Practical articles on CPU, GPU, memory and engineering process optimization.
Lazy shader compilation is a launch-day stutter trap. This module shows how to catch CreateGPUProgram in a real build, when ShaderVariantCollection is not enough, and how to prewarm the right variants during startup or a loading screen before players ever feel the hitch.
Asset loading is a three-stage problem, not a mystery. Learn how to profile CPU Usage, File Access, and Asset Loading together, where Awake and asset integration create hitches, and how to use backgroundLoadingPriority and lightmap discipline to cut real load pain.
Decals can make a scene look shipped or quietly wreck your frame time. This module shows why BIRP Projector does not scale, when URP DBuffer vs Screen Space makes sense, and when decal stamping is the smarter call if the result does not need to move.
Heat and battery drain are a retention problem. This module shows how to measure power at the wall, inspect CPU activity with Battery Historian, and cut waste with simple Unity-side moves like lower frame rates and rendering on demand.
A solid build pipeline for your projects that actually helps you save 20h/week or more. Is this too much of an ask? Look inside.
In this blog post I will show you how to use object pooling in Unity 2021 with this new official API so you don't have to mess with 3rd party code that breaks on every Unity upgrade.
Compute Shaders are the move you make when CPU-side optimization is still not enough. This module shows the real Tape to Tape use case, the API/support gates, the SIMD mindset, and the buffer and readback decisions that separate a real GPU win from a slow science project.
Manual Unity builds are a delivery tax. This module shows the GitLab CI plus Docker plus Game CI path, why a Linux VPS beats blocking your workstation, and how to get from commit to build artifact and team notifications without the usual pipeline mess.
Static batching will not always reduce your draw call count. But it will always do something else that is highly beneficial for your game
High-performance collections can cut managed overhead, but only if you stop treating NativeList, Burst, safety checks, and allocators like magic. This module shows when native containers win, when they lose, and how to make the switch without fooling yourself.
Cannot have many real-time lights with Forward Shading in URP... can you? (obviously yes)
Forward+ in URP is not a magic switch. This module shows where classic Forward breaks at 8 real-time lights per object, when Forward+ starts to make sense, and why mobile memory bandwidth can still make deferred the wrong call.