12 — Game Development, GPU Computing & ML Pipelines

T E C H N I C A L B R I E F I N G Game Development, GPU Computing and ML Delivery Engines, real-time graphics, GPU compute and shipping practice for working engineers Eight sections · 110 slides · Conceptual and reference depth By Md Ahbab Hamid Khan

Agenda 01 The Production Pipeline Phases, milestones, and why games are not web 02 Unreal Engine Layout, Blueprints vs C++, framework, Lumen, Nanite 03 Unity Scenes, prefabs, lifecycle, DOTS, Addressables, IL2CPP 04 C, C++ and Python RAII, allocators, cache layout, CMake, tooling 05 Graphics and the GPU Pipeline, ray tracing, shaders, LODs, frame budgets 06 GPU Compute and ML CUDA, PyTorch, ONNX, in-game inference 07 Art and Asset Pipelines DCC tools, FBX/glTF/USD, LFS and Perforce 08 Shipping and Live-Ops Netcode, servers, submission, telemetry, QA AGENDA 2

Assumed prerequisites Assumed ▪ Fluency in at least one language, and comfort reading another ▪ Functions, classes, interfaces, generics and asynchronous execution ▪ HTTP, client-server architecture and basic distributed systems ideas ▪ Git branching, merging, pull requests and continuous integration Not assumed — defined here on first use ▪ Any engine, rendering, GPU, shader or netcode background ▪ Linear algebra beyond vectors, matrices and dot products ▪ Machine learning theory beyond gradient descent as a concept PREREQUISITES 3