How to restart graphics drivers in Windows 11

Everyone who’s ever worked on a PC knows that they are extremely susceptible to all kinds of strange problems that drain up to 20% of your time. Some of the most prominent ones that feature in almost every IT support and troubleshooting guide are display and graphics problems like blank screens, artifacts, and resolution issues.

There could be a few reasons for display issues, some of which need some extensive troubleshooting steps, but the quickest method that usually fixes a bunch of issues is to just restart your graphics drivers. Here are the simplest ways to do that.

What to know before restarting your graphics drivers in Windows 11

Restarting graphics drivers in Windows 11 commonly resolves the following IT issues, since they are typically tied to the driver’s display pipeline or memory state:

• Screen flickering
• Slow or unresponsive UI animations
• Graphical glitches (artifacts or corrupted rendering)
• Sudden display blanking

However, problems like persistent driver crashes with Error Code 43, warning lights on the GPU, or artifacts on external monitors despite safe mode may indicate deeper hardware faults such as a failing GPU, VRAM damage, or faulty display cable rather than something a driver restart alone can fix. In those cases, a full diagnostic report or replacement is required.

This means that a graphics driver restart may temporarily stabilize performance when the GPU enters a TDR (Timeout Detection and Recovery) event, but frequent TDRs often signal deeper instability in the driver or GPU hardware. Microsoft’s telemetry standards state that drivers that cause TDRs on more than 0.6% of devices within a 7-day window fail quality thresholds, and five or more TDRs within 60 seconds trigger a system bugcheck that signals serious instability rather than a temporary glitch.

Similarly, restarting can briefly resolve stuttering after waking from sleep or frame drops in 3D apps, but these symptoms often indicate outdated or incompatible drivers, thermal throttling, or VRAM issues.

» Don’t miss the best options for hardware monitoring software

Restarting drivers can mess with your system

Some driver restart methods activate the Timeout Detection and Recovery (TDR) mechanism. This temporarily halts and reloads the GPU driver without rebooting the system. It clears GPU memory, resets display pipelines, and reinitializes DirectX sessions, helping recover from frozen screens, flickering, or blackouts.

However, during this restart, GPU-accelerated applications like Adobe Premiere, CAD software, or games may freeze or crash if they fail to rebind to the refreshed driver. This means that you risk losing unsaved work or experiencing display context corruption, especially in multi-monitor or virtual desktop setups.

Be sure to save work and close GPU-intensive apps (if possible) before triggering the restart.

» Here’s how you can restart a remote computer using Windows

Different manufacturers influence driver restart behavior

Windows 11 uses WDDM (Windows Display Driver Model) as its graphics architecture baseline, but vendors like NVIDIA, AMD, and Intel add proprietary layers that affect restart behavior:

• NVIDIA: NVAPI modules (e.g., G-Sync, CUDA) may delay reinitialization during TDR events.
  
• AMD: Adrenalin drivers can reset secondary display configurations or cause brief UI instability.
  
• Intel: Integrated GPUs often recover faster but may show frame stutter post-restart.

Methods for restarting graphics drivers

Windows 11 offers multiple ways to restart your graphics driver, each suited to different scenarios and levels of system access.

1. Keyboard shortcut: The fastest method

The Ctrl + Win + Shift + B shortcut in Windows 11 activates a graphics driver reset through the Windows Display Driver Model (WDDM). This action briefly pauses and reloads the GPU stack, clears video memory (VRAM), and reboots the display pipeline, helping to resolve issues such as screen freezes, flickering, or unresponsive apps without requiring a full system reboot. you’ll typically hear a short beep and see the screen flash, confirming the reset.

However, this shortcut has limitations in enterprise environments. It won’t fix corrupted drivers, hardware-level faults, or systems where Group Policy blocks local resets. In setups like virtual desktop infrastructures (VDIs) or GPU virtualization, the command may not work at all.

All you need to do is press Ctrl + Win + Shift + B.

» Learn more about virtualization by enabling CPU virtualization in BIOS and using Microsoft Hyper-V

2. Device manager: More control than the keyboard shortcut

This process unloads and reloads the driver kernel, clears GPU memory allocations, and resets active rendering tasks. It gives you more control and visibility since Device Manager shows you exactly which GPU is being restarted, lets you see driver version info, and confirms the device status before and after. It’s a great option if the keyboard shortcut didn’t work.

Follow these steps:

1. Press Windows + X and select Device Manager

2. Expand the “Display adapters” section

3. Right click the GPU or graphics driver you want to restart > Disable device

4. Wait a few seconds

5. Right click the GPU again > Enable device

To ensure stability afterward, run <dxdiag> in PowerShell or Command Prompt, or check Event Viewer System logs for display-related warnings.

It’s also important to verify that the latest certified driver is installed via Windows Update or your vendor’s utility (e.g., GeForce Experience, AMD Adrenalin, Intel Arc Control).

» Here’s how to disable Windows updates and re-enable Windows updates

3. PowerShell/Command prompt: The remote/enterprise method

PowerShell is more complex but offers critical advantages for IT administrators managing graphics driver issues at scale as they can be executed remotely across multiple endpoints. Additionally, every PowerShell action can be logged with timestamps and results, meeting compliance requirements and helping diagnose patterns across your infrastructure.

Follow these steps:

1. Open PowerShell or Command Prompt as an Admin

2. (Optional) Identify your display adapter and GPU by checking HWID or with this command: Get-PnpDevice -Class Display

Remember to note the exact name in case you use a multi-GPU setup and need to disable a specific GPU driver.

3. Disable the graphics driver with this command: Get-PnpDevice -Class Display | Disable-PnpDevice -Confirm:$false

This command disables all display adapters. The <-Confirm:$false> flag skips the confirmation prompt (useful for automation).

4. Input the following command to ensure the driver fully unloads, then wait 4-5 seconds: Start-Sleep -Seconds 3

5. Re-enable the graphics driver with this command: Get-PnpDevice -Class Display | Enable-PnpDevice -Confirm:$false

6. Verify the restart happened with this command: Get-PnpDevice -Class Display | Select-Object Status, FriendlyName

Check that the status shows “OK” for all the relevant drivers.

Unlike manual methods, PowerShell enables IT teams to address GPU driver issues at scale by restarting drivers on hundreds of endpoints simultaneously or setting up conditional automation that responds to telemetry, maintaining uptime while minimizing support burden.

Tools like Atera’s automation engine can execute PowerShell scripts remotely across managed endpoints. Through custom scripting and automation profiles, IT teams can implement conditional logic such as triggering driver restarts based on GPU temperature thresholds or repeated TDR events.

» Did you know you can paste in PowerShell?

From quick fixes to long-term GPU health

A graphics driver restart in Windows 11 shouldn’t be part of routine maintenance. It’s a reactive measure, not preventive. IT administrators should only trigger it after temporary display issues, such as black screens, flickering, or application freezes, typically caused by TDR (Timeout Detection and Recovery) events. Performing restarts too frequently can interrupt GPU memory management, destabilize active rendering tasks, and accelerate driver cache corruption.

The smarter approach uses proactive monitoring instead of reactive fixes. Rather than waiting for users to report display problems or manually restarting drivers across your fleet, platforms like Atera shift the paradigm from break-fix to prevention.

Atera’s RMM capabilities continuously monitor endpoint performance. Through custom alerts and PowerShell scripts generated by AI Copilot, IT teams can configure monitoring to detect early signs of GPU driver instability. When issues do arise, enterprise IT teams can execute automated PowerShell scripts remotely via Atera’s automation engine to restart drivers without disrupting users, or configure automation workflows that trigger driver restarts when specific Event Viewer thresholds are met.

» Ready to try it out? Start your free trial with Atera