new rn AMD Ryzen Threadripper 3970X 32-Core testing with a ASUS ROG ZENITH II EXTREME (1802 BIOS) and AMD Radeon RX 5700 8GB on Ubuntu 22.04 via the Phoronix Test Suite. a: Processor: AMD Ryzen Threadripper 3970X 32-Core @ 3.70GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH II EXTREME (1802 BIOS), Chipset: AMD Starship/Matisse, Memory: 4 x 16GB DDR4-3600MT/s Corsair CMT64GX4M4Z3600C16, Disk: Samsung SSD 980 PRO 500GB, Graphics: AMD Radeon RX 5700 8GB, Audio: AMD Navi 10 HDMI Audio, Monitor: ASUS VP28U, Network: Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200 OS: Ubuntu 22.04, Kernel: 6.8.0-45-generic (x86_64), Desktop: GNOME Shell 42.9, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.1 (LLVM 13.0.1 DRM 3.57), Vulkan: 1.2.204, Compiler: GCC 11.4.0, File-System: ext4, Screen Resolution: 3840x2160 b: Processor: AMD Ryzen Threadripper 3970X 32-Core @ 3.70GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH II EXTREME (1802 BIOS), Chipset: AMD Starship/Matisse, Memory: 4 x 16GB DDR4-3600MT/s Corsair CMT64GX4M4Z3600C16, Disk: Samsung SSD 980 PRO 500GB, Graphics: AMD Radeon RX 5700 8GB, Audio: AMD Navi 10 HDMI Audio, Monitor: ASUS VP28U, Network: Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200 OS: Ubuntu 22.04, Kernel: 6.8.0-45-generic (x86_64), Desktop: GNOME Shell 42.9, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.1 (LLVM 13.0.1 DRM 3.57), Vulkan: 1.2.204, Compiler: GCC 11.4.0, File-System: ext4, Screen Resolution: 3840x2160 c: Processor: AMD Ryzen Threadripper 3970X 32-Core @ 3.70GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH II EXTREME (1802 BIOS), Chipset: AMD Starship/Matisse, Memory: 4 x 16GB DDR4-3600MT/s Corsair CMT64GX4M4Z3600C16, Disk: Samsung SSD 980 PRO 500GB, Graphics: AMD Radeon RX 5700 8GB, Audio: AMD Navi 10 HDMI Audio, Monitor: ASUS VP28U, Network: Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200 OS: Ubuntu 22.04, Kernel: 6.8.0-45-generic (x86_64), Desktop: GNOME Shell 42.9, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.1 (LLVM 13.0.1 DRM 3.57), Vulkan: 1.2.204, Compiler: GCC 11.4.0, File-System: ext4, Screen Resolution: 3840x2160 d: Processor: AMD Ryzen Threadripper 3970X 32-Core @ 3.70GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH II EXTREME (1802 BIOS), Chipset: AMD Starship/Matisse, Memory: 4 x 16GB DDR4-3600MT/s Corsair CMT64GX4M4Z3600C16, Disk: Samsung SSD 980 PRO 500GB, Graphics: AMD Radeon RX 5700 8GB, Audio: AMD Navi 10 HDMI Audio, Monitor: ASUS VP28U, Network: Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200 OS: Ubuntu 22.04, Kernel: 6.8.0-45-generic (x86_64), Desktop: GNOME Shell 42.9, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.1 (LLVM 13.0.1 DRM 3.57), Vulkan: 1.2.204, Compiler: GCC 11.4.0, File-System: ext4, Screen Resolution: 3840x2160 oneDNN 3.6 Harness: IP Shapes 1D - Engine: CPU ms < Lower Is Better a . 1.11186 |================================================================== b . 1.10833 |================================================================== c . 1.10933 |================================================================== d . 1.10728 |================================================================== oneDNN 3.6 Harness: IP Shapes 3D - Engine: CPU ms < Lower Is Better a . 6.26589 |========================================= b . 9.69956 |================================================================ c . 9.83021 |================================================================= d . 9.98705 |================================================================== oneDNN 3.6 Harness: Convolution Batch Shapes Auto - Engine: CPU ms < Lower Is Better a . 5.41074 |============================================================ b . 5.94199 |================================================================== c . 5.92981 |================================================================== d . 5.95880 |================================================================== oneDNN 3.6 Harness: Deconvolution Batch shapes_1d - Engine: CPU ms < Lower Is Better a . 8.92496 |================================================================== b . 8.96791 |================================================================== c . 8.68858 |================================================================ d . 8.80404 |================================================================= oneDNN 3.6 Harness: Deconvolution Batch shapes_3d - Engine: CPU ms < Lower Is Better a . 2.74673 |================================================================== b . 2.71999 |================================================================= c . 2.71566 |================================================================= d . 2.74986 |================================================================== oneDNN 3.6 Harness: Recurrent Neural Network Training - Engine: CPU ms < Lower Is Better a . 1063.25 |======================= b . 1065.95 |======================= c . 1068.48 |======================= d . 3079.07 |================================================================== oneDNN 3.6 Harness: Recurrent Neural Network Inference - Engine: CPU ms < Lower Is Better a . 544.59 |================================================================== b . 545.19 |================================================================== c . 546.14 |================================================================== d . 551.72 |=================================================================== LiteRT 2024-10-15 Model: DeepLab V3 Microseconds < Lower Is Better a . 3924.43 |================================================================= b . 3884.51 |================================================================= c . 3892.22 |================================================================= d . 3963.25 |================================================================== LiteRT 2024-10-15 Model: SqueezeNet Microseconds < Lower Is Better a . 2506.99 |================================================================= b . 2507.88 |================================================================= c . 2533.05 |================================================================== d . 2524.93 |================================================================== LiteRT 2024-10-15 Model: Inception V4 Microseconds < Lower Is Better a . 30460.8 |================================================================== b . 30192.8 |================================================================= c . 30102.7 |================================================================= d . 30176.1 |================================================================= LiteRT 2024-10-15 Model: NASNet Mobile Microseconds < Lower Is Better a . 21130.2 |================================================================== b . 21246.2 |================================================================== c . 21105.6 |================================================================== d . 21148.3 |================================================================== LiteRT 2024-10-15 Model: Mobilenet Float Microseconds < Lower Is Better a . 1820.83 |================================================================ b . 1890.63 |================================================================== c . 1830.85 |================================================================ d . 1827.86 |================================================================ LiteRT 2024-10-15 Model: Mobilenet Quant Microseconds < Lower Is Better a . 1405.64 |================================================================== b . 1299.31 |============================================================= c . 1300.82 |============================================================= d . 1291.57 |============================================================= LiteRT 2024-10-15 Model: Inception ResNet V2 Microseconds < Lower Is Better a . 29125.0 |================================================================== b . 29065.3 |================================================================= c . 29257.6 |================================================================== d . 29288.1 |================================================================== LiteRT 2024-10-15 Model: Quantized COCO SSD MobileNet v1 Microseconds < Lower Is Better a . 2356.92 |================================================================== b . 2344.66 |================================================================= c . 2356.07 |================================================================== d . 2366.82 |================================================================== XNNPACK b7b048 Model: FP32MobileNetV1 us < Lower Is Better a . 1929 |===================================================================== b . 1920 |==================================================================== c . 1925 |===================================================================== d . 1935 |===================================================================== XNNPACK b7b048 Model: FP32MobileNetV2 us < Lower Is Better a . 2695 |=================================================================== b . 2678 |================================================================== c . 2788 |===================================================================== d . 2692 |=================================================================== XNNPACK b7b048 Model: FP32MobileNetV3Large us < Lower Is Better a . 3828 |=================================================================== b . 3872 |==================================================================== c . 3938 |===================================================================== d . 3904 |==================================================================== XNNPACK b7b048 Model: FP32MobileNetV3Small us < Lower Is Better a . 2310 |===================================================================== b . 2209 |================================================================== c . 2266 |==================================================================== d . 2219 |================================================================== XNNPACK b7b048 Model: FP16MobileNetV1 us < Lower Is Better a . 1614 |===================================================================== b . 1606 |===================================================================== c . 1616 |===================================================================== d . 1597 |==================================================================== XNNPACK b7b048 Model: FP16MobileNetV2 us < Lower Is Better a . 2200 |==================================================================== b . 2156 |================================================================== c . 2194 |==================================================================== d . 2242 |===================================================================== XNNPACK b7b048 Model: FP16MobileNetV3Large us < Lower Is Better a . 3100 |===================================================================== b . 3099 |===================================================================== c . 3106 |===================================================================== d . 3080 |==================================================================== XNNPACK b7b048 Model: FP16MobileNetV3Small us < Lower Is Better a . 2270 |================================================================ b . 2151 |============================================================ c . 2074 |========================================================== d . 2455 |===================================================================== XNNPACK b7b048 Model: QS8MobileNetV2 us < Lower Is Better a . 2086 |===================================================================== b . 2058 |==================================================================== c . 2060 |==================================================================== d . 2097 |=====================================================================