Implementing GPU-Driven Rendering

In this chapter, we embark on a deep dive into the intricacies of manipulating geometry specifically for GPU rendering. The traditional approach heavily relies on the CPU for various tasks, which can be a bottleneck in many scenarios. Our goal here is to liberate your rendering techniques from such constraints. We aim to put the GPU in the driver’s seat, ensuring efficient processing by capitalizing on its parallel processing prowess. We will unravel the technique of generating and drawing lines, not from the traditional space of the CPU but directly from shaders, such as vertex or fragment shaders. This not only enhances efficiency but also opens new realms of creativity. Taking a step further, we’ll demonstrate how to extend this novel line-drawing feature to display numbers right from shaders. This capability paves the way for real-time displays and feedback without toggling between GPU and CPU. We then shift our gaze to a more intricate...