How does X11 forwarding work?

Question

How does X11 forwarding work? I am interested in knowing whether the processing to render graphics is done at the end of the host running the application or the host displaying the graphical interface?

Should I use a GPU intensive application (game) - where should I have the GPU installed (server end / client end)? Of course the server will need the GPU if it is running CUDA / openCL applications, but what about the display?

---

This question was closed on StackOverflow. This was a link pointed to me, but I am looking to understand more of the underlying protocol and performance tweaks.

Answer 1

Assuming you're using OpenGL, the GPU should be installed on the host where the X server is running. The client will send rendering commands to the X server, which will then take advantage of the GPU to process the rendering commands.