While playing on Giant World Map, caused Windows 11 WatchDog to crash my PC.

[deputyryan32]

Describe the bug
OpenFront slows down over time on the Giant World Map. Input delay builds. FPS drops. System locks up. No response from mouse or keyboard. Windows triggers a watchdog reset after the freeze. System restarts on its own. After restart, Windows runs updates and restarts more than once.

To Reproduce
Steps to reproduce the behavior:

1. Open Brave Browser
2. Go to OpenFront
3. Start a match on Giant World Map
4. Use basic game settings
5. Play for several minutes
6. Notice lag increase and FPS drop
7. System freezes
8. Windows watchdog triggers reset
9. System restarts

Expected behavior
Game runs stable on large maps. No system freeze. No watchdog reset. No forced restart.

Screenshots
Not available. System froze before capture.

Desktop (please complete the following information):

• OS: Windows 11
• Browser: Brave
• Version: Latest

Smartphone (please complete the following information):

• Device: N/A
• OS: N/A
• Browser: N/A
• Version: N/A

Additional context

• GPU: NVIDIA RTX 4060
• CPU: Ryzen 5 7600
• RAM: 32 GB DDR5
• Other apps open: Steam, Discord
• No other tabs open
• Issue shows under load on large map
• Likely GPU driver timeout detected by Windows watchdog
• System entered Windows update cycle after restart

[samisbakedham]

Hey @deputyryan32, thanks for the detailed report — the hardware specs and symptom description made this straightforward to diagnose.

Root cause: the render loop was calling requestAnimationFrame unconditionally with no frame rate cap. On a fast GPU like your RTX 4060, especially on a large map with a lot going on, this can push the renderer to hundreds or even thousands of frames per second. Sustained at that level, the GPU driver stops responding within Windows’ ~2 second timeout window, which triggers TDR (Timeout Detection and Recovery) — the watchdog reset you were seeing.

I’ve opened a PR that adds a configurable FPS limit to the render loop. It defaults to 60 FPS and can be adjusted in Settings (30 / 60 / 120 / 144 / 240 / Uncapped). The throttle skips frames that arrive ahead of schedule rather than doing any extra work, so it should meaningfully reduce GPU load on large maps without any impact on gameplay or tick rate.

In the meantime, if you want a workaround before the fix ships: enable hardware-accelerated GPU scheduling in Windows settings and cap your frame rate at the driver level via the NVIDIA Control Panel (set a max frame rate for your browser). That will achieve the same effect for now.

[deputyryan32]

Thanks!

[iiamlewis]

Added to roadmap

[VariableVince]

Hey @deputyryan32, thanks for the detailed report — the hardware specs and symptom description made this straightforward to diagnose.

Root cause: the render loop was calling requestAnimationFrame unconditionally with no frame rate cap. On a fast GPU like your RTX 4060, especially on a large map with a lot going on, this can push the renderer to hundreds or even thousands of frames per second. Sustained at that level, the GPU driver stops responding within Windows’ ~2 second timeout window, which triggers TDR (Timeout Detection and Recovery) — the watchdog reset you were seeing.

I’ve opened a PR that adds a configurable FPS limit to the render loop. It defaults to 60 FPS and can be adjusted in Settings (30 / 60 / 120 / 144 / 240 / Uncapped). The throttle skips frames that arrive ahead of schedule rather than doing any extra work, so it should meaningfully reduce GPU load on large maps without any impact on gameplay or tick rate.

In the meantime, if you want a workaround before the fix ships: enable hardware-accelerated GPU scheduling in Windows settings and cap your frame rate at the driver level via the NVIDIA Control Panel (set a max frame rate for your browser). That will achieve the same effect for now.

Update on this from some time ago by @scamiv:

"also, no. https://developer.mozilla.org/en-US/docs/Web/API/Window/requestAnimationFrame"

But, that was already addessed by @samisbakedham in the PR. Current state is this: #3462 (comment)