nvpro-samples / vk_raytrace Goto Github PK

Hello. Please explain how rayquery variant of renderer works.
Created acceleration structure and used or so?
Where is no fully compute renderer with self-made bvh, not use vulkan acceleration structure?

Hello,

During compilation with VS2022 of vk_raytrace.exe I ended with an error I can't fix.

\vk_raytrace\out\build\x64-Debug\nvml.lib(nvml.dll) : error LNK2001: symbole externe non résolu __delayLoadHelper2

I tried to add

C:\Program Files\NVIDIA Corporation\NVSMI\nvml.dll
C:\Program Files\NVIDIA Corporation\NVSMI\nvml.lib

I have installed latest CUDA 12.2
https://developer.nvidia.com/gpu-deployment-kit says :
The GPU Deployment Kit is part of the CUDA Toolkit, in versions 8 and later, and is no longer available as a separate download.

I do not know how to proceed

I notice some problems in pathtrace.glsl. I am sorry I have no time to fix them myself. If these problems exist, I hope this issue post can help you out.

1. The shadow rays of point light sources are not handled correctly. The maximum distance of the shadow rays should not be 1e32 for them.
2. The environment map evaluation in PathTrace() does not have MIS weight.
3. The direct light contribution is accumulated after RR. So if RR is activated, the direct light of the last bounce may not be added.

Hi, thanks for this greate project!

I encountered a problem which seems like a bug.

Take the robot-toon scene for example, when you zoom in to observe the model, some black lines will occur around the contour, as shown in the figure below.

This will also happen to other contours. Is that a bug?

cmake configuration failed:

$ cmake ..
-- Building for: Ninja
-- The C compiler identification is GNU 12.2.0
-- The CXX compiler identification is GNU 12.2.0
-- Detecting C compiler ABI info
-- Detecting C compiler ABI info - done
-- Check for working C compiler: C:/msys64/mingw64/bin/cc.exe - skipped
-- Detecting C compile features
-- Detecting C compile features - done
-- Detecting CXX compiler ABI info
-- Detecting CXX compiler ABI info - done
-- Check for working CXX compiler: C:/msys64/mingw64/bin/c++.exe - skipped
-- Detecting CXX compile features
-- Detecting CXX compile features - done
-- -------------------------------
-- Processing Project vk_raytrace:
CMake Error at CMakeLists.txt:30 (find_path):
  Could not find BASE_DIRECTORY using the following files:
  nvpro_core/cmake/setup.cmake


-- Configuring incomplete, errors occurred!

Hi there, just to let you know, this sample is hanging for me during startup after the debug statement "Create RtCore" in updatePipeline, specifically : vk_raytrace\rtx_pipeline.cpp (line 134).

The other vulkan KHR ray tracing samples seem to work fine with the mainline nvidia VK drivers, so I figured I'd try it here. No dice.

Hello.
My question is about final image.
Is used float or integer format for image data?

Hi! For which purposes is declared functions
vec3 getPunctualRadianceSubsurface()
vec3 getPunctualRadianceTransmission()
vec3 getPunctualRadianceClearCoat()
vec3 getPunctualRadianceSheen()
but not used?
How to plan use it?

Hello.
I cant find in code

• how to make panel fixed width in pixels
• how to start renderer with floating panel, not docked
• how to save panel states between renderer runs

The thread include is missing in main.cpp:

In file included from /home/chris/lab/ray-tracing/vk_raytrace/main.cpp:32:
/usr/include/vulkan/vulkan.hpp:14218:5: note: declared here
14218 |     operator T const& () const & VULKAN_HPP_NOEXCEPT
      |     ^~~~~~~~
/home/chris/lab/ray-tracing/vk_raytrace/main.cpp: In function ‘int main(int, char**)’:
/home/chris/lab/ray-tracing/vk_raytrace/main.cpp:175:8: error: ‘thread’ is not a member of ‘std’
  175 |   std::thread([&] {
      |        ^~~~~~
/home/chris/lab/ray-tracing/vk_raytrace/main.cpp:46:1: note: ‘std::thread’ is defined in header ‘<thread>’; did you forget to ‘#include <thread>’?
   45 | #include "sample_example.hpp"
  +++ |+#include <thread>
   46 | 

The README.md incorrectly refers to the repo as "vk_raytracing":

git clone https://github.com/nvpro-samples/vk_raytracing.git

It should be "vk_raytrace" instead.

I get a compilation error for just this shader:

1>Generating D:/poojan/projects/nvpro-samples/vk_raytrace/autogen/pathtrace.rgen.spv
1>C:/VulkanSDK/1.2.189.2/bin/glslangValidator.exe -g --target-env vulkan1.2 -o D:/poojan/projects/nvpro-samples/vk_raytrace/autogen/pathtrace.rgen.spv D:/poojan/projects/nvpro-samples/vk_raytrace/shaders/pathtrace.rgen
1>D:/poojan/projects/nvpro-samples/vk_raytrace/shaders/pathtrace.rgen
1>CUSTOMBUILD : error : D:/poojan/projects/nvpro-samples/vk_raytrace/shaders/traceray_rtx.glsl:46: 'no rayPayloadEXT/rayPayloadInEXT declared' : with layout(location = 0)
1>CUSTOMBUILD : error : D:/poojan/projects/nvpro-samples/vk_raytrace/shaders/random.glsl:34: '' : syntax error, unexpected UINT, expecting COMMA or SEMICOLON
1>CUSTOMBUILD : error : 2 compilation errors. No code generated.

Ofcourse, if I don't compile that shader, the sample runs (atleast the compute pipeline works as expected), so I suspect it's probably something different with my validator/version/sdk.

I twiddled around a lot with moving code, etc. No luck. I can't seem to understand what is so different than say the ray tracing gltf sample (all the samples except this one compile/run fine incidentally)

Any ideas?

The cfgmgr.h is a windows-specific dependency preventing compilation on Linux.

[ 86%] Building CXX object CMakeFiles/vk_raytrace.dir/home/chris/lab/ray-tracing/shared_sources/imgui/imgui_camera_widget.cpp.o
[ 88%] Building CXX object CMakeFiles/vk_raytrace.dir/home/chris/lab/ray-tracing/shared_sources/imgui/imgui_impl_vk.cpp.o
In file included from /home/chris/lab/ray-tracing/vk_raytrace/sample_example.cpp:55:
/home/chris/lab/ray-tracing/vk_raytrace/nvml_monitor.hpp:19:10: fatal error: cfgmgr32.h: No such file or directory
   19 | #include <cfgmgr32.h>
      |          ^~~~~~~~~~~

This works on Jetson nano or xavier?

Please explain how "Remove banding" from shaders/post.frag works:
// Remove banding
uvec3 r = pcg3d(uvec3(gl_FragCoord.xy, 0));
vec3 noise = uintBitsToFloat(0x3f800000 | (r >> 9)) - 1.0f;
color = dither(sRGBToLinear(color), noise, 1. / 255.);

and give please example file to check this code.
I dont see differencies in many scenes

Commit

The issue was: both libdl and pthreads were missing. I am not sure what the optimal route would be in the context of the overall tutorial/example system. But, I got things working with the following change in CMakeLists.txt circa line 151 (the Linkage section):

From:

 #####################################################################################
# Linkage
#
target_link_libraries(${PROJNAME} ${PLATFORM_LIBRARIES} shared_sources)

To:

#####################################################################################
# Linkage
#
find_package(Threads REQUIRED)
set(PROJECT_LIBRARIES ${CMAKE_DL_LIBS} -lpthread)
target_link_libraries(${PROJNAME} ${PLATFORM_LIBRARIES}  ${PROJECT_LIBRARIES} shared_sources)

←[0mVulkan Version:
←[0m - available: 1.3.268
←[0m - requesting: 1.3.0
←[0m______________________
←[0mUsed Instance Layers :
←[0m
←[0mUsed Instance Extensions :
←[0mVK_KHR_surface
←[0mVK_KHR_win32_surface
←[0mVK_EXT_debug_utils
←[0m____________________
←[0mDevices : 2
←[0m0: NVIDIA GeForce RTX 3060 Laptop GPU
←[0m - Compatible
←[0m1: Intel(R) UHD Graphics
←[0m - Missing extensions: ←[0mVK_KHR_acceleration_structure ←[0mVK_KHR_ray_tracing_pipeline ←[0mVK_KHR_deferred_host_operations ←[0m
←[0mCompatible physical devices found : ←[0m1
←[0mUsing Device:
←[0m - Device Name : NVIDIA GeForce RTX 3060 Laptop GPU
←[0m - Vendor : NVIDIA
←[0m - Driver Version : 551.92.0
←[0m - API Version : 1.3.271
←[0m - Device Type : Discrete GPU
←[0m________________________
←[0mUsed Device Extensions :
←[0mVK_KHR_swapchain
←[0mVK_KHR_shader_clock
←[0mVK_KHR_acceleration_structure
←[0mVK_KHR_ray_tracing_pipeline
←[0mVK_KHR_ray_query
←[0mVK_KHR_deferred_host_operations
←[0mVK_KHR_buffer_device_address
←[0m
←[0mCreate Offscreen←[0m --> (5.602 ms)
←[0mLoading HDR and converting C:/msys64/home/user/from-git/nv-feb24/bin_x64/Release/../../downloaded_resources//std_env.hdr
←[0m --> (57.192 ms)
←[0mLoading scene: C:/msys64/home/user/from-git/nv-feb24/bin_x64/Release/../../downloaded_resources//robot_toon/robot-toon.gltf←[0m --> (99.653 ms)
←[0m←[33m←[0mConvert to internal GLTF←[0m --> (61.516 ms)
←[0mCreate Buffers
←[0m - Create 18 Material Buffer←[0m --> (0.020 ms)
←[0m - Create 2 Textures, 2 Images←[0m --> (0.734 ms)
←[0m - Create 50 Vertex Buffers←[0m --> (94.095 ms)
←[0m ←[0m --> (6.171 ms)
←[0mCreate acceleration structure
←[0m BLAS(50)←[0mRT BLAS: reducing from: 64887040 to: 25686016 = 39201024 (60.41% smaller)
←[0m TLAS(52)←[0m --> (35.301 ms)
←[0mSwitching renderer, from 2 to 0
←[0mCreate RtxPipeline←[0m --> (17.305 ms)
←[0m

Ran into this error compiling the shaders:

[ 53%] Linking CXX static library /home/chris/ray-tracing/bin_x64/libshared_sources.a
[ 53%] Built target shared_sources
[ 55%] Generating ../autogen/shaders/post.frag.h
/usr/bin/glslangValidator -o autogen/shaders/post.frag.h -V /home/chris/ray-tracing/vk_raytrace/shaders/post.frag --vn post_frag --target-env vulkan1.2
/home/chris/ray-tracing/vk_raytrace/shaders/post.frag
ERROR: Failed to open file: autogen/shaders/post.frag.h
[ 57%] Generating ../autogen/shaders/passthrough.vert.h
/usr/bin/glslangValidator -o autogen/shaders/passthrough.vert.h -V /home/chris/ray-tracing/vk_raytrace/shaders/passthrough.vert --vn passthrough_vert --target-env vulkan1.2
/home/chris/ray-tracing/vk_raytrace/shaders/passthrough.vert
ERROR: Failed to open file: autogen/shaders/passthrough.vert.h
[ 59%] Generating ../autogen/shaders/pathtrace.rahit.h
/usr/bin/glslangValidator -o autogen/shaders/pathtrace.rahit.h -V /home/chris/ray-tracing/vk_raytrace/shaders/pathtrace.rahit --vn pathtrace_rahit --target-env vulkan1.2
/home/chris/ray-tracing/vk_raytrace/shaders/pathtrace.rahit
ERROR: /home/chris/ray-tracing/vk_raytrace/shaders/pathtrace.rahit:88: 'ignoreIntersectionEXT' : undeclared identifier 
ERROR: /home/chris/ray-tracing/vk_raytrace/shaders/pathtrace.rahit:88: '' : compilation terminated 
ERROR: 2 compilation errors.  No code generated.

GLSL is installed:

AMDGPU Packages 
Installed Packages
glslc.x86_64  2020.4-1.fc33
Available Packages
glslc.x86_64  2020.2-1.fc33  
glslc.x86_64  2020.4-1.fc33  

GLSL Lang is @11.0

Installed Packages
glslang.x86_64 11.0.0-1.20200803.git5743eed.fc33

I cant find credits of used method in source codes

Declared as needed in line