User:SurajYadav27/sandbox

CUDA CORES IN NVIDIA’s GPU CUDA(Compute Unified Device Architecture) is a parallel computing platform and Application Programming Interface(API) model created by Nvidia. Cuda cores are just small part of the whole GPU(Graphics Processing Unit).The “CUDA Cores” are nvidia’s equivalent to AMD’s Stream Processors. The CUDA is designed to work with programming languages like C, C++ and Fortran. Each GPU can contain hundreds of thousands of CUDA cores. Architecture changes in a fashion that makes upcoming generation comparison often non-linear, but generally, more CUDA core will equate more raw compute power from the GPU. The Kepler to Maxwell Architecture jump saw nearly a 40% efficiency gain in CUDA Cores processing ability, illustrating the difficulty of linearly drawing comparisons without proper benchmarks. CUDA Cores are parallel processors just like your CPU might be a dual-core or a quad-core device, nVidia GPU’s host hundreds of thousands of cores. The cores are responsible for processing all the data that is fed in and out of the GPU, performing game graphics calculations that are resolved visually to the end-user. An example of something a CUDA core might do would include rendering scenery in game, drawing character models, or resolving complex lighting and shading within an environment. It also doesn’t mean that a GPU with more CUDA core will be more powerfull than with the one which is having less CUDA cores, there only a minor changes not a drastic change in the performance.