Volume 04 / Unity & Performance

The Indie Game Optimization Handbook

Name: The Indie Game Optimization Handbook
Author: Vertex Arc

Build Bigger Worlds Without Destroying Performance

Find the actual bottleneck and make measured improvements. A practical Unity field guide to rendering, simulation, memory, loading, and platform budgets.

Length: 19 pages
Reading time: 17 minutes
Format: PDF · 118 KB
Price: Free

Download PDF ↓Read a preview →

First edition · October 2026 · Published by Vertex Arc

Inside the guide

CPU and GPU profilingRendering and visibilityMemory and loadingMobile, WebGL and PC

Also includes an introduction, practical exercises, a final checklist, and a conclusion.

Introduction

Optimization is the process of changing a measured cost while preserving the behavior and presentation that matter. It is not a collection of switches that always improve performance. A technique can help one project, do nothing in another, or move work into a different bottleneck. Start with a specific build, device, scenario, and target.

This book follows an original hypothetical Unity project called Relay Valley: a compact exploration game with an outdoor route, a busy machinery yard, and an interior control room. The example gives us different visibility, simulation, and loading conditions. Any numbers used to explain budgets are illustrative, not measured results from a shipped Vertex Arc game.

Keep three artifacts: a repeatable test route, a baseline capture, and a change log. Record build version, resolution, quality settings, device, and test conditions. Compare changes under similar conditions. A faster run on a cooler phone or a different scene state is not a fair comparison.

The workflow is simple to state: reproduce, measure, form a hypothesis, change one cause, and verify. The difficult part is resisting a familiar optimization before you know whether it addresses the problem. Use this book as a diagnostic companion. Read the chapter that matches the evidence, then return to the same test and check the result.

Read chapter 01

01 / Profile before optimizing

The idea

Describe the symptom precisely. Does the game run slowly throughout an encounter, hitch when a new effect appears, or consume more memory after every transition? These are different problems. Capture the event on the intended platform where possible. The Unity Editor performs additional work, and development instrumentation can also affect results.

Use frame time alongside frame rate. A 60 FPS target corresponds to roughly 16.7 milliseconds per frame, but an average can conceal occasional long frames. Inspect the timing around visible hitches and distinguish steady costs from first-use costs. Preserve the trace before making changes so later comparisons have a reference.

Example

Relay Valley pauses briefly when the player first enters the machinery yard, then runs smoothly. Reducing every prop's polygon count might lower sustained rendering cost without fixing the entry hitch. The investigation instead captures the transition and separates asset preparation from ongoing simulation. The fix follows the measured event, not the most visually complicated part of the scene.

Key takeaway

Name the scenario and capture the cost before choosing a technique. A repeatable baseline is part of the optimization work.

Try this

Create a one-minute route with a quiet area, a busy encounter, and a transition. Capture it on a target device. Mark the most expensive sustained interval and the longest visible hitch. Write a separate hypothesis for each. Do not change both at once.

Developer note

Deep profiling can add substantial overhead. Use detailed instrumentation to investigate a narrowed question, then validate the final result with representative profiling and a release-like build. Record the capture mode with the result.

Keep reading — download the book ↓