2
0
Fork 0
mirror of https://github.com/burgerindividual/hypercull synced 2025-12-10 10:02:58 +00:00
Replaces all chunk culling with a native tile-based replacement https://github.com/burgerindividual/hypercull
Find a file
2025-08-04 13:43:38 -04:00
buildSrc add partial Iris support, add debug info 2025-08-03 20:02:47 -04:00
common add partial Iris support, add debug info 2025-08-03 20:02:47 -04:00
fabric add partial Iris support, add debug info 2025-08-03 20:02:47 -04:00
gradle/wrapper initial working commit 2025-07-16 10:57:28 -04:00
native rewrite frustum test_box documentation 2025-08-04 13:43:38 -04:00
neoforge add partial Iris support, add debug info 2025-08-03 20:02:47 -04:00
.gitattributes Initial commit 2025-06-19 13:25:52 -04:00
.gitignore add partial Iris support, add debug info 2025-08-03 20:02:47 -04:00
build.gradle Move rust code to this repository, set up gradle to compile it with gradle-rust 2025-07-17 17:47:03 -04:00
gradle.properties add partial Iris support, add debug info 2025-08-03 20:02:47 -04:00
gradlew Initial commit 2025-06-19 13:25:52 -04:00
gradlew.bat Initial commit 2025-06-19 13:25:52 -04:00
README.md Slightly change readme 2025-07-17 17:50:36 -04:00
settings.gradle Move rust code to this repository, set up gradle to compile it with gradle-rust 2025-07-17 17:47:03 -04:00

HyperCull

An attempt at a faster culling algorithm for Minecraft, built to slot into the existing architecture of the Sodium world renderer

This repository contains an implementation of tile-based culling, which was initially theorized by JellySquid. While only occlusion culling was considered in the original theory, this has been expanded to include tile-based implementations of fog culling, angle culling, and frustum culling successfully. Combined, the number of sections culled should be on-par with the current Java implementation, while being much more efficient.

Building

This project is not ready to be tested!

However, if you still would like to build the project, the following instructions are provided:

// TODO

Native Compiled Platforms:

linux-x64-avx2+fma
macos-x64-avx2+fma
windows-x64-avx2+fma
linux-x64-sse4_1+ssse3
macos-x64-sse4_1+ssse3
windows-x64-sse4_1+ssse3
linux-arm64
macos-arm64
windows-arm64

The Java code which loads the native library will automatically choose the best option of the above list for the system it's running on.

How it works

(TODO)