How to Use NVIDIA Reflex and AMD Anti-Lag for Lower Latency
Key Takeaway
- They eliminate the render queue: Both technologies prevent the CPU from preparing frames faster than the GPU can render them, eliminating the "backpressure" that causes input lag.
- NVIDIA Reflex is superior: Reflex is integrated directly into the game engine, allowing for more precise timing and lower latency than driver-level solutions.
- Crucial for GPU-bound scenarios: These technologies provide the most massive latency reductions when your GPU is maxed out at 99% utilization.
- Enable "Reflex + Boost": If your game supports it, "On + Boost" prevents the GPU from downclocking during CPU-bound moments, keeping latency consistently low.
You bought a 240Hz monitor, a high-end mouse, and a powerful GPU. You are getting 200 frames per second, but your aim still feels slightly "floaty" or disconnected. The culprit is likely the render queue. To achieve the absolute lowest input lag and response time, you need to manage how your CPU and GPU talk to each other. This is where NVIDIA Reflex and AMD Anti-Lag come in.
The Problem: The Render Queue
In a traditional rendering pipeline, the CPU prepares frames and sends them to the GPU. If the CPU is faster than the GPU (a GPU-bound scenario), the CPU will queue up several frames in advance. This ensures the GPU always has work to do, maximizing frame rates.
However, this queue is terrible for latency. If you click your mouse, that input is applied to the frame currently being prepared by the CPU. But that frame has to wait in line behind 2 or 3 other frames before the GPU renders it and sends it to your monitor. This creates noticeable input lag.
NVIDIA Reflex: The Engine-Level Solution
NVIDIA Reflex is the gold standard for latency reduction. Instead of operating at the driver level, Reflex is integrated directly into the game's code by the developers (available in games like Valorant, Apex Legends, Overwatch 2, and CS2).
Reflex dynamically paces the CPU so that it only prepares a frame exactly when the GPU is ready to render it. This completely eliminates the render queue. The moment you click, the frame is generated and rendered immediately.
Reflex "On" vs. "On + Boost"
- On: Eliminates the render queue. Best for most situations.
- On + Boost: Eliminates the queue AND forces the GPU to maintain high clock speeds even in CPU-bound scenarios. This prevents the GPU from going to sleep and adding latency when the action suddenly gets intense. Use this if your thermals allow it.
AMD Radeon Anti-Lag
AMD's Anti-Lag operates similarly but traditionally works at the driver level rather than the game engine level. It monitors the pace of the CPU and GPU and attempts to align them to reduce the queue.
While highly effective (especially in DirectX 11 games), it generally cannot match the precision of NVIDIA's engine-integrated Reflex. However, AMD has recently introduced Anti-Lag 2, which, like Reflex, integrates directly into game engines (starting with CS2) to provide comparable latency reductions.
When Do You Need Them Most?
Reflex and Anti-Lag provide the most dramatic benefits when you are GPU-bound (your GPU usage is at 95-99%). In these scenarios, they can cut system latency in half.
If you are heavily CPU-bound (e.g., playing CS2 at 1080p low settings with an RTX 4090), the render queue is already empty because the GPU is waiting on the CPU. In this case, Reflex won't reduce latency much, but "Reflex + Boost" will keep your GPU clocks high for consistency.
Frequently Asked Questions
Does NVIDIA Reflex lower FPS?
Generally, no. It might cause a negligible 1-2% drop in average FPS in some titles, but the massive reduction in input lag makes the game feel significantly smoother and more responsive.
Should I use Ultra Low Latency Mode (ULLM) in the NVIDIA Control Panel?
If a game supports NVIDIA Reflex natively, use Reflex. Reflex overrides ULLM and is much more effective. Only use ULLM for older games that do not have Reflex built-in.
Can I use AMD Anti-Lag on an NVIDIA GPU?
No, Anti-Lag is an exclusive feature of the AMD Radeon driver software and requires an AMD graphics card.
Does Reflex work with V-Sync?
Yes. If you must use V-Sync (preferably combined with G-Sync), enabling Reflex will automatically cap your frame rate slightly below your monitor's refresh rate to prevent V-Sync latency from kicking in.
Is Anti-Lag 2 better than Anti-Lag 1?
Yes. Anti-Lag 2 is integrated into the game engine (like Reflex), allowing for much tighter synchronization and lower latency than the driver-level Anti-Lag 1.
How do I test my system latency?
While true system latency requires specialized hardware (like NVIDIA LDAT), you can use our Refresh Rate Test below to ensure your browser and monitor are communicating as fast as possible.
Ready to test your newly optimized, low-latency setup?
Is your monitor performing as advertised?
Don't just trust the box. Verify your true refresh rate and check for frame skips.
Run the Refresh Rate Test NowRefreshRateTest Engineering Team
A specialized collective of display hardware researchers and low-latency engineers dedicated to providing objective performance metrics for the high-refresh rate era.