Hello, world. It’s Siraj. And today we’re going to do some GPU programming to add the elements of two arrays together. Sounds super simple, I know. But it will be a way for us to introduce ourselves to the key concepts that make up the incredible world of parallel computing used heavily in deep learning. The field of computer graphics traces its roots back to the earliest days of computer science. HP and IBM were some of the first companies to introduce silicon hardware capable of generating imagery, which helped fuel the decades long transformation of the southern SF Bay area into the technology hub known as Silicon Valley. There were endless applications of computer graphics. From scientific simulations to movies to video games, Programmers created graphical software. And that software was broken down Into a series of instructions That the central processing unit of a computer was capable of carrying out. The CPU is the main chip in a computer. Responsible for telling all the other components in a computer. What to do using it’s given instructions? Just like your friendly neighborhood dictator. If we look at the difference between early games and modern-day games, It is truly remarkable. How much of a difference in realism there is? The process of rendering a scene consists of a number of steps. A computer must perform. In a specific sequence and the sequence is called the graphics pipeline. The first step is to set up the objects in the scene. Objects models are a mesh of triangles that describe its shape. This is done by listing the exact coordinates. Of every corner of every triangle.