📑 Table of Contents
🎮 The Crash That Keeps Coming Back
Nothing kills momentum in a game like the screen freezing for a second, going black, and dumping you to the desktop with a cryptic error box. If that box mentions DXGI_ERROR_DEVICE_HUNG, you're dealing with one of the more common — and more confusing — PC gaming crashes, because it doesn't point to one single cause.
This error basically means your GPU stopped responding to DirectX commands mid-render. The tricky part is that three or four completely different problems can trigger the exact same message. Rather than making you try every fix blindly, this post is organized by likely cause, so you can jump to the branch that matches your situation first.
I've hit this exact error in two completely different situations myself — once from a stale GPU driver after a Windows feature update, and once from a mild overclock that was fine in benchmarks but unstable an hour into a long session. Those two fixes looked nothing alike, which is exactly why a single "try this one thing" article never quite covers it. Read through the symptoms section first so you can pick the right starting branch instead of working through all four in order.
⚠️ Symptoms & the Exact Error
Typically you'll see something like this right before the game closes or your display driver resets:
The application's device failed due to badly formed commands sent by the application.
Along with the error, watch for these accompanying signs — they help you figure out which branch below applies to you:
- A brief screen flicker or "Display driver stopped responding and has recovered" notification right before the crash
- The crash happens more often during demanding scenes — explosions, crowded areas, ray tracing
- It only happens in one specific game versus happening across multiple games
- It started right after you overclocked, updated a driver, or updated Windows
Take a moment to actually write these details down before moving to the next section — which games are affected, whether it's every session or occasional, and what changed on your system recently. That short list does most of the diagnostic work for you.
🔍 Why This Happens
DXGI_ERROR_DEVICE_HUNG happens when the GPU takes too long to respond to a command and Windows' Timeout Detection and Recovery (TDR) system kills and resets the graphics driver to prevent a full system freeze. What causes the GPU to hang in the first place usually falls into one of these buckets:
- Driver instability — an outdated, corrupted, or freshly-broken GPU driver
- Overclocking — a GPU or CPU pushed past a stable clock/voltage under sustained load
- Outdated DirectX components — missing legacy runtime files a specific game still depends on
- Thermal throttling or power delivery issues — the GPU hitting a thermal or power limit mid-scene
- Game-specific bugs — a poorly optimized DirectX 12 path in that particular title, which is common enough that developers sometimes patch it out entirely in a later update
Knowing that TDR is the mechanism behind the crash (rather than the cause itself) matters, because it explains why the same error message can come from such different root problems — TDR is just the messenger reporting that something upstream took too long.
🛠️ Match Your Cause to a Fix
Pick the branch below that best matches what you noticed in the symptoms section. If it doesn't resolve things, work down through the others in order.
Branch A: You saw "Display driver stopped responding" or it happens across multiple games
This points to the driver itself. Do a clean reinstall rather than a regular update — remove the current driver first, then install fresh.
⬇️ Download the Latest NVIDIA Drivers — Official NVIDIA Page nvidia.com/drivers → auto-detect your GPU model, choose the WHQL-certified version ⬇️ Download the Latest AMD Drivers — Official AMD Page amd.com/support → AMD Software: Adrenalin Edition, auto-detect availableIf your crashes stop, this was it. If that didn't help, try the next branch.
Branch B: It started right after you overclocked anything
Reset your GPU and CPU to stock clocks and voltages in your overclocking utility or BIOS, then run the game again for at least 30 minutes in the area where it used to crash. Overclocks that pass a quick benchmark can still be unstable under a game's specific, sustained load. If that didn't help, try the next branch.
Branch C: It's isolated to one specific game
First, make sure that game's DirectX dependencies are current — many older titles bundle a legacy DirectX End-User Runtime installer, but reinstalling it manually can fix missing components:
⬇️ Download the DirectX End-User Runtime — Official Microsoft Download Center microsoft.com/download → dxwebsetup.exe, installs legacy D3DX/XAudio components some games still needAlso try switching the game's graphics API from DirectX 12 to DirectX 11 in its video settings if that option exists — it sidesteps DX12-specific bugs in a lot of titles. If that didn't help, try the last branch.
Branch D: It happens mostly during heavy, sustained scenes
This points toward thermals or power. Check GPU temperature and clock behavior with a monitoring tool during a crash-prone scene — if temps are pushing 85°C+ or clocks are fluctuating wildly, clean your case's dust filters, reapply thermal paste if your GPU is a few years old, or cap your frame rate slightly to ease sustained load.
🧩 Still Hanging? Extra Checks
| Extra Check | Why It Helps |
|---|---|
| Run the game as Administrator | Fixes rare permission conflicts that can interrupt GPU command queues |
| Increase TDR delay (advanced) | Gives the GPU more time to respond before Windows force-resets the driver |
| Verify game files | Corrupted shader cache or game files can trigger malformed GPU commands |
| Check PSU wattage headroom | A marginal power supply can cause micro voltage drops under GPU load spikes |
Most Steam and Epic launchers have a "Verify integrity of game files" option in the game's properties menu — it's quick and rules out corrupted installs entirely.
❓ FAQ
Q1. Is DXGI_ERROR_DEVICE_HUNG a sign my GPU is dying?
Rarely. It's almost always driver, overclock, or thermal related rather than actual hardware failure, so there's no need to assume the worst before working through the branches above.
Q2. Does this only happen with NVIDIA cards?
No, it's a DirectX-level error defined by Microsoft that can occur with NVIDIA, AMD, or Intel graphics — the underlying causes and fixes are essentially the same across all three vendors.
Q3. Should I downgrade my driver instead of reinstalling the same version?
Worth trying if a specific driver version is known to have issues with your game — check the game's community forums or the vendor's release notes for reports of regressions in that version.
Q4. Can lowering in-game graphics settings help?
Yes, especially if the cause is thermal or power related, since it reduces sustained GPU load. Turning down ray tracing or resolution scaling specifically tends to help the most, since those features push the GPU hardest.
Q5. Is switching from DirectX 12 to DirectX 11 a permanent fix?
It's more of a workaround than a fix, but it's a legitimate one if the game runs stable and you're not missing DX12-exclusive features you care about, like certain ray tracing effects.
✅ Wrap-Up
DXGI_ERROR_DEVICE_HUNG isn't one bug with one fix — it's a symptom that shows up whenever your GPU stops responding, whatever the reason. Match your symptoms to a branch (driver, overclock, single game, or thermal) and work through that path first instead of trying everything at random.
Most people land the fix in the first branch they try once they've honestly identified which pattern matches their crashes, so take a minute with the symptoms section before diving into fixes. And if a fix works, keep an eye on things for a week or two afterward — some causes, like marginal thermal throttling, only show up again once the weather warms up or your case gets a bit dustier.
#DXGIERRORDEVICEHUNG #GameCrash #DirectX #GPUDriver #PCGaming #GameErrors