So much differences in code between chapter 1 and 2, undocumented

[Gaetz]

Hello, I am using your book, and comparing chapter 1's and 2's code, I have spotted a lot of differences, not directly linked with chapter's 2 content.

I can see the lighting changed, and there is mipmap support. Would it be possible to get a list of intended changes so that I can understand the code by comparing it ?

[theWatchmen]

Yes, some of the changes are not strictly related to the chapter - we should have probably added them to chapter 1 as a starting point. I admit it's a bit hard to isolate the code that is purely related to the chapter. However I think we highlighted most (if not all) of the relevant code for chapter 2 in the book. There might be some lines missing to avoid making the code too long or verbose, but it should provide a good guide to isolate the intended changes.

These boil down to:

• Bindless Rendering: we tagged these changes with [TAG: BINDLESS] (although not in the shaders). Most of the code is in gpu_device.cpp and, for the shaders, the main change of note is the use of the nonuniformEXT qualifier. There is also some plumbing - i.e. passing the index of the material textures instead of the textures themselves for a draw. You can follow the changes made to the MeshDraw structure for this aspect.
• Automating pipeline layout generation: most of the code is in the SPIR-V parser, spirv_parser.h/cpp. For the actual usage, you can find it here and here, where we create the descriptor set layout from the reflected data.
• Pipeline caching: the code changes for this feature are concentrated here and should be easy to follow.

Hope this helps, let us know if you have any further questions!

[Gaetz]

Ok, thank you very much. I code-compared chapter 2 and 1 and reintegrated changes.