calculate-loan

01 Loan payments

Advanced Accelerated GPU Scheduling Explained

Guide 01-14495 · Updated · Step by step

Advanced accelerated GPU scheduling is a Windows graphics feature that lets a compatible GPU manage more of its own video-memory scheduling, reducing some work the operating system would otherwise perform. It can lower CPU overhead and, in some workloads, improve responsiveness or frame-time consistency, but the benefit depends on the hardware, driver, and application.

What GPU scheduling means

A graphics processor runs many operations concurrently, while games and other programs submit commands for it to execute. The operating system and graphics driver coordinate those commands, decide how GPU resources are shared, and manage video memory. Scheduling matters because several processes may request GPU time, and because commands and data have to move between system memory and dedicated graphics memory.

Microsoft introduced Hardware-accelerated GPU scheduling (HAGS) with the Windows 10 May 2020 Update, version 2004. The feature shifts some high-frequency scheduling work from the CPU to a GPU-based scheduling processor. “Advanced accelerated GPU scheduling” is not a separate, standardized Windows feature name; it is often a search phrase for HAGS, newer scheduling options in graphics software, or general GPU scheduling techniques.

How hardware-accelerated scheduling works

In a conventional arrangement, the CPU and operating system handle much of the work involved in organizing GPU tasks and managing their submission. With HAGS enabled, compatible hardware takes over some of that scheduling responsibility. Windows still coordinates graphics work; the feature does not remove the operating system or driver from the process, nor does it make the GPU execute every task independently.

The potential advantage is reduced scheduling overhead. In a situation where the CPU spends meaningful time managing graphics work, shifting part of that burden may free CPU resources or reduce latency. The effect is usually small, however, and it can be difficult to notice when a game is limited by GPU rendering, memory bandwidth, or another part of the system.

GPU scheduling is also distinct from features such as variable refresh rate, frame generation, or a game’s low-latency mode. Those may affect how frames are displayed or queued, but they are not the same mechanism as Windows HAGS.

Requirements and where to find the setting

HAGS requires a supported Windows version, compatible graphics hardware, and a suitable driver. Availability varies with the graphics adapter and driver; a setting missing from Windows is not necessarily a fault. Updating Windows and installing a driver from the graphics-device manufacturer may add support, but an update cannot make unsupported hardware compatible.

On supported Windows installations, look under Settings > System > Display > Graphics > Change default graphics settings. The control may be labelled “Hardware-accelerated GPU scheduling.” Its exact location can vary by Windows version. Turning it on or off generally requires a restart before the change takes effect.

  • Check that Windows is up to date and identify the installed graphics adapter.
  • Install a compatible graphics driver, then restart if requested.
  • Change the scheduling setting, restart, and test the same workload under comparable conditions.

What performance changes to expect

There is no universal performance gain. Results depend on the game or application, driver, CPU and GPU combination, display settings, and whether the workload is CPU- or GPU-limited. Some systems show little measurable difference. In others, a benchmark may show changes in average frame rate or frame-time behaviour, but results from one machine do not establish what another machine will experience.

For example, a game with a CPU bottleneck may be more sensitive to scheduling overhead than a visually demanding game that keeps the GPU fully occupied. Conversely, if the graphics card is already the limiting factor, changing how work is scheduled may not increase frame rate. A benchmark that reports only average frames per second can also miss occasional stutters; frame-time graphs and repeatable play sessions provide useful context.

To compare settings, use the same game scene, resolution, graphics options, background applications, and driver. Run multiple passes and compare averages as well as frame-time consistency. Avoid changing several settings at once, since that makes the cause of any difference unclear.

Compatibility, troubleshooting, and trade-offs

HAGS is an optional scheduling path, not a general-purpose speed switch. Driver and application behaviour can differ, and a change may introduce instability or worse performance on a particular configuration. If crashes, visual glitches, stuttering, or unusual performance begin after enabling it, restart and test with the setting disabled. If the problem remains, investigate recent driver or application changes separately.

When the control is unavailable, check the Windows version, graphics driver, and adapter support. Laptops with both integrated and discrete graphics can have extra driver and power-management considerations. On such systems, verify which adapter is running the application before drawing conclusions from a benchmark.

HAGS does not replace good driver maintenance, fix inadequate cooling, increase the physical amount of video memory, or guarantee a higher frame rate. Nor does enabling it by itself improve every graphics application. Treat it as a configuration option whose value must be checked on the system and software you actually use.

HAGS compared with other scheduling options

Windows scheduling coordinates access to the GPU across applications. Graphics APIs such as DirectX 12 and Vulkan give applications more explicit control over aspects of graphics work than older APIs, but they do not simply turn HAGS on. An application’s API, engine, and driver all influence how work is submitted and managed.

Some GPU vendors also provide driver-level controls for queueing or latency. These controls can have overlapping goals, but their names and effects differ. Do not assume that a “low latency” or “maximum performance” option is equivalent to hardware-accelerated scheduling; consult the driver documentation and test settings individually.

Frequently Asked Questions

Does hardware-accelerated GPU scheduling increase FPS?
It can affect performance in some configurations, but it does not guarantee a frame-rate increase. The result depends on whether scheduling overhead is a meaningful limit in the workload.

Should I turn HAGS on or off?
There is no setting that is right for every computer. If it is supported, compare the same workload with it enabled and disabled, and keep the option that behaves reliably for your use.

Why can’t I see the GPU scheduling setting?
Your Windows version, graphics adapter, or installed driver may not support the feature, or the setting may be located differently in that Windows release. Check for compatible updates and consult the device documentation.

Does HAGS reduce input lag?
It may change scheduling overhead in some circumstances, but it is not a guaranteed input-lag fix. Input latency also depends on the game, frame queue, display, peripherals, and other system settings.

Is HAGS the same as GPU hardware acceleration?
No. Hardware acceleration is a broad term for using a GPU to perform certain tasks. HAGS refers specifically to Windows shifting some GPU scheduling work to compatible GPU hardware.

Related

General steps for current mainstream versions; menus can differ by device and version. See the disclaimer or report an error.

More in this section

The whole Loan payments section →