Hacker News new | past | comments | ask | show | jobs | submit login

Buffers see extensive use in modern shaders. A frame will get rasterized several times for different shaders before those buffers are all composited into an output buffer. Several types of things can only be known if you've got the full scene rasterized or at least more than a few lines.

If you've only got fixed function shaders and a relatively simple rasterizer the raster-to-display works fine. The DS also only has a single contributor to the output. A game on a PS4 or a PC is just one of several contributors to a compositing window/display manager which is actually doing final composition to the output device(s).




Right, but my point here was mainly aimed at the compositor. Those shaders are much much simpler. They're just combining existing buffers in a single pass every time I've seen it. I'm not saying swapping out the main GPU, but instead the programmability that already exists in the scan out engine.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: