TR 3970X oneDNN + clash AMD Ryzen Threadripper 3970X 32-Core testing with a ASUS ROG ZENITH II EXTREME (1201 BIOS) and AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 8GB on Ubuntu 20.10 via the Phoronix Test Suite. 1: Processor: AMD Ryzen Threadripper 3970X 32-Core @ 3.70GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH II EXTREME (1201 BIOS), Chipset: AMD Starship/Matisse, Memory: 64GB, Disk: Samsung SSD 980 PRO 500GB, Graphics: AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 8GB (1750/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: ASUS VP28U, Network: Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200 OS: Ubuntu 20.10, Kernel: 5.8.0-29-generic (x86_64), Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: amdgpu 19.1.0, OpenGL: 4.6 Mesa 20.2.1 (LLVM 11.0.0), Vulkan: 1.2.131, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 3840x2160 2: Processor: AMD Ryzen Threadripper 3970X 32-Core @ 3.70GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH II EXTREME (1201 BIOS), Chipset: AMD Starship/Matisse, Memory: 64GB, Disk: Samsung SSD 980 PRO 500GB, Graphics: AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 8GB (1750/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: ASUS VP28U, Network: Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200 OS: Ubuntu 20.10, Kernel: 5.8.0-29-generic (x86_64), Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: amdgpu 19.1.0, OpenGL: 4.6 Mesa 20.2.1 (LLVM 11.0.0), Vulkan: 1.2.131, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 3840x2160 3: Processor: AMD Ryzen Threadripper 3970X 32-Core @ 3.70GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH II EXTREME (1201 BIOS), Chipset: AMD Starship/Matisse, Memory: 64GB, Disk: Samsung SSD 980 PRO 500GB, Graphics: AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 8GB (1750/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: ASUS VP28U, Network: Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200 OS: Ubuntu 20.10, Kernel: 5.8.0-29-generic (x86_64), Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: amdgpu 19.1.0, OpenGL: 4.6 Mesa 20.2.1 (LLVM 11.0.0), Vulkan: 1.2.131, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 3840x2160 Libplacebo 2.72.2 Test: deband_heavy FPS > Higher Is Better 1 . 351.71 |================================================================== 2 . 350.84 |================================================================== 3 . 355.35 |=================================================================== Libplacebo 2.72.2 Test: polar_nocompute FPS > Higher Is Better 1 . 700.10 |=================================================================== 2 . 698.31 |=================================================================== 3 . 700.09 |=================================================================== Libplacebo 2.72.2 Test: hdr_peakdetect FPS > Higher Is Better 1 . 3262.67 |================================================================== 2 . 3261.05 |================================================================== 3 . 3262.45 |================================================================== Libplacebo 2.72.2 Test: av1_grain_lap FPS > Higher Is Better 1 . 602.60 |=================================================================== 2 . 600.12 |=================================================================== 3 . 601.56 |=================================================================== oneDNN 2.0 Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU ms < Lower Is Better 1 . 1.20958 |================================================================== 2 . 1.21820 |================================================================== 3 . 1.20900 |================================================================== oneDNN 2.0 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU ms < Lower Is Better 1 . 4.38972 |========================================================== 2 . 4.89684 |================================================================ 3 . 5.01124 |================================================================== oneDNN 2.0 Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better 1 . 0.936915 |================================================================= 2 . 0.935449 |================================================================= 3 . 0.937624 |================================================================= oneDNN 2.0 Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better 1 . 0.844063 |================================================================= 2 . 0.841168 |================================================================= 3 . 0.832783 |================================================================ oneDNN 2.0 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU ms < Lower Is Better 1 . 5.80922 |============================================================= 2 . 6.14756 |================================================================= 3 . 6.23653 |================================================================== oneDNN 2.0 Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU ms < Lower Is Better 1 . 1.60368 |================================================================= 2 . 1.61715 |================================================================== 3 . 1.60088 |================================================================= oneDNN 2.0 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU ms < Lower Is Better 1 . 2.72913 |================================================================== 2 . 2.70732 |================================================================= 3 . 2.70346 |================================================================= oneDNN 2.0 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better 1 . 6.20119 |============================================================ 2 . 6.74754 |================================================================= 3 . 6.79948 |================================================================== oneDNN 2.0 Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better 1 . 1.76063 |================================================================== 2 . 1.74974 |================================================================== 3 . 1.75098 |================================================================== oneDNN 2.0 Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better 1 . 1.56172 |================================================================== 2 . 1.56151 |================================================================== 3 . 1.56305 |================================================================== oneDNN 2.0 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU ms < Lower Is Better 1 . 3980.36 |================================================================== 2 . 3967.73 |================================================================== 3 . 3977.39 |================================================================== oneDNN 2.0 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU ms < Lower Is Better 1 . 929.53 |=================================================================== 2 . 930.29 |=================================================================== 3 . 930.50 |=================================================================== oneDNN 2.0 Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better 1 . 3974.30 |================================================================== 2 . 3975.07 |================================================================== 3 . 3980.60 |================================================================== oneDNN 2.0 Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better 1 . 930.36 |=================================================================== 2 . 932.47 |=================================================================== 3 . 931.09 |=================================================================== oneDNN 2.0 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU ms < Lower Is Better 1 . 0.403642 |================================================================ 2 . 0.404541 |================================================================ 3 . 0.408374 |================================================================= oneDNN 2.0 Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better 1 . 3977.18 |================================================================== 2 . 3961.17 |================================================================== 3 . 3980.54 |================================================================== oneDNN 2.0 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better 1 . 931.50 |=================================================================== 2 . 926.13 |================================================================== 3 . 933.87 |=================================================================== oneDNN 2.0 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better 1 . 0.927523 |================================================================= 2 . 0.927820 |================================================================= 3 . 0.927802 |================================================================= Timed Clash Compilation Time To Compile Seconds < Lower Is Better 1 . 386.33 |=================================================================== 2 . 387.23 |=================================================================== 3 . 387.16 |===================================================================