vulkan benchmarks AMD Ryzen 9 7950X 16-Core testing with a ASUS ROG STRIX X670E-E GAMING WIFI (1416 BIOS) and AMD Radeon RX 6700 XT on Ubuntu 23.04 via the Phoronix Test Suite. a: Processor: AMD Ryzen 9 7950X 16-Core @ 4.50GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG STRIX X670E-E GAMING WIFI (1416 BIOS), Chipset: AMD Device 14d8, Memory: 32GB, Disk: Western Digital WD_BLACK SN850X 1000GB + 4001GB, Graphics: AMD Radeon RX 6700 XT (2855/1000MHz), Audio: AMD Navi 21/23, Monitor: ASUS MG28U, Network: Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411 OS: Ubuntu 23.04, Kernel: 6.4.6-060406-generic (x86_64), Desktop: GNOME Shell 44.2, Display Server: X Server 1.21.1.7 + Wayland, OpenGL: 4.6 Mesa 23.3~git2307260600.87109c~oibaf~l (git-87109c3 2023-07-26 lunar-oibaf-ppa) (LLVM 15.0.7 DRM 3.52), Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 3840x2160 b: Processor: AMD Ryzen 9 7950X 16-Core @ 4.50GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG STRIX X670E-E GAMING WIFI (1416 BIOS), Chipset: AMD Device 14d8, Memory: 32GB, Disk: Western Digital WD_BLACK SN850X 1000GB + 4001GB, Graphics: AMD Radeon RX 6700 XT (2855/1000MHz), Audio: AMD Navi 21/23, Monitor: ASUS MG28U, Network: Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411 OS: Ubuntu 23.04, Kernel: 6.4.6-060406-generic (x86_64), Desktop: GNOME Shell 44.2, Display Server: X Server 1.21.1.7 + Wayland, OpenGL: 4.6 Mesa 23.3~git2307260600.87109c~oibaf~l (git-87109c3 2023-07-26 lunar-oibaf-ppa) (LLVM 15.0.7 DRM 3.52), Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 3840x2160 c: Processor: AMD Ryzen 9 7950X 16-Core @ 4.50GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG STRIX X670E-E GAMING WIFI (1416 BIOS), Chipset: AMD Device 14d8, Memory: 32GB, Disk: Western Digital WD_BLACK SN850X 1000GB + 4001GB, Graphics: AMD Radeon RX 6700 XT (2855/1000MHz), Audio: AMD Navi 21/23, Monitor: ASUS MG28U, Network: Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411 OS: Ubuntu 23.04, Kernel: 6.4.6-060406-generic (x86_64), Desktop: GNOME Shell 44.2, Display Server: X Server 1.21.1.7 + Wayland, OpenGL: 4.6 Mesa 23.3~git2307260600.87109c~oibaf~l (git-87109c3 2023-07-26 lunar-oibaf-ppa) (LLVM 15.0.7 DRM 3.52), Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 3840x2160 vkpeak 20230730 fp32-scalar GFLOPS > Higher Is Better a . 13190.09 |================================================================= b . 12807.06 |=============================================================== c . 12860.56 |=============================================================== vkpeak 20230730 fp32-vec4 GFLOPS > Higher Is Better a . 12730.08 |================================================================= b . 12808.59 |================================================================= c . 12822.01 |================================================================= vkpeak 20230730 fp16-scalar GFLOPS > Higher Is Better a . 13154.15 |================================================================= b . 13145.19 |================================================================= c . 13136.79 |================================================================= vkpeak 20230730 fp16-vec4 GFLOPS > Higher Is Better a . 23232.42 |================================================================= b . 23390.44 |================================================================= c . 23387.26 |================================================================= vkpeak 20230730 fp64-scalar GFLOPS > Higher Is Better a . 841.40 |=================================================================== b . 839.20 |=================================================================== c . 839.01 |=================================================================== vkpeak 20230730 fp64-vec4 GFLOPS > Higher Is Better a . 841.80 |=================================================================== b . 836.55 |=================================================================== c . 836.16 |=================================================================== vkpeak 20230730 int32-scalar GIOPS > Higher Is Better a . 2272.62 |================================================================== b . 2269.25 |================================================================== c . 2269.06 |================================================================== vkpeak 20230730 int32-vec4 GIOPS > Higher Is Better a . 2658.73 |================================================================== b . 2640.08 |================================================================== c . 2638.69 |================================================================== vkpeak 20230730 int16-scalar GIOPS > Higher Is Better a . 13102.75 |================================================================= b . 13070.81 |================================================================= c . 13063.86 |================================================================= vkpeak 20230730 int16-vec4 GIOPS > Higher Is Better a . 23123.77 |================================================================ b . 23396.59 |================================================================= c . 23385.44 |================================================================= VkFFT 1.2.31 Test: FFT + iFFT R2C / C2R Benchmark Score > Higher Is Better a . 42105 |=================================================================== b . 42163 |=================================================================== c . 43021 |==================================================================== VkFFT 1.2.31 Test: FFT + iFFT C2C 1D batched in half precision Benchmark Score > Higher Is Better a . 91597 |==================================================================== b . 91812 |==================================================================== c . 91744 |==================================================================== VkFFT 1.2.31 Test: FFT + iFFT C2C Bluestein in single precision Benchmark Score > Higher Is Better a . 11340 |==================================================================== b . 11273 |==================================================================== c . 11311 |==================================================================== VkFFT 1.2.31 Test: FFT + iFFT C2C 1D batched in double precision Benchmark Score > Higher Is Better a . 20816 |==================================================================== b . 20822 |==================================================================== c . 20847 |==================================================================== VkFFT 1.2.31 Test: FFT + iFFT C2C 1D batched in single precision Benchmark Score > Higher Is Better a . 47887 |==================================================================== b . 47948 |==================================================================== c . 47971 |==================================================================== VkFFT 1.2.31 Test: FFT + iFFT C2C multidimensional in single precision Benchmark Score > Higher Is Better a . 33001 |==================================================================== b . 32751 |=================================================================== c . 32812 |==================================================================== VkFFT 1.2.31 Test: FFT + iFFT C2C Bluestein benchmark in double precision Benchmark Score > Higher Is Better a . 4717 |===================================================================== b . 4695 |===================================================================== c . 4670 |==================================================================== VkFFT 1.2.31 Test: FFT + iFFT C2C 1D batched in single precision, no reshuffling Benchmark Score > Higher Is Better a . 50504 |==================================================================== b . 50643 |==================================================================== c . 50596 |==================================================================== VkResample 1.0 Upscale: 2x - Precision: Single ms < Lower Is Better a . 11.69 |==================================================================== b . 11.69 |==================================================================== c . 11.69 |==================================================================== NCNN 20230517 Target: CPU - Model: mobilenet ms < Lower Is Better a . 8.05 |===================================================================== b . 7.97 |==================================================================== c . 8.02 |===================================================================== NCNN 20230517 Target: CPU-v2-v2 - Model: mobilenet-v2 ms < Lower Is Better a . 3.16 |===================================================================== b . 3.16 |===================================================================== c . 3.18 |===================================================================== NCNN 20230517 Target: CPU - Model: shufflenet-v2 ms < Lower Is Better a . 3.34 |===================================================================== b . 3.34 |===================================================================== c . 3.35 |===================================================================== NCNN 20230517 Target: CPU - Model: mnasnet ms < Lower Is Better a . 2.97 |===================================================================== b . 2.97 |===================================================================== c . 2.99 |===================================================================== NCNN 20230517 Target: CPU - Model: efficientnet-b0 ms < Lower Is Better a . 3.90 |===================================================================== b . 3.85 |==================================================================== c . 3.88 |===================================================================== NCNN 20230517 Target: CPU - Model: blazeface ms < Lower Is Better a . 1.38 |===================================================================== b . 1.38 |===================================================================== c . 1.39 |===================================================================== NCNN 20230517 Target: CPU - Model: googlenet ms < Lower Is Better a . 7.94 |===================================================================== b . 7.82 |==================================================================== c . 7.93 |===================================================================== NCNN 20230517 Target: CPU - Model: vgg16 ms < Lower Is Better a . 23.75 |==================================================================== b . 23.49 |=================================================================== c . 23.45 |=================================================================== NCNN 20230517 Target: CPU - Model: resnet18 ms < Lower Is Better a . 5.29 |===================================================================== b . 5.20 |==================================================================== c . 5.24 |==================================================================== NCNN 20230517 Target: CPU - Model: alexnet ms < Lower Is Better a . 4.41 |===================================================================== b . 4.32 |==================================================================== c . 4.31 |=================================================================== NCNN 20230517 Target: CPU - Model: resnet50 ms < Lower Is Better a . 10.20 |==================================================================== b . 10.01 |=================================================================== c . 10.11 |=================================================================== NCNN 20230517 Target: CPU - Model: yolov4-tiny ms < Lower Is Better a . 12.90 |==================================================================== b . 12.74 |=================================================================== c . 12.81 |==================================================================== NCNN 20230517 Target: CPU - Model: squeezenet_ssd ms < Lower Is Better a . 7.07 |===================================================================== b . 7.06 |===================================================================== c . 7.10 |===================================================================== NCNN 20230517 Target: CPU - Model: regnety_400m ms < Lower Is Better a . 8.18 |==================================================================== b . 8.18 |==================================================================== c . 8.27 |===================================================================== NCNN 20230517 Target: CPU - Model: vision_transformer ms < Lower Is Better a . 32.49 |==================================================================== b . 31.95 |=================================================================== c . 31.77 |================================================================== NCNN 20230517 Target: CPU - Model: FastestDet ms < Lower Is Better a . 3.62 |============================================================= b . 4.05 |==================================================================== c . 4.11 |===================================================================== NCNN 20230517 Target: CPU-v3-v3 - Model: mobilenet-v3 ms < Lower Is Better a . 3.18 |===================================================================== c . 3.17 |===================================================================== NCNN 20230517 Target: Vulkan GPU - Model: mobilenet ms < Lower Is Better a . 8.05 |===================================================================== b . 8.04 |===================================================================== c . 8.03 |===================================================================== NCNN 20230517 Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2 ms < Lower Is Better a . 3.17 |===================================================================== b . 3.14 |==================================================================== c . 3.13 |==================================================================== NCNN 20230517 Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3 ms < Lower Is Better a . 3.17 |==================================================================== c . 3.20 |===================================================================== NCNN 20230517 Target: Vulkan GPU - Model: shufflenet-v2 ms < Lower Is Better a . 3.35 |===================================================================== b . 3.33 |===================================================================== c . 3.32 |==================================================================== NCNN 20230517 Target: Vulkan GPU - Model: mnasnet ms < Lower Is Better a . 2.98 |===================================================================== b . 2.95 |==================================================================== c . 2.96 |===================================================================== NCNN 20230517 Target: Vulkan GPU - Model: efficientnet-b0 ms < Lower Is Better a . 3.86 |===================================================================== b . 3.82 |==================================================================== c . 3.83 |==================================================================== NCNN 20230517 Target: Vulkan GPU - Model: blazeface ms < Lower Is Better a . 1.38 |===================================================================== b . 1.37 |===================================================================== c . 1.37 |===================================================================== NCNN 20230517 Target: Vulkan GPU - Model: googlenet ms < Lower Is Better a . 7.90 |===================================================================== b . 7.85 |===================================================================== c . 7.80 |==================================================================== NCNN 20230517 Target: Vulkan GPU - Model: vgg16 ms < Lower Is Better a . 23.51 |==================================================================== b . 23.56 |==================================================================== c . 23.54 |==================================================================== NCNN 20230517 Target: Vulkan GPU - Model: resnet18 ms < Lower Is Better a . 5.28 |===================================================================== b . 5.23 |==================================================================== c . 5.21 |==================================================================== NCNN 20230517 Target: Vulkan GPU - Model: alexnet ms < Lower Is Better a . 4.31 |===================================================================== b . 4.33 |===================================================================== c . 4.33 |===================================================================== NCNN 20230517 Target: Vulkan GPU - Model: resnet50 ms < Lower Is Better a . 10.01 |==================================================================== b . 10.00 |==================================================================== c . 10.00 |==================================================================== NCNN 20230517 Target: Vulkan GPU - Model: yolov4-tiny ms < Lower Is Better a . 12.84 |==================================================================== b . 12.87 |==================================================================== c . 12.81 |==================================================================== NCNN 20230517 Target: Vulkan GPU - Model: squeezenet_ssd ms < Lower Is Better a . 7.09 |===================================================================== b . 7.07 |===================================================================== c . 7.06 |===================================================================== NCNN 20230517 Target: Vulkan GPU - Model: regnety_400m ms < Lower Is Better a . 8.16 |===================================================================== b . 8.21 |===================================================================== c . 8.00 |=================================================================== NCNN 20230517 Target: Vulkan GPU - Model: vision_transformer ms < Lower Is Better a . 31.88 |==================================================================== b . 31.85 |==================================================================== c . 31.79 |==================================================================== NCNN 20230517 Target: Vulkan GPU - Model: FastestDet ms < Lower Is Better a . 4.10 |===================================================================== b . 4.07 |==================================================================== c . 4.09 |===================================================================== NCNN 20230517 Target: CPU-v3-v3-v3 - Model: mobilenet ms < Lower Is Better b . 8.01 |===================================================================== c . 8.00 |===================================================================== NCNN 20230517 Target: CPU-v3-v3-v3-v2-v2 - Model: mobilenet-v2 ms < Lower Is Better b . 3.15 |===================================================================== c . 3.14 |===================================================================== NCNN 20230517 Target: CPU-v3-v3-v3-v3-v3 - Model: mobilenet-v3 ms < Lower Is Better b . 3.16 |===================================================================== c . 3.16 |===================================================================== NCNN 20230517 Target: CPU-v3-v3-v3 - Model: shufflenet-v2 ms < Lower Is Better b . 3.33 |===================================================================== c . 3.34 |===================================================================== NCNN 20230517 Target: CPU-v3-v3-v3 - Model: mnasnet ms < Lower Is Better b . 2.97 |===================================================================== c . 2.97 |===================================================================== NCNN 20230517 Target: CPU-v3-v3-v3 - Model: efficientnet-b0 ms < Lower Is Better b . 3.82 |==================================================================== c . 3.89 |===================================================================== NCNN 20230517 Target: CPU-v3-v3-v3 - Model: blazeface ms < Lower Is Better b . 1.37 |===================================================================== c . 1.38 |===================================================================== NCNN 20230517 Target: CPU-v3-v3-v3 - Model: googlenet ms < Lower Is Better b . 7.84 |===================================================================== c . 7.88 |===================================================================== NCNN 20230517 Target: CPU-v3-v3-v3 - Model: vgg16 ms < Lower Is Better b . 23.50 |=================================================================== c . 23.99 |==================================================================== NCNN 20230517 Target: CPU-v3-v3-v3 - Model: resnet18 ms < Lower Is Better b . 5.21 |==================================================================== c . 5.26 |===================================================================== NCNN 20230517 Target: CPU-v3-v3-v3 - Model: alexnet ms < Lower Is Better b . 4.29 |===================================================================== c . 4.28 |===================================================================== NCNN 20230517 Target: CPU-v3-v3-v3 - Model: resnet50 ms < Lower Is Better b . 10.01 |================================================================== c . 10.33 |==================================================================== NCNN 20230517 Target: CPU-v3-v3-v3 - Model: yolov4-tiny ms < Lower Is Better b . 12.98 |==================================================================== c . 12.89 |==================================================================== NCNN 20230517 Target: CPU-v3-v3-v3 - Model: squeezenet_ssd ms < Lower Is Better b . 7.03 |===================================================================== c . 7.04 |===================================================================== NCNN 20230517 Target: CPU-v3-v3-v3 - Model: regnety_400m ms < Lower Is Better b . 8.05 |==================================================================== c . 8.14 |===================================================================== NCNN 20230517 Target: CPU-v3-v3-v3 - Model: vision_transformer ms < Lower Is Better b . 31.65 |==================================================================== c . 31.78 |==================================================================== NCNN 20230517 Target: CPU-v3-v3-v3 - Model: FastestDet ms < Lower Is Better b . 4.07 |===================================================================== c . 4.08 |===================================================================== NCNN 20230517 Target: Vulkan GPU-v3-v3-v3 - Model: mobilenet ms < Lower Is Better b . 8.00 |===================================================================== c . 7.95 |===================================================================== NCNN 20230517 Target: Vulkan GPU-v3-v3-v3-v2-v2 - Model: mobilenet-v2 ms < Lower Is Better b . 3.15 |===================================================================== c . 3.15 |===================================================================== NCNN 20230517 Target: Vulkan GPU-v3-v3-v3-v3-v3 - Model: mobilenet-v3 ms < Lower Is Better b . 3.16 |===================================================================== c . 3.17 |===================================================================== NCNN 20230517 Target: Vulkan GPU-v3-v3-v3 - Model: shufflenet-v2 ms < Lower Is Better b . 3.33 |===================================================================== c . 3.33 |===================================================================== NCNN 20230517 Target: Vulkan GPU-v3-v3-v3 - Model: mnasnet ms < Lower Is Better b . 2.96 |===================================================================== c . 2.96 |===================================================================== NCNN 20230517 Target: Vulkan GPU-v3-v3-v3 - Model: efficientnet-b0 ms < Lower Is Better b . 3.85 |===================================================================== c . 3.82 |==================================================================== NCNN 20230517 Target: Vulkan GPU-v3-v3-v3 - Model: blazeface ms < Lower Is Better b . 1.37 |===================================================================== c . 1.36 |==================================================================== NCNN 20230517 Target: Vulkan GPU-v3-v3-v3 - Model: googlenet ms < Lower Is Better b . 7.97 |===================================================================== c . 7.83 |==================================================================== NCNN 20230517 Target: Vulkan GPU-v3-v3-v3 - Model: vgg16 ms < Lower Is Better b . 23.42 |==================================================================== c . 23.54 |==================================================================== NCNN 20230517 Target: Vulkan GPU-v3-v3-v3 - Model: resnet18 ms < Lower Is Better b . 5.42 |===================================================================== c . 5.23 |=================================================================== NCNN 20230517 Target: Vulkan GPU-v3-v3-v3 - Model: alexnet ms < Lower Is Better b . 4.42 |===================================================================== c . 4.30 |=================================================================== NCNN 20230517 Target: Vulkan GPU-v3-v3-v3 - Model: resnet50 ms < Lower Is Better b . 9.87 |=================================================================== c . 10.03 |==================================================================== NCNN 20230517 Target: Vulkan GPU-v3-v3-v3 - Model: yolov4-tiny ms < Lower Is Better b . 12.77 |==================================================================== c . 12.86 |==================================================================== NCNN 20230517 Target: Vulkan GPU-v3-v3-v3 - Model: squeezenet_ssd ms < Lower Is Better b . 7.14 |===================================================================== c . 7.07 |==================================================================== NCNN 20230517 Target: Vulkan GPU-v3-v3-v3 - Model: regnety_400m ms < Lower Is Better b . 8.27 |===================================================================== c . 7.98 |=================================================================== NCNN 20230517 Target: Vulkan GPU-v3-v3-v3 - Model: vision_transformer ms < Lower Is Better b . 31.71 |==================================================================== c . 31.66 |==================================================================== NCNN 20230517 Target: Vulkan GPU-v3-v3-v3 - Model: FastestDet ms < Lower Is Better b . 4.06 |===================================================================== c . 3.69 |===============================================================