Nvidia Kepler Compute Capability

1 2 application compatibility on kepler the nvidia cuda c compiler nvcc can be used to generate both architecture specific cubin files and forward compatible ptx versions of each kernel.

Nvidia kepler compute capability. 2 the features of gk107 are similar to those of gk104. Compute capability of fermi and kepler gpus fermi gf100 fermi gf104 kepler gk104 kepler gk110 kepler gk210 compute capability 2 0 2 1 3 0 3 5 3 7 threads warp 32 max threads thread block 1024. It also includes 24 gb of gpu memory for training neural networks. 10 2 is the last official release for macos as support will not be available for macos in newer releases.

Nvidia kepler gpu computing accelerators are the world s fastest and most efficient high performance computing hpc companion processors. Recommended gpu for developers nvidia titan rtx nvidia titan rtx is built for data science ai research content creation and general gpu development. Built on the turing architecture it features 4608 576 full speed mixed precision tensor cores for accelerating ai and 72 rt cores for accelerating ray tracing. Last version with support for compute capability 3 x kepler.

Most geforce 600 series most geforce 700 series and some geforce 800m series gpus were based on kepler all manufactured in 28 nm. Based on the kepler compute architecture which is 3 times higher performance per watt than the previous fermi compute architecture 1 the tesla kepler gpu computing accelerators make hybrid computing. Kepler is nvidia s 3 rd generation architecture for cuda compute applications kepler retains and extends the same cuda programming model as in earlier nvidia architectures such as fermi and applications that follow the best practices for the fermi architecture should typically see speedups on the kepler architecture without any code changes. Kepler was nvidia s first microarchitecture to focus on energy efficiency.

2880 cuda cores and compute capability 3 5 gpu raycaster demo using nvidia cuda nvidia cuda compute capabilities in r334 67 and maxwell gpu codenames. Gk110 has compute capability 3 5. 1 throughout this guide fermi refers to devices of compute capability 2 x and kepler refers to devices of compute capability 3 x. Each cubin file targets a specific compute capability version and is forward compatible only with cuda architectures of the same major version number.

Cuda sdk 11 0 11 1 support for compute capability 3 5 8 6 kepler in part maxwell pascal volta turing ampere 33 new data types. Bfloat16 and tf32 on third generations tensor cores. The following table compares parameters of different compute capabilities for fermi and kepler gpu architectures.