Performance of Parallel Jobs, Burst Compiler and Pointers in a Real Game Setting

After some comments on my recent culling video, and discussions about the performance of different approaches to threading, I thought it would be good to run some empirical tests to compare the results. I did not want to run the tests in isolation, but rather as part of an actual, complex, real-game setting, where lots of other calculations happen before, after, and in-between the tests. This highlights a few very interesting factors to keep in mind when planning a multi-threaded approach. Let’s have a look. This is the story of implementing what I've learned into a gridless, interactive, world-building sandbox game called Minor Deity. Minor Deity on Steam: https://store.steampowered.com/app/3876240/Minor_Deity/ Minor Deity Discord Server: https://discord.gg/2NEb4HxwhF 00:00 Introduction 01:16 Context is Important 02:08 Multi-threaded Native Arrays 03:30 Impact of Memory Caches 04:46 Pointers 05:34 One Job Per Frame 06:50 Setting up the Jobs 08:33 A More Complex Job 09:21 Outro