AMD Anti-Lag is a driver-level feature designed to reduce input latency by managing when the CPU submits work to the GPU. Understanding how it works helps you combine it correctly with system-level tweaks.
How Anti-Lag works
Normally, the CPU runs ahead of the GPU — it submits several frames worth of work to the GPU queue in advance. This keeps the GPU fed and maximizes throughput, but means the GPU is rendering frames based on slightly stale game state.
Anti-Lag throttles the CPU submission rate to keep the CPU-GPU distance minimal. The GPU renders frames based on more recent game state, reducing the gap between your input and what appears on screen.
Anti-Lag vs Anti-Lag+
Anti-Lag+ (introduced later) works at the game engine level rather than the driver level, giving more precise timing. It requires game integration.
Combining with system tweaks
Anti-Lag works at the driver/flip queue level. System tweaks like Win32PrioritySeparation, MMCSS priority, and process priority work at the OS scheduler level. These are complementary — they address different layers of the latency stack.
MPO and Anti-Lag: ensure MPO is disabled (OverlayTestMode=5) when using Anti-Lag. MPO can interfere with Anti-Lag's frame timing on some driver versions.
HAGS and Anti-Lag: HAGS and Anti-Lag both touch GPU scheduling. On Navi GPUs (where HAGS causes issues), keep HAGS disabled even if Anti-Lag is enabled.
Enable Anti-Lag in Radeon Software → Gaming → Global Settings → Radeon Anti-Lag.
System-level latency tweaks at https://terweb.lt/
