2970WX July AMD Ryzen Threadripper 2970WX 24-Core testing with a Gigabyte X399 AORUS Gaming 7 (F12h BIOS) and Sapphire AMD Radeon RX 550 640SP / 560/560X 4GB on Ubuntu 20.04 via the Phoronix Test Suite. 1: Processor: AMD Ryzen Threadripper 2970WX 24-Core @ 3.00GHz (24 Cores / 48 Threads), Motherboard: Gigabyte X399 AORUS Gaming 7 (F12h BIOS), Chipset: AMD 17h, Memory: 16GB, Disk: 120GB Corsair Force MP500, Graphics: Sapphire AMD Radeon RX 550 640SP / 560/560X 4GB (1300/1750MHz), Audio: Realtek ALC1220, Monitor: VA2431, Network: Qualcomm Atheros Killer E2500 + 2 x QLogic cLOM8214 1/10GbE + Intel 8265 / 8275 OS: Ubuntu 20.04, Kernel: 5.9.0-050900rc6daily20200926-generic (x86_64) 20200925, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.9, OpenGL: 4.6 Mesa 20.2.6 (LLVM 11.0.0), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080 2: Processor: AMD Ryzen Threadripper 2970WX 24-Core @ 3.00GHz (24 Cores / 48 Threads), Motherboard: Gigabyte X399 AORUS Gaming 7 (F12h BIOS), Chipset: AMD 17h, Memory: 16GB, Disk: 120GB Corsair Force MP500, Graphics: Sapphire AMD Radeon RX 550 640SP / 560/560X 4GB (1300/1750MHz), Audio: Realtek ALC1220, Monitor: VA2431, Network: Qualcomm Atheros Killer E2500 + 2 x QLogic cLOM8214 1/10GbE + Intel 8265 / 8275 OS: Ubuntu 20.04, Kernel: 5.9.0-050900rc6daily20200926-generic (x86_64) 20200925, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.9, OpenGL: 4.6 Mesa 20.2.6 (LLVM 11.0.0), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080 3: Processor: AMD Ryzen Threadripper 2970WX 24-Core @ 3.00GHz (24 Cores / 48 Threads), Motherboard: Gigabyte X399 AORUS Gaming 7 (F12h BIOS), Chipset: AMD 17h, Memory: 16GB, Disk: 120GB Corsair Force MP500, Graphics: Sapphire AMD Radeon RX 550 640SP / 560/560X 4GB (1300/1750MHz), Audio: Realtek ALC1220, Monitor: VA2431, Network: Qualcomm Atheros Killer E2500 + 2 x QLogic cLOM8214 1/10GbE + Intel 8265 / 8275 OS: Ubuntu 20.04, Kernel: 5.9.0-050900rc6daily20200926-generic (x86_64) 20200925, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.9, OpenGL: 4.6 Mesa 20.2.6 (LLVM 11.0.0), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080 4: Processor: AMD Ryzen Threadripper 2970WX 24-Core @ 3.00GHz (24 Cores / 48 Threads), Motherboard: Gigabyte X399 AORUS Gaming 7 (F12h BIOS), Chipset: AMD 17h, Memory: 16GB, Disk: 120GB Corsair Force MP500, Graphics: Sapphire AMD Radeon RX 550 640SP / 560/560X 4GB (1300/1750MHz), Audio: Realtek ALC1220, Monitor: VA2431, Network: Qualcomm Atheros Killer E2500 + 2 x QLogic cLOM8214 1/10GbE + Intel 8265 / 8275 OS: Ubuntu 20.04, Kernel: 5.9.0-050900rc6daily20200926-generic (x86_64) 20200925, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.9, OpenGL: 4.6 Mesa 20.2.6 (LLVM 11.0.0), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080 5: Processor: AMD Ryzen Threadripper 2970WX 24-Core @ 3.00GHz (24 Cores / 48 Threads), Motherboard: Gigabyte X399 AORUS Gaming 7 (F12h BIOS), Chipset: AMD 17h, Memory: 16GB, Disk: 120GB Corsair Force MP500, Graphics: Sapphire AMD Radeon RX 550 640SP / 560/560X 4GB (1300/1750MHz), Audio: Realtek ALC1220, Monitor: VA2431, Network: Qualcomm Atheros Killer E2500 + 2 x QLogic cLOM8214 1/10GbE + Intel 8265 / 8275 OS: Ubuntu 20.04, Kernel: 5.9.0-050900rc6daily20200926-generic (x86_64) 20200925, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.9, OpenGL: 4.6 Mesa 20.2.6 (LLVM 11.0.0), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080 Renaissance 0.12 Test: Akka Unbalanced Cobwebbed Tree ms < Lower Is Better 1 . 21703.7 |================================================================= 2 . 21813.4 |================================================================= 3 . 21476.3 |================================================================ 4 . 22068.2 |================================================================== 5 . 21886.5 |================================================================= Apache Cassandra 4.0 Test: Mixed 1:1 Op/s > Higher Is Better 1 . 107795 |================================================================== 2 . 98990 |============================================================= 3 . 105033 |================================================================= 4 . 109065 |=================================================================== 5 . 105682 |================================================================= Renaissance 0.12 Test: ALS Movie Lens ms < Lower Is Better 1 . 9836.8 |================================================================= 2 . 9878.2 |================================================================= 3 . 10001.3 |================================================================== 4 . 9814.8 |================================================================= 5 . 9857.0 |================================================================= TNN 0.3 Target: CPU - Model: DenseNet ms < Lower Is Better 1 . 3034.70 |================================================================== 2 . 3023.79 |================================================================== 3 . 3026.06 |================================================================== 4 . 3035.01 |================================================================== 5 . 3035.69 |================================================================== Mobile Neural Network 1.2 Model: inception-v3 ms < Lower Is Better 1 . 59.53 |================================================================== 2 . 61.55 |==================================================================== 3 . 59.60 |================================================================== 4 . 59.25 |================================================================= 5 . 57.16 |=============================================================== Mobile Neural Network 1.2 Model: mobilenet-v1-1.0 ms < Lower Is Better 1 . 5.114 |==================================================================== 2 . 4.948 |================================================================== 3 . 5.109 |==================================================================== 4 . 5.011 |=================================================================== 5 . 4.996 |================================================================== Mobile Neural Network 1.2 Model: MobileNetV2_224 ms < Lower Is Better 1 . 7.707 |================================================================= 2 . 8.046 |==================================================================== 3 . 7.615 |================================================================ 4 . 7.590 |================================================================ 5 . 7.292 |============================================================== Mobile Neural Network 1.2 Model: SqueezeNetV1.0 ms < Lower Is Better 1 . 12.28 |================================================================= 2 . 11.02 |========================================================== 3 . 11.36 |============================================================ 4 . 12.68 |=================================================================== 5 . 12.81 |==================================================================== Mobile Neural Network 1.2 Model: resnet-v2-50 ms < Lower Is Better 1 . 44.87 |============================================================ 2 . 50.67 |==================================================================== 3 . 46.24 |============================================================== 4 . 46.48 |============================================================== 5 . 46.22 |============================================================== Mobile Neural Network 1.2 Model: squeezenetv1.1 ms < Lower Is Better 1 . 9.868 |================================================================ 2 . 8.954 |========================================================== 3 . 8.024 |==================================================== 4 . 10.299 |=================================================================== 5 . 9.974 |================================================================= Mobile Neural Network 1.2 Model: mobilenetV3 ms < Lower Is Better 1 . 5.165 |=================================================================== 2 . 4.324 |======================================================== 3 . 5.225 |==================================================================== 4 . 4.359 |========================================================= 5 . 4.266 |======================================================== Apache Cassandra 4.0 Test: Reads Op/s > Higher Is Better 1 . 91943 |============================================= 2 . 105525 |=================================================== 3 . 138409 |=================================================================== 4 . 102666 |================================================== 5 . 106201 |=================================================== GravityMark 1.2 Resolution: 1920 x 1080 - Renderer: Vulkan Frames Per Second > Higher Is Better 1 . 33.2 |===================================================================== 2 . 33.2 |===================================================================== 3 . 33.2 |===================================================================== 4 . 33.2 |===================================================================== 5 . 33.2 |===================================================================== Renaissance 0.12 Test: Savina Reactors.IO ms < Lower Is Better 1 . 12292.6 |================================================================ 2 . 12326.5 |================================================================ 3 . 11693.4 |============================================================= 4 . 12667.5 |================================================================== 5 . 11704.2 |============================================================= NCNN 20210720 Target: CPU - Model: regnety_400m ms < Lower Is Better 1 . 38.70 |==================================================================== 2 . 36.84 |================================================================= 3 . 36.10 |=============================================================== 4 . 36.25 |================================================================ 5 . 36.55 |================================================================ NCNN 20210720 Target: CPU - Model: squeezenet_ssd ms < Lower Is Better 1 . 42.91 |==================================================================== 2 . 36.43 |========================================================== 3 . 38.83 |============================================================== 4 . 41.15 |================================================================= 5 . 41.54 |================================================================== NCNN 20210720 Target: CPU - Model: yolov4-tiny ms < Lower Is Better 1 . 52.00 |==================================================================== 2 . 47.08 |============================================================== 3 . 49.51 |================================================================= 4 . 49.12 |================================================================ 5 . 48.67 |================================================================ NCNN 20210720 Target: CPU - Model: resnet50 ms < Lower Is Better 1 . 68.54 |==================================================================== 2 . 53.95 |====================================================== 3 . 50.10 |================================================== 4 . 54.60 |====================================================== 5 . 67.92 |=================================================================== NCNN 20210720 Target: CPU - Model: alexnet ms < Lower Is Better 1 . 24.56 |=================================================== 2 . 26.93 |======================================================== 3 . 22.99 |=============================================== 4 . 27.83 |========================================================= 5 . 32.93 |==================================================================== NCNN 20210720 Target: CPU - Model: resnet18 ms < Lower Is Better 1 . 44.83 |====================================================== 2 . 56.06 |==================================================================== 3 . 47.96 |========================================================== 4 . 51.53 |=============================================================== 5 . 31.22 |====================================== NCNN 20210720 Target: CPU - Model: vgg16 ms < Lower Is Better 1 . 97.15 |=================================================================== 2 . 87.61 |============================================================ 3 . 80.57 |======================================================= 4 . 98.93 |==================================================================== 5 . 83.42 |========================================================= NCNN 20210720 Target: CPU - Model: googlenet ms < Lower Is Better 1 . 35.87 |==================================================================== 2 . 30.90 |=========================================================== 3 . 31.94 |============================================================= 4 . 33.67 |================================================================ 5 . 31.59 |============================================================ NCNN 20210720 Target: CPU - Model: blazeface ms < Lower Is Better 1 . 5.79 |===================================================================== 2 . 5.33 |================================================================ 3 . 5.58 |================================================================== 4 . 5.61 |=================================================================== 5 . 5.53 |================================================================== NCNN 20210720 Target: CPU - Model: efficientnet-b0 ms < Lower Is Better 1 . 17.03 |==================================================================== 2 . 16.06 |================================================================ 3 . 17.12 |==================================================================== 4 . 16.68 |================================================================== 5 . 16.52 |================================================================== NCNN 20210720 Target: CPU - Model: mnasnet ms < Lower Is Better 1 . 12.19 |================================================================ 2 . 12.86 |==================================================================== 3 . 11.84 |=============================================================== 4 . 12.29 |================================================================= 5 . 11.71 |============================================================== NCNN 20210720 Target: CPU - Model: shufflenet-v2 ms < Lower Is Better 1 . 14.00 |==================================================================== 2 . 12.78 |============================================================== 3 . 13.56 |================================================================== 4 . 13.81 |=================================================================== 5 . 13.53 |================================================================== NCNN 20210720 Target: CPU-v3-v3 - Model: mobilenet-v3 ms < Lower Is Better 1 . 13.01 |================================================================= 2 . 13.65 |==================================================================== 3 . 12.74 |=============================================================== 4 . 12.59 |=============================================================== 5 . 13.56 |==================================================================== NCNN 20210720 Target: CPU-v2-v2 - Model: mobilenet-v2 ms < Lower Is Better 1 . 13.38 |=================================================================== 2 . 13.53 |==================================================================== 3 . 13.02 |================================================================= 4 . 13.28 |=================================================================== 5 . 13.50 |==================================================================== NCNN 20210720 Target: CPU - Model: mobilenet ms < Lower Is Better 1 . 29.59 |================================================================= 2 . 31.08 |==================================================================== 3 . 28.71 |=============================================================== 4 . 28.82 |=============================================================== 5 . 28.75 |=============================================================== Renaissance 0.12 Test: Apache Spark PageRank ms < Lower Is Better 1 . 5088.7 |=============================================================== 2 . 5443.7 |=================================================================== 3 . 5041.3 |============================================================== 4 . 4485.8 |======================================================= 5 . 5102.0 |=============================================================== Apache Cassandra 4.0 Test: Mixed 1:3 Op/s > Higher Is Better 1 . 97181 |========================================================== 2 . 105086 |============================================================== 3 . 112837 |=================================================================== 4 . 96271 |========================================================= 5 . 111614 |================================================================== Renaissance 0.12 Test: Genetic Algorithm Using Jenetics + Futures ms < Lower Is Better 1 . 2665.9 |================================================================== 2 . 2700.4 |=================================================================== 3 . 2679.7 |================================================================== 4 . 2620.0 |================================================================= 5 . 2660.0 |================================================================== Apache Cassandra 4.0 Test: Writes Op/s > Higher Is Better 1 . 137190 |=================================================================== 2 . 130153 |================================================================ 3 . 131113 |================================================================ 4 . 131846 |================================================================ 5 . 130920 |================================================================ Renaissance 0.12 Test: In-Memory Database Shootout ms < Lower Is Better 1 . 6178.5 |================================================================ 2 . 6439.5 |=================================================================== 3 . 6178.3 |================================================================ 4 . 6142.0 |================================================================ 5 . 6457.2 |=================================================================== YafaRay 3.5.1 Total Time For Sample Scene Seconds < Lower Is Better 1 . 96.09 |================================================================== 2 . 98.40 |==================================================================== 3 . 98.83 |==================================================================== 4 . 98.45 |==================================================================== 5 . 96.47 |================================================================== Renaissance 0.12 Test: Apache Spark ALS ms < Lower Is Better 1 . 2359.6 |=================================================================== 2 . 2279.6 |================================================================= 3 . 2305.2 |================================================================= 4 . 2202.1 |=============================================================== 5 . 2220.6 |=============================================================== NCNN 20210720 Target: Vulkan GPU - Model: regnety_400m ms < Lower Is Better 1 . 10.19 |=================================================================== 2 . 10.37 |==================================================================== 3 . 10.12 |================================================================== 4 . 10.34 |==================================================================== 5 . 10.15 |=================================================================== NCNN 20210720 Target: Vulkan GPU - Model: squeezenet_ssd ms < Lower Is Better 1 . 13.47 |================================================================= 2 . 13.18 |================================================================ 3 . 14.09 |==================================================================== 4 . 13.19 |================================================================ 5 . 13.25 |================================================================ NCNN 20210720 Target: Vulkan GPU - Model: yolov4-tiny ms < Lower Is Better 1 . 20.00 |=================================================================== 2 . 20.24 |==================================================================== 3 . 20.00 |=================================================================== 4 . 19.98 |=================================================================== 5 . 20.01 |=================================================================== NCNN 20210720 Target: Vulkan GPU - Model: resnet50 ms < Lower Is Better 1 . 15.32 |==================================================================== 2 . 15.16 |=================================================================== 3 . 15.32 |==================================================================== 4 . 15.20 |=================================================================== 5 . 15.23 |==================================================================== NCNN 20210720 Target: Vulkan GPU - Model: alexnet ms < Lower Is Better 1 . 10.29 |=================================================================== 2 . 10.43 |==================================================================== 3 . 10.39 |==================================================================== 4 . 10.33 |=================================================================== 5 . 10.38 |==================================================================== NCNN 20210720 Target: Vulkan GPU - Model: resnet18 ms < Lower Is Better 1 . 7.05 |==================================================================== 2 . 7.11 |===================================================================== 3 . 6.99 |==================================================================== 4 . 6.92 |=================================================================== 5 . 7.06 |===================================================================== NCNN 20210720 Target: Vulkan GPU - Model: vgg16 ms < Lower Is Better 1 . 28.93 |==================================================================== 2 . 29.10 |==================================================================== 3 . 28.80 |=================================================================== 4 . 29.06 |==================================================================== 5 . 28.90 |==================================================================== NCNN 20210720 Target: Vulkan GPU - Model: googlenet ms < Lower Is Better 1 . 11.31 |================================================================== 2 . 11.34 |=================================================================== 3 . 11.58 |==================================================================== 4 . 11.38 |=================================================================== 5 . 11.28 |================================================================== NCNN 20210720 Target: Vulkan GPU - Model: blazeface ms < Lower Is Better 1 . 3.11 |============================================================== 2 . 3.10 |============================================================== 3 . 2.97 |============================================================ 4 . 3.03 |============================================================= 5 . 3.44 |===================================================================== NCNN 20210720 Target: Vulkan GPU - Model: efficientnet-b0 ms < Lower Is Better 1 . 15.48 |==================================================================== 2 . 15.27 |=================================================================== 3 . 15.26 |=================================================================== 4 . 15.21 |=================================================================== 5 . 15.30 |=================================================================== NCNN 20210720 Target: Vulkan GPU - Model: mnasnet ms < Lower Is Better 1 . 5.85 |==================================================================== 2 . 5.83 |==================================================================== 3 . 5.91 |===================================================================== 4 . 5.88 |===================================================================== 5 . 5.86 |==================================================================== NCNN 20210720 Target: Vulkan GPU - Model: shufflenet-v2 ms < Lower Is Better 1 . 4.97 |===================================================================== 2 . 4.97 |===================================================================== 3 . 4.97 |===================================================================== 4 . 4.93 |==================================================================== 5 . 4.99 |===================================================================== NCNN 20210720 Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3 ms < Lower Is Better 1 . 7.35 |=================================================================== 2 . 7.43 |==================================================================== 3 . 7.59 |===================================================================== 4 . 7.39 |=================================================================== 5 . 7.35 |=================================================================== NCNN 20210720 Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2 ms < Lower Is Better 1 . 5.56 |==================================================================== 2 . 5.60 |===================================================================== 3 . 5.61 |===================================================================== 4 . 5.50 |==================================================================== 5 . 5.46 |=================================================================== NCNN 20210720 Target: Vulkan GPU - Model: mobilenet ms < Lower Is Better 1 . 13.12 |==================================================================== 2 . 12.92 |=================================================================== 3 . 13.17 |==================================================================== 4 . 12.98 |=================================================================== 5 . 12.97 |=================================================================== Facebook RocksDB 6.22.1 Test: Sequential Fill Op/s > Higher Is Better 1 . 629253 |================================================================ 2 . 593316 |============================================================ 3 . 654095 |================================================================== 4 . 586679 |=========================================================== 5 . 662215 |=================================================================== Renaissance 0.12 Test: Scala Dotty ms < Lower Is Better 1 . 1132.9 |=================================================================== 2 . 1111.5 |================================================================== 3 . 1038.3 |============================================================= 4 . 968.0 |========================================================= 5 . 965.4 |========================================================= Facebook RocksDB 6.22.1 Test: Random Fill Sync Op/s > Higher Is Better 1 . 16365 |======================================================= 2 . 17701 |=========================================================== 3 . 12380 |========================================== 4 . 20282 |==================================================================== 5 . 19465 |================================================================= Facebook RocksDB 6.22.1 Test: Read Random Write Random Op/s > Higher Is Better 1 . 1929291 |=============================================================== 2 . 2007692 |================================================================== 3 . 1985772 |================================================================= 4 . 1986790 |================================================================= 5 . 1986640 |================================================================= Facebook RocksDB 6.22.1 Test: Update Random Op/s > Higher Is Better 1 . 516006 |=================================================================== 2 . 518649 |=================================================================== 3 . 497709 |================================================================ 4 . 498715 |================================================================ 5 . 515847 |=================================================================== Facebook RocksDB 6.22.1 Test: Random Fill Op/s > Higher Is Better 1 . 559506 |============================================================ 2 . 629001 |=================================================================== 3 . 558560 |=========================================================== 4 . 563861 |============================================================ 5 . 524031 |======================================================== Facebook RocksDB 6.22.1 Test: Random Read Op/s > Higher Is Better 1 . 105202520 |============================================================= 2 . 106197835 |============================================================== 3 . 107926092 |=============================================================== 4 . 110355654 |================================================================ 5 . 108371044 |=============================================================== Facebook RocksDB 6.22.1 Test: Read While Writing Op/s > Higher Is Better 1 . 3684442 |================================================================ 2 . 3795875 |================================================================== 3 . 3593614 |============================================================== 4 . 3656007 |================================================================ 5 . 3634278 |=============================================================== Renaissance 0.12 Test: Finagle HTTP Requests ms < Lower Is Better 1 . 4099.4 |=================================================================== 2 . 3906.7 |================================================================ 3 . 4055.7 |================================================================== 4 . 4040.6 |================================================================== 5 . 4006.9 |================================================================= C-Blosc 2.0 Compressor: blosclz MB/s > Higher Is Better 1 . 15950.7 |================================================================== 2 . 15733.0 |================================================================= 3 . 15895.6 |================================================================= 4 . 16025.3 |================================================================== 5 . 15930.8 |================================================================== Renaissance 0.12 Test: Random Forest ms < Lower Is Better 1 . 920.0 |==================================================================== 2 . 892.7 |================================================================== 3 . 902.9 |=================================================================== 4 . 911.1 |=================================================================== 5 . 891.3 |================================================================== Renaissance 0.12 Test: Apache Spark Bayes ms < Lower Is Better 1 . 1209.1 |================================================================== 2 . 1184.2 |================================================================ 3 . 1180.9 |================================================================ 4 . 1162.1 |=============================================================== 5 . 1231.9 |=================================================================== Natron 2.4 Input: Spaceship FPS > Higher Is Better 1 . 3.4 |====================================================================== 2 . 3.2 |================================================================== 3 . 3.3 |==================================================================== 4 . 3.2 |================================================================== 5 . 2.8 |========================================================== TNN 0.3 Target: CPU - Model: MobileNet v2 ms < Lower Is Better 1 . 306.27 |=================================================================== 2 . 305.28 |=================================================================== 3 . 306.19 |=================================================================== 4 . 305.57 |=================================================================== 5 . 304.27 |=================================================================== TNN 0.3 Target: CPU - Model: SqueezeNet v1.1 ms < Lower Is Better 1 . 256.36 |================================================================== 2 . 256.49 |================================================================== 3 . 256.32 |================================================================== 4 . 256.30 |================================================================== 5 . 260.29 |=================================================================== TNN 0.3 Target: CPU - Model: SqueezeNet v2 ms < Lower Is Better 1 . 72.12 |=================================================================== 2 . 72.51 |==================================================================== 3 . 71.38 |=================================================================== 4 . 71.22 |=================================================================== 5 . 72.72 |====================================================================