Vuda aims to bring cuda to vulkan

Jgbit, a developer at GitHub, has launched an ambitious open source project called VUDA, which is inspired by Nvidia's CUDA API to bring an easy-to-access GPU computing interface to the open source world.

VUDA will bring the full potential of CUDA to Vulkan

VUDA is implemented as a wrapper over the very popular Vulkan next-generation graphics API, which provides low-level access to Nvidia's hardware. VUDA comes as a C ++ library for the header only, which means it is compatible with all platforms that have a C ++ compiler and that are compatible with Vulkan. While the project is still young, its potential is enormous, especially due to the open source nature of using the MIT license. The page on GitHub comes with a very basic sample, which could be a good start to use the library.

We recommend reading our post about Nvidia RTX 2080 Ti Review in Spanish

Most people confuse CUDA with a language or perhaps an API, but it is not. CUDA is a parallel computing platform and programming model, which makes using a GPU for general-purpose computing simple and elegant. The developer still programs in C, C ++, Fortran, or an ever-expanding list of supported languages, and incorporates extensions of these in the form of some basic keywords. These keywords allow the developer to express large amounts of parallelism, and direct the compiler to the part of the application that maps to the GPU.

The arrival of CUDA in Vulkan will open a great and important door for developers who use this API, which is increasingly popular for its benefits and for being cross-platform.