Reverse-Engineering GTA V (adriancourreges.com)

An anonymous reader writes: Software engineer Adrian Courrèges posted on his blog a breakdown of the rendering of a frame in Grand Theft Auto: V. Each rendering pass is explained in detail, with all the techniques and the tricks Rockstar used to make the game run on 8-year-old consoles. It's a fascinating trip through the making of a frame and reminds us of how far GPU computing power has come. Here's a brief snippet from the beginning: "As a first step, the game renders a cubemap of the environment. This cubemap is generated in realtime at each frame, its purpose is to help render realistic reflections later. This part is forward-rendered. How is such cubemap rendered? For those not familiar with the technique, this is just like you would do in the real world when taking a panoramic picture: put the camera on a tripod, imagine you’re standing right in the middle of a big cube and shoot at the 6 faces of the cube, one by one, rotating by 90 degrees each time. This is exactly how the game does: each face is rendered into a 128x128 HDR texture."

Comments

Re:Is this all just speculation?

[UnknownSoldier]

Easy. By

a) examining the assembly from the pixel shader.
b) discarding all rendering until the end of frame.

The only "interpretation" is how the HDR tone mapping is applied but it is close enough to the real game that it is a minor point.