thecherno/raytracing

YouTube ray tracing series

raytracing's Issues

Wont Build for me ?

Thanks for providing this for free.

I followed the instructions here, using latest vs 2022

1. Clone recursively: git clone --recursive https://github.com/TheCherno/RayTracing
2. Run scripts/Setup.bat
3. Open RayTracing.sln and hit F5

I cant get the sln to build, get various files not found, not sure if I have missed something or there is a path missing in the sln setup ?

A selection of the errors I get
1>C:\RayTracing\Walnut\vendor\imgui\imgui.h(60,10): fatal error C1083: Cannot open include file: 'float.h': No such file or directory
2>C:\RayTracing\Walnut\vendor\glfw\include\GLFW\glfw3.h(103,10): fatal error C1083: Cannot open include file: 'stddef.h': No such file or directory

4>C:\Program Files\Microsoft Visual Studio\2022\Community\VC\Tools\MSVC\14.31.31103\include\yvals.h(12,10): fatal error C1083: Cannot open include file: 'crtdbg.h': No such file or directory
4>Done building project "RayTracing.vcxproj" -- FAILED.

Any thoughts on how to fix this would be appreciated, thanks.

Converting code to run on the GPU

Hi, I'm quite new to Vulkan, but I wanted to see if anyone could give me directives on how to modify the project to run on the GPU.

There's a bunch of resources out there (I've listed a few below), but I'm not sure how all of these tie in to work with the current codebase.

• https://nvpro-samples.github.io/vk_raytracing_tutorial_KHR/
• https://github.com/nvpro-samples/vk_mini_path_tracer/tree/main

Is there any way I can modify the current code to reproduce something similar to the examples above?

Multithreading and Optimization Discussion

Hey all! For those following along with the series on YouTube, I hope you've been enjoying it thus far! In the latest episode (Ep 11), we introduced multithreading into our code base by using std::for_each with the parallel execution policy. I mentioned that if the community has any other suggestions, and wants to do some testing to see if we can multithread in a more efficient way, I'd open a GitHub issue for this discussion - and here we are!

I figured crowd-sourcing this would be a good idea since your mileage may vary - certain techniques might be better or worse on certain hardware and architectures. Feel free to fork this repository and implement something, and then drop it in a comment here so we can test. If your method is faster than our std::for_each method, make sure to include some profiling data and your hardware specifications.

Thanks all! ❤️

• Cherno

Render() throws an error (image_state was nullptr) because Render() renders the image before the data is available

Easy fix with:
if(m_Image)
Render();

m_AccumulationData of 'Renderer' class will not initialize to a value of 0

Background

In my PC (CPU I5-8400), debug mode
m_AccumulationData of 'Renderer' class will not initialize to a value of 0

Process

The first 'Renderer::OnResize' event is not a problem:
Because the value of 'm_FrameIndex' is 1, 'memset' will be called to initialize

The second call to 'Renderer::OnResize' will not call memset because 'm_FrameIndex' is not equal to 1, resulting in 'm_AccumulationData' being all is a negative number:

Thanks

Getting error when compiling, something about the input

I get this error when I try to compile:
Severity Code Description Project File Line Suppression State
Error LNK2019 unresolved external symbol "public: static class Walnut::Application & __cdecl Walnut::Application::Get(void)" (?Get@Application@Walnut@@SAAEAV12@XZ) referenced in function "public: static struct glm::vec<2,float,0> __cdecl Walnut::Input::GetMousePosition(void)" (?GetMousePosition@Input@Walnut@@sa?AU?$vec@$01M$0A@@glm@@xz) RayTracing C:\Users\Student\Documents\SA RayTracing\RayTracing\RayTracing\Walnut.lib(Input.obj) 1

I don't know what to do, I haven't found anything helpful online either.