Optimising Data Building In Game Development
summary
Describes building a sparse 3D grid cache for fast spatial queries, a job system using counters to parallelize hundreds of small jobs, and moving file serialization onto its own dependent job to remove lock contention; extracts sort keys into a separate array to cut large-array sort time roughly in half on a 16-core/32-logical test machine; emphasizes output cache and content-hash + binary-determinism for reusing work, uses C++17/20 and profiles with Intel Vtune.
AI-written summary · about
chapters (12)
- 0:00Data building intro
- 1:49What is data building
- 5:02Game code vs data code
- 10:28Minimize build time
- 14:45Parallel processing basics
- 18:45Balanced job sizes
- 20:34Pre-process world data
- 23:26Spatial queries with grid cache
- 34:57Parallel array sorting
- 42:07Avoid thread blocking
- 49:43Parallel init and cleanup
- 55:46Optimization summary
related talks
57mThis Videogame Programmer Used the STL and You Will Never Guess What Happened Next
41mBeyond OOP: A Journey From Cache Misses to MASS Hits
28mRuntime optimization patterns - from games to your application
01h 04mParallel Game Engine Design
01h 02mMaking Games Start Fast: A Story About Concurrency
01h 18mThis Videogame Developer Used the STL and You'll Never Guess What Happened
50mProductivity in C++ Game Development: The Mind of an AAA Game Developer
51mOptimising Games for a Complex World