glad VS GLFW

Just link to GLAD if you're going to link to it! https://github.com/Dav1dde/glad

glew is a GL extension/loading library, OP didn't write glew, personally I use GLAD (https://glad.dav1d.de/) in my projects, which is pretty much the same thing but auto-generated for you based on your project requirements.

Just use a library like glad which loads everything at runtime under the hood (even system OpenGL runtime, no need to link anything). For a quick start, there is a website to generate headers and a single glad.c to put in your project: https://glad.dav1d.de/

I installed glfw3 through vcpkg on Windows and used the GLAD generator at https://glad.dav1d.de/ to generate a loader for OGL 4.6. The tile renderer example builds, but it just displays a black window without drawing anything.

This is a personal thing, but I write all my code starting off using OpenGL with GLFW as my main library and GLAD as my plugin loader.

Coming from C++ you may have heard of glad before, the glad2 branch also supports generating OpenGL bindings for Rust.

Going parallel on the GPU is where you see these massive performance increases. OpenGL's GLSL shader pipeline is great for that, but if you decide to build up your own OpenGL back-end from scratch it's a steep learning curve. It's really well documented though. I would recommend SDL2 + GLAD for window management, context creation, and extension wrangling.

Even in C++, you have to generate the right OpenGL headers for the GL version you want to use. I use GLAD, but sadly I don't see an option to use it for C# bindings. There isn't really any magic involved, though; OpenGL publishes machine-readable lists of C functions and symbols so just about any software can access the full power of OpenGL using those (assuming their video drivers are up to snuff). There is also a small bit of code needed to create the GL context but it's pretty simple.

glad

Hey there. I'm learning opengl and using cmake too. I just got my build with a few of the same libraries working a week ago and maybe I can help. I wouldn't worry about using vcpkg with glad. It's just a simple header and source file combo and can be copied to your source/include directories. It's much better if you generate your own on their website, because you can specify what versions of opengl you need. The generator is satisfyingly simple to use, just select what version of opengl your target is and include the extensions you want. I've just been including everything until I more know what I need. [Link]

I'm not familiar with LWJGL, my 5 seconds on their website makes me think you might be looking for something like GLFW https://www.glfw.org/ to handle I/O and window creation/management.

Consider GLFW3 for windowing and GLAD for function loading. I've used this combination myself successfully. Granted, that is "two things" you need, but OTOH it'll work well.

There is a pattern to it which is usable with every GitHub repository. For example, GLFW v3.3.8 can be fetched from: https://github.com/glfw/glfw/archive/refs/tags/3.3.8.tar.gz

What exactly would I fetch with this method? He's fetching a raylib url, but I don't really see a pattern in it I can follow when fetching glfw. I assume I can use this repository? https://github.com/glfw/glfw

I'm not sure how to fix it, sorry. I searched for some of the messages and found this bug: . I think overriding the LWJGL library might fix the issue, but I don't see any way to do that with the CurseForge launcher.

To go multiplatform with OpenGL / Vulkan (harder for a beginner) you can use a simple window library like GLFW and Dear ImGui for UI. I use these for Avoyd, along with a few other libraries.

cmake_minimum_required(VERSION 3.16) project(P1_PR) add_executable(P1_PR src/main.cpp ) find_package(GLFW QUIET) if(NOT GLFW_FOUND) message(STATUS "GLFW not found, fetching from GitHub...") include(FetchContent) FETCHCONTENT_DECLARE( GLFW GIT_REPOSITORY https://github.com/glfw/glfw/ GIT_TAG 3.3.8 ) FetchContent_GetProperties(GLFW) if (NOT GLFW_POPULATED) set(FETCHCONTENT_QUIET NO) FetchContent_Populate(GLFW) set(BUILD_EXAMPLES OFF CACHE BOOL "" FORCE) set(BUILD_GAMES OFF CACHE BOOL "" FORCE) add_subdirectory(${glfw_SOURCE_DIR} ${glfw_BINARY_DIR}) endif() endif() find_package(GLEW QUIET) if(NOT GLEW_FOUND) message(STATUS "GLEW not found, fetching from GitHub...") include(FetchContent) FETCHCONTENT_DECLARE( GLEW GIT_REPOSITORY https://github.com/Perlmint/glew-cmake GIT_TAG glew-cmake-2.2.0 ) FetchContent_GetProperties(GLEW) if (NOT GLEW_POPULATED) set(FETCHCONTENT_QUIET NO) FetchContent_Populate(GLEW) set(BUILD_EXAMPLES OFF CACHE BOOL "" FORCE) set(BUILD_GAMES OFF CACHE BOOL "" FORCE) add_subdirectory(${glew_SOURCE_DIR} ${glew_BINARY_DIR}) endif() endif() target_include_directories(${PROJECT_NAME} PRIVATE ${GLFW_SOURCE_DIR} PRIVATE ${GLEW_SOURCE_DIR} ) target_link_libraries(${PROJECT_NAME} glfw glew32 opengl32 gdi32 glm ) target_compile_features(${PROJECT_NAME} PRIVATE cxx_std_20)

I recently started following the Learn OpenGL tutorial and noticed that when I create a window with GLFW in the Wayland GNOME session the cursor becomes much larger when hovering over the window. This is because a Wayland client is expected to define its own pointer (cursor) and that seems to lead to inconsistencies between implementations. I do not have a problem with CSD being default on Wayland (I even prefer it when done really well, like Telegram Desktop, or any GTK app with GtkHeaderBar) but the cursor switching scale while moving it over surfaces (windows) is much more jarring.