移植性:
CUDA支持WinXP、XP64、Vista、Vista64、Linux32、Linux64、Mac OS
CAL支持WinXP、XP64、Vista、Vista64、Linux32、Linux64
官方开发文档:
CUDA:
Programming Documentation
CUDA Programming Guide 1.1 (revision history )
CUDA CUBLAS Library 1.1
CUDA CUFFT Library 1.1
MathWorks MATLAB® Plug-in
CUDA Roll Users Guide Version 4.3 (August 2007): The purpose of the NVIDIA® CUDA™ Roll is to install and configure the device driver and full toolchain necessary to run and develop CUDA programs on a Rocks™ cluster.
CUDA Toolkit Release Notes – Version 1.1 for Linux
CUDA Toolkit Release Notes – Version 1.1 for Windows XP
CUDA SDK Release Notes – Version 1.1 for Linux
CUDA SDK Release Notes – Version 1.1 for Windows XP
CUDA Programming Tools
CUDA Compiler Driver NVCC
PTX: Parallel Thread Execution ISA Version 1.1
CUDA Occupancy Calculator v1.2 : The CUDA Occupancy Calculator is a programmer tool that allows you to compute the multiprocessor occupancy of a GPU by a given CUDA kernel. The multiprocessor occupancy is the ratio of active warps to the maximum number of warps supported on a multiprocessor of the GPU, and is helpful in determining how efficient the kernel will be on the GPU.
Fortran
FORTRAN using pinned memory calling CUBLAS (Fortran_Cuda_Blas.tgz)
FORTRAN interface to CUDA kernel (Fortran_Cuda.tgz)
White Papers and Articles
Accelerating MATLAB with CUDA using MEX Files
Fast N-body Simulation with CUDA (from GPU Gems 3)
PARALLEL PROCESSING WITH CUDA
Trainings and Tutorials
SC07 Tutorial: High Performance Computing with CUDA
ECE 498 AL1: Programming Massively Parallel Processors
CAL:
Stream Computing Resources
Learn more about stream computing from these presentations and whitepapers.
Quadro
FX 5600
FX 4600
FX 3700
FX 1700
FX 570
FX 370
NVS 290
FX 3600M
FX 1600M
FX 570M
FX 360M
Quadro Plex 1000 Model IV
Quadro Plex 1000 Model S4
NVS 320M
NVS 140M
NVS 135M
NVS 130M
CAL:
R600 and newer GPUs are found with ATI Radeon™ HD2400, HD2600, HD2900 and HD3800 graphics board
CAL是一页里面提到一条 http://ati.amd.com/technology/streamcomputing/faq.html#11 Will the AMD FireStream SDK work on previous generation hardware? To run the CAL/Brook+ SDK, you need a platform based on the AMD R600 GPU or later. R600 and newer GPUs are found with ATI Radeon™ HD2400, HD2600, HD2900 and HD3800 graphics board.