Hardware-Accelerated GPU Scheduling (HAGS) is a Windows feature that shifts GPU work scheduling from the CPU to the GPU itself. It has been available since Windows 10 version 2004 and is often cited as a way to reduce input lag in games. The reality is more nuanced: HAGS delivers a measurable latency improvement on modern hardware, but the raw FPS change is typically small — and on older GPUs it can introduce instability. This guide covers exactly what HAGS does, which hardware benefits, and how to configure it correctly. For a full overview of Windows and driver-level optimisations, see the complete PC game settings optimisation guide.
What Hardware-Accelerated GPU Scheduling Does
In the traditional Windows GPU scheduling model, the CPU’s VRAM manager is responsible for scheduling work packets to the GPU. Every time a game submits a draw call, the CPU aggregates these packets, schedules them, and hands them off to the GPU driver. This process introduces a small but measurable scheduling overhead — typically 1–3 ms of additional CPU-side latency per frame.
With HAGS enabled, the GPU’s own hardware scheduler takes over this work. The GPU receives work packets directly and manages its own execution queue without waiting for the CPU scheduler. The result is a reduction in scheduling latency and, in many cases, more consistent frame delivery. The feature requires WDDM 2.7 (Windows Display Driver Model 2.7) and a compatible GPU — specifically any NVIDIA RTX 20 series or newer, any AMD RX 5000 series or newer, or any Intel Arc GPU.
What HAGS does not do is significantly increase raw render performance. The GPU still runs the same workload at the same speed — what changes is how efficiently that workload is delivered to it.
Should You Enable or Disable HAGS?
The answer depends on your GPU generation. Use this table as your quick reference:
| GPU Generation | Recommendation | Reason |
|---|---|---|
| NVIDIA RTX 40 series (Ada Lovelace) | Enable | Full HAGS support; works with NVIDIA Reflex for best latency |
| NVIDIA RTX 30 series (Ampere) | Enable | Stable, small latency benefit; no significant downsides |
| NVIDIA RTX 20 series (Turing) | Enable (test first) | Generally stable but some older driver combos cause instability |
| NVIDIA GTX 16 series | Disable | No hardware scheduler; HAGS adds overhead without benefit |
| NVIDIA GTX 10 series and older | Disable | Not supported; can cause crashes or black screens |
| AMD RX 7000 / RX 6000 (RDNA 3/2) | Enable | Full HAGS support; pairs well with AMD Anti-Lag+ |
| AMD RX 5000 (RDNA 1) | Enable (test first) | Supported but some titles show marginal instability on RDNA 1 |
| AMD RX 580 / Vega and older | Disable | Not supported; driver emulation adds latency instead of reducing it |
| Intel Arc (all generations) | Enable | Intel Arc was designed with HAGS in mind; clear benefit on Arc A and B series |
FPS Impact: What to Actually Expect
Independent benchmarks consistently show that HAGS produces a FPS change of approximately −2% to +3% depending on the game and GPU. In most titles on modern hardware, the change is within margin of error. HAGS is not a performance optimisation — it is a latency optimisation. Frame times become more consistent, and the average input lag (the time between a mouse movement and the resulting frame appearing on screen) drops by approximately 1–3 ms.
That 1–3 ms reduction may sound negligible, but it compounds with other latency-reducing features. On NVIDIA hardware, HAGS is a prerequisite for getting the full benefit from NVIDIA Reflex. On AMD hardware, HAGS works alongside AMD Anti-Lag+ in supported titles. For competitive gaming — where reducing every millisecond of input latency matters — HAGS should always be enabled on modern GPUs alongside these driver-level latency tools.
For a full breakdown of which in-game and driver settings affect frame timing and input latency, see the game settings explained guide.
Which Games Benefit Most from HAGS
HAGS delivers the most noticeable benefit in DirectX 12 and Vulkan titles, where the scheduling overhead from the traditional model is highest. Games built on DX12 or Vulkan submit higher volumes of smaller draw calls compared to DX11 games, which means the CPU scheduler is invoked more frequently — and HAGS’ latency reduction is proportionally larger.
DirectX 11 titles show minimal HAGS impact. The traditional CPU scheduling model is well-optimised for DX11’s rendering patterns, so shifting the scheduler to the GPU provides little benefit. Titles like older Counter-Strike builds or legacy MMOs on DX11 are unlikely to benefit.
Some games have historically had compatibility issues with HAGS enabled. Notably, certain Destiny 2 builds and older Halo Infinite versions exhibited shader compilation stutters with HAGS on. If you notice intermittent stuttering after enabling HAGS in a specific title, disable it and test — the issue is usually title-specific rather than a system-wide problem.
How to Enable or Disable HAGS in Windows
- Open Windows Settings (Win + I)
- Go to System → Display
- Scroll down to Graphics and click it
- Click Change default graphics settings
- Toggle Hardware-accelerated GPU scheduling on or off
- Restart your PC — the change does not apply until after a full reboot
If the toggle is greyed out, your GPU driver does not support HAGS or your driver is outdated. Update to the latest GPU driver from the NVIDIA, AMD, or Intel website and check again. On laptops with switchable graphics (integrated + discrete), the toggle controls both GPUs — enabling it applies to the discrete GPU when active.
HAGS and NVIDIA Reflex
NVIDIA Reflex is the most impactful latency-reduction tool available on RTX hardware, and it works best when HAGS is enabled. Reflex operates by optimising the render queue and reducing the CPU render-ahead — the number of frames the CPU prepares before the GPU processes them. When HAGS is active, the GPU’s scheduler handles work delivery more efficiently, giving Reflex a more consistent pipeline to optimise. The result is a lower and more stable system latency measurement in the NVIDIA Reflex Latency Analyser.
In competitive titles that support both HAGS and NVIDIA Reflex — including Valorant, Apex Legends, Fortnite, and Overwatch 2 — enabling HAGS is one part of a two-step latency configuration: enable HAGS in Windows settings, then enable Reflex (or Reflex + Boost on lower-end GPUs) in-game.
Frequently Asked Questions
Does enabling HAGS increase FPS?
Not meaningfully in most games. Expect 0–3% FPS change on modern hardware — often within benchmark noise. The real benefit is lower and more consistent input latency, not raw frame rate. If you are chasing FPS, focus on in-game settings, resolution scaling, and DLSS/FSR before adjusting HAGS.
Can HAGS cause stuttering or crashes?
On unsupported GPUs (GTX 10 series, older AMD cards), yes — HAGS should be disabled on those. On supported modern hardware, HAGS is stable for the vast majority of users. If you experience crashes after enabling HAGS, first update your GPU driver. If instability persists, disable HAGS and check for game-specific reports in the title’s community forums.
Should I enable HAGS for competitive gaming in 2026?
Yes, on any supported GPU. Competitive gaming is where HAGS makes the clearest case for itself — it reduces input latency and works synergistically with NVIDIA Reflex and AMD Anti-Lag+. Enable HAGS, then enable your GPU’s in-game latency feature, and cap your frame rate slightly below your monitor’s refresh rate for the optimal competitive configuration.
Sources
- Microsoft DirectX Developer Blog. Hardware Accelerated GPU Scheduling — how it works and what it changes. Microsoft Corporation.
- NVIDIA. NVIDIA Reflex Low Latency Platform — System Latency Explained. NVIDIA Corporation.
- Tom’s Hardware. GPU Hierarchy 2026 — Benchmark and Performance Rankings. Future Publishing.