet-b0 ms < Lower Is Better baseline . 5.00 |============================================================== NCNN 20230517 Target: Vulkan GPU - Model: blazeface ms < Lower Is Better baseline . 1.43 |============================================================== NCNN 20230517 Target: Vulkan GPU - Model: googlenet ms < Lower Is Better baseline . 8.71 |============================================================== NCNN 20230517 Target: Vulkan GPU - Model: vgg16 ms < Lower Is Better baseline . 34.68 |============================================================= NCNN 20230517 Target: Vulkan GPU - Model: resnet18 ms < Lower Is Better baseline . 6.14 |============================================================== NCNN 20230517 Target: Vulkan GPU - Model: alexnet ms < Lower Is Better baseline . 6.42 |============================================================== NCNN 20230517 Target: Vulkan GPU - Model: resnet50 ms < Lower Is Better baseline . 16.13 |============================================================= NCNN 20230517 Target: Vulkan GPUv2-yolov3v2-yolov3 - Model: mobilenetv2-yolov3 ms < Lower Is Better baseline . 16.10 |============================================================= NCNN 20230517 Target: Vulkan GPU - Model: yolov4-tiny ms < Lower Is Better baseline . 25.06 |============================================================= NCNN 20230517 Target: Vulkan GPU - Model: squeezenet_ssd ms < Lower Is Better baseline . 10.82 |============================================================= NCNN 20230517 Target: Vulkan GPU - Model: regnety_400m ms < Lower Is Better baseline . 9.84 |============================================================== NCNN 20230517 Target: Vulkan GPU - Model: vision_transformer ms < Lower Is Better baseline . 267.88 |============================================================ NCNN 20230517 Target: Vulkan GPU - Model: FastestDet ms < Lower Is Better baseline . 3.56 |============================================================== ViennaCL 1.7.1 Test: CPU BLAS - sCOPY GB/s > Higher Is Better baseline . 64.4 |============================================================== ViennaCL 1.7.1 Test: CPU BLAS - sAXPY GB/s > Higher Is Better baseline . 79.4 |============================================================== ViennaCL 1.7.1 Test: CPU BLAS - sDOT GB/s > Higher Is Better baseline . 63.9 |============================================================== ViennaCL 1.7.1 Test: CPU BLAS - dCOPY GB/s > Higher Is Better baseline . 64.1 |============================================================== ViennaCL 1.7.1 Test: CPU BLAS - dAXPY GB/s > Higher Is Better baseline . 76.0 |============================================================== ViennaCL 1.7.1 Test: CPU BLAS - dDOT GB/s > Higher Is Better baseline . 61.8 |============================================================== ViennaCL 1.7.1 Test: CPU BLAS - dGEMV-N GB/s > Higher Is Better baseline . 47.4 |============================================================== ViennaCL 1.7.1 Test: CPU BLAS - dGEMV-T GB/s > Higher Is Better baseline . 62.1 |============================================================== ViennaCL 1.7.1 Test: CPU BLAS - dGEMM-NN GFLOPs/s > Higher Is Better baseline . 27.3 |============================================================== ViennaCL 1.7.1 Test: CPU BLAS - dGEMM-NT GFLOPs/s > Higher Is Better baseline . 23.5 |============================================================== ViennaCL 1.7.1 Test: CPU BLAS - dGEMM-TN GFLOPs/s > Higher Is Better baseline . 26.3 |============================================================== ViennaCL 1.7.1 Test: CPU BLAS - dGEMM-TT GFLOPs/s > Higher Is Better baseline . 26.6 |============================================================== VkFFT 1.3.4 Test: FFT + iFFT R2C / C2R Benchmark Score > Higher Is Better baseline . 11818 |============================================================= VkFFT 1.3.4 Test: FFT + iFFT C2C 1D batched in half precision Benchmark Score > Higher Is Better baseline . 31914 |============================================================= VkFFT 1.3.4 Test: FFT + iFFT C2C Bluestein in single precision Benchmark Score > Higher Is Better baseline . 3867 |============================================================== VkFFT 1.3.4 Test: FFT + iFFT C2C 1D batched in double precision Benchmark Score > Higher Is Better baseline . 3569 |============================================================== VkFFT 1.3.4 Test: FFT + iFFT C2C 1D batched in single precision Benchmark Score > Higher Is Better baseline . 26448 |============================================================= VkFFT 1.3.4 Test: FFT + iFFT C2C multidimensional in single precision Benchmark Score > Higher Is Better baseline . 10844 |============================================================= VkFFT 1.3.4 Test: FFT + iFFT C2C Bluestein benchmark in double precision Benchmark Score > Higher Is Better baseline . 734 |=============================================================== VkFFT 1.3.4 Test: FFT + iFFT C2C 1D batched in single precision, no reshuffling Benchmark Score > Higher Is Better baseline . 28743 |============================================================= VkResample 1.0 Upscale: 2x - Precision: Double ms < Lower Is Better baseline . 512.02 |============================================================ VkResample 1.0 Upscale: 2x - Precision: Single ms < Lower Is Better baseline . 56.74 |=============================================================