Tags » Cuda

Getting Started With CUDA Programming

Graphics Processing Units or GPUs were traditionally used for the purpose of rendering graphics on screen. They were optimized for throughput and could render millions of pixels simultaneously, by performing the same computations on millions of individual data elements in parallel. 534 more words

Guides And Tips

Material System

Just added more material types into the current implementation, here are some results:

Gold xyzrgb dragon

Copper xyzrgb dragon

Silver xyzrgb dragon

Rough xyzrgb dragon … 19 more words

Gpu

New tracing routine

After implementing a path tracing routine in OpenGL 4.3+ compute shader, I decided to do a counterpart implementation in cuda, just to see the capability of it and the flexibility for developers. 71 more words

Gpu

A simple CUDA program

This is my first CUDA program. It just adds two vector arrays. Vector addition is a simple program to begin with to start learning GPU programming using CUDA. 548 more words

20160412

동시성 프로그램의 이해
– 행렬곱을 메모리 접근을 고려하지 않고 만들면 메모리 접근이 과다하게 발생한다. 이를 해결하기 위해서는 blocking을 사용. Blocking을 사용하면 데이터 접근의 locality를 높일 수 있음. 101 more words

1) Memo

First steps on CUDA

On this post I’m going to use the programming language “C” for our first steps in CUDA. Remember that “C++” is also compatible.

To start we have to know the basic structure of any CUDA program: is like a regular program but in this one we have to communicate with our Nvidia GPU, keeping in mind memory assignment, memory transactions and calling the kernel itself. 976 more words

Sin Categoría

CUDA template function for creating texture objects (cudaTextureObject_t)

I just wrote this function for creating lots of cudaTextureObject_t for 1-D, linear arrays of differently typed data. I did this to convert my old, static global… 172 more words

Code Bits