cuda-testing
Intel Xeon E3-1280 v5 testing with a MSI C236A WORKSTATION (MS-7998) v1.0 and eVGA NVIDIA GeForce GTX 960 2043MB on Ubuntu 16.04 via the Phoronix Test Suite.


GeForce GTX 1080: 

	Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: Samsung SSD 950 PRO 256GB, Graphics: GeForce GTX 1080 8187MB (909/5005MHz), Audio: Realtek ALC1150, Network: Intel Connection

	OS: Ubuntu 16.04, Kernel: 4.4.0-22-generic (x86_64), Desktop: Unity 7.4.0, Display Driver: NVIDIA 367.18, OpenGL: 4.5.0, Vulkan: 1.0.8, Compiler: GCC 5.3.1 20160413 + CUDA 8.0, File-System: ext4, Screen Resolution: 3840x2160

GeForce GTX 980: 

	Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: Samsung SSD 950 PRO 256GB, Graphics: NVIDIA GeForce GTX 980 4091MB (1126/3505MHz), Audio: Realtek ALC1150, Network: Intel Connection

	OS: Ubuntu 16.04, Kernel: 4.4.0-22-generic (x86_64), Desktop: Unity 7.4.0, Display Driver: NVIDIA 367.18, OpenGL: 4.5.0, Vulkan: 1.0.8, Compiler: GCC 5.3.1 20160413 + CUDA 8.0, File-System: ext4, Screen Resolution: 3840x2160

GeForce GTX 960: 

	Processor: Intel Xeon E3-1280 v5 @ 4.00GHz (8 Cores), Motherboard: MSI C236A WORKSTATION (MS-7998) v1.0, Chipset: Intel Sky Lake, Memory: 16384MB, Disk: Samsung SSD 950 PRO 256GB, Graphics: eVGA NVIDIA GeForce GTX 960 2043MB (1277/3505MHz), Audio: Realtek ALC1150, Network: Intel Connection

	OS: Ubuntu 16.04, Kernel: 4.4.0-22-generic (x86_64), Desktop: Unity 7.4.0, Display Driver: NVIDIA 367.18, OpenGL: 4.5.0, Vulkan: 1.0.8, Compiler: GCC 5.3.1 20160413 + CUDA 8.0, File-System: ext4, Screen Resolution: 3840x2160


SHOC Scalable HeterOgeneous Computing 2015-11-10
Target: CUDA - Benchmark: Triad
GB/s > Higher Is Better
GeForce GTX 1080 . 14.86 |=====================================================
GeForce GTX 980 .. 14.74 |=====================================================
GeForce GTX 960 .. 14.36 |===================================================


SHOC Scalable HeterOgeneous Computing 2015-11-10
Target: CUDA - Benchmark: FFT SP
GFLOPS > Higher Is Better
GeForce GTX 1080 . 461.28 |====================================================
GeForce GTX 980 .. 292.78 |=================================
GeForce GTX 960 .. 189.14 |=====================


SHOC Scalable HeterOgeneous Computing 2015-11-10
Target: CUDA - Benchmark: MD5 Hash
GHash/s > Higher Is Better
GeForce GTX 1080 . 11.98 |=====================================================
GeForce GTX 980 .. 6.53  |=============================
GeForce GTX 960 .. 3.88  |=================


SHOC Scalable HeterOgeneous Computing 2015-11-10
Target: CUDA - Benchmark: Max SP Flops
GFLOPS > Higher Is Better
GeForce GTX 1080 . 9397.41 |===================================================
GeForce GTX 980 .. 4999.85 |===========================
GeForce GTX 960 .. 2944.94 |================


SHOC Scalable HeterOgeneous Computing 2015-11-10
Target: CUDA - Benchmark: Bus Speed Download
GB/s > Higher Is Better
GeForce GTX 1080 . 12.53 |=====================================================
GeForce GTX 980 .. 12.53 |=====================================================
GeForce GTX 960 .. 12.53 |=====================================================


SHOC Scalable HeterOgeneous Computing 2015-11-10
Target: CUDA - Benchmark: Bus Speed Readback
GB/s > Higher Is Better
GeForce GTX 1080 . 13.22 |=====================================================
GeForce GTX 980 .. 13.22 |=====================================================
GeForce GTX 960 .. 13.21 |=====================================================


SHOC Scalable HeterOgeneous Computing 2015-11-10
Target: CUDA - Benchmark: Texture Read Bandwidth
GB/s > Higher Is Better
GeForce GTX 1080 . 528.41 |====================================================
GeForce GTX 980 .. 332.16 |=================================
GeForce GTX 960 .. 381.05 |=====================================


Caffe AlexNet 2016-06-11
Build: CUDA
Milli-Seconds < Lower Is Better
Xeon E3-1280 v5 - CPU Only . 1787207.00 |======================================
GeForce GTX 1080 ........... 8959.77    |
GeForce GTX 980 ............ 15504.53   |
GeForce GTX 960 ............ 28134.07   |=


CUDA Mini-Nbody 2015-11-10
Test: Original
Seconds < Lower Is Better
GeForce GTX 1080 . 30.51 |====================
GeForce GTX 980 .. 46.51 |==============================
GeForce GTX 960 .. 82.29 |=====================================================


CUDA Mini-Nbody 2015-11-10
Test: Cache Blocking
Seconds < Lower Is Better
GeForce GTX 1080 . 14.02 |====================
GeForce GTX 980 .. 24.91 |====================================
GeForce GTX 960 .. 36.30 |=====================================================


CUDA Mini-Nbody 2015-11-10
Test: Loop Unrolling
Seconds < Lower Is Better
GeForce GTX 1080 . 14.52 |======================
GeForce GTX 980 .. 24.63 |=====================================
GeForce GTX 960 .. 35.71 |=====================================================


CUDA Mini-Nbody 2015-11-10
Test: SOA Data Layout
Seconds < Lower Is Better
GeForce GTX 1080 . 28.58 |===================
GeForce GTX 980 .. 51.02 |=================================
GeForce GTX 960 .. 81.27 |=====================================================


CUDA Mini-Nbody 2015-11-10
Test: Flush Denormals To Zero
Seconds < Lower Is Better
GeForce GTX 1080 . 28.58 |===================
GeForce GTX 980 .. 50.44 |=================================
GeForce GTX 960 .. 81.19 |=====================================================