Apex Legends: 'DXGI_ERROR_DEVICE_HUNG' Crash - Repair Diagnosis

⚠️ SAFETY WARNING. Incorrect intervention on this device can lead to unforeseen risks or void the warranty. Before opening the computer case for internal cleaning or component replacement, always disconnect the power cable from the wall outlet (230V) and press the power button for a few seconds to discharge residual electricity in the capacitors. It is recommended not to proceed without adequate technical skills. ReeFix provides this diagnosis EXCLUSIVELY for educational and informational purposes.

IF the 'DXGI_ERROR_DEVICE_HUNG' crash occurs predominantly or exclusively during intense Apex Legends sessions (or similar games with high graphical load), while the system is stable in other less demanding activities (browsing, videos, light games).

Key signs/indicators:

• The crash occurs after a variable period of gameplay, often during intense action.
• No crashes or visual artifacts during desktop use or in non-graphically intensive applications.
• The system recovers after the crash, without blue screens (BSOD) or complete reboots.

Likelihood of causes:

1. Corrupted graphics drivers or outdated / Software conflicts (overlays, shader cache): 45-55%
   • Why: The low-level management of DirectX 12 and the sensitivity of Apex Legends' Source Engine make the system particularly vulnerable to unstable drivers or third-party interference.
   • Often happens when: Drivers are updated "over" previous versions without a complete cleanup, or non-essential graphics utilities are installed.
2. GPU operating frequencies (factory/user overclock) too aggressive: 25-35%
   • Why: The GPU cannot maintain stability at the set frequencies under the specific load and peaks required by the game, causing a calculation error and TDR timeout.
3. Inefficient GPU thermal management (degraded thermal paste, obstructed fans): 15-20%
   • Why: Excessive temperatures, especially in localized "hotspots" on the GPU or VRAM, force the chip to reduce performance or generate irreversible errors.

Quick checks (DIY - low risk):

1. Clean reinstallation of graphics drivers: Use Display Driver Uninstaller (DDU) in safe mode to completely remove current drivers, then install the latest stable version from the manufacturer's website (Nvidia/AMD).
2. Disable overlays: Disable all in-game overlays (Discord, Steam, GeForce Experience, Xbox Game Bar, RivaTuner Statistics Server).
3. Temperature monitoring: Use software like HWMonitor or MSI Afterburner to check GPU temperatures during gaming. If they exceed 85-90°C (core) or 95-100°C (hotspot/VRAM), thermal management is critical.
4. Physical cleaning: With a Compressed Air Can, clean the fans and heatsinks of the video card and case.

Estimated costs (for this section):

• DIY repair: Low (cost of compressed air: 5-15€, time).
• Technician (software diagnosis, cleaning, driver reinstallation): 40-80€.
• Technician (software diagnosis, cleaning, driver reinstallation, thermal paste reapplication): 80-150€ (includes cost of Noctua NT-H1 Thermal Paste and labor).

IF the 'DXGI_ERROR_DEVICE_HUNG' crash occurs less predictably, even with moderate loads, or after prolonged use, or with visual artifacts preceding the crash, or extends to multiple applications.

Key signs/indicators:

• The crash can also occur in other 3D applications or benchmarks.
• Possible micro-freezes or visual artifacts (corrupted textures, flickering) before the crash.
• The system might show instability even in idle or with light loads, albeit rarely.

Likelihood of causes:

1. Power Supply Unit (PSU) instability on the 12V line: 30-45%
   • Why: Modern GPUs require significant current peaks. A PSU with degraded capacitors or insufficient capacity cannot provide stable power, causing voltage drops that destabilize the GPU.
   • Common error: Underestimating the quality of one's power supply or its ability to handle the "transient spikes" of latest-generation video cards.
2. GPU or VRAM hardware degradation: 30-40%
   • Why: Over time or due to excessive thermal stress, the GPU silicon or VRAM memory modules can degrade, compromising their operational stability, especially at high frequencies.
3. System RAM instability (aggressive XMP/EXPO profiles, suboptimal voltages): 15-20%
   • Why: Corrupted data sent to the GPU from system RAM can cause a logical stall, triggering the TDR.

Quick checks (DIY - low risk):

1. RAM stability test: Run an in-depth RAM test (e.g., MemTest86 from a USB stick) for several hours.
2. BIOS/UEFI reset: Restore BIOS/UEFI factory settings, disabling XMP/EXPO profiles for RAM and checking if the problem persists.
3. GPU power cable check: Ensure that the GPU's PCI-E cables are securely inserted into both the video card and the PSU.

Estimated costs (for this section):

• DIY repair: Low (time, free software).
• Technician (PSU/RAM/GPU diagnosis, in-depth tests): 80-180€.
• Power Supply Replacement: 100-250€ (cost of Corsair RM750x Power Supply + labor).
• Video Card Replacement: 300-1500+€ (high cost, often not convenient for older GPUs).

THEN:

• If the quick checks from the first scenario (drivers, overlays, surface temperatures, cleaning) resolve the issue: The cause was likely software or related to slight thermal inefficiency. You can manage the repair yourself.
• If the quick checks from the first scenario do not resolve, but the problem remains confined to intensive games: It could be GPU instability (factory overclock) or a deeper thermal issue (dry thermal paste). It is advisable to consult a technician for a more in-depth diagnosis and, if necessary, reapplication of thermal paste or a slight GPU downclock.
• If the problem persists or extends to other contexts (second scenario): The probability of a significant hardware failure (PSU, degraded GPU, unstable RAM) increases. A technician is essential for instrumental diagnosis (PSU test with oscilloscope, prolonged GPU stress tests) and to evaluate the replacement of the faulty component.

CONCLUSION:

The highest probability (45-55%) is that the problem can be resolved independently through a clean reinstallation of graphics drivers and disabling overlays. If this is not enough, GPU frequency instability (factory overclock) is the second most probable cause (25-35%), resolvable with a slight downclock. If temperatures are abnormal, reapplying thermal paste is the next step (15-20%). If the problem is generalized, an insufficient or degraded power supply is a relevant cause (30-45% in the second scenario) requiring technician intervention. Video card replacement remains the last option for irreversible hardware failures.

Operational decision: Perform software and physical cleaning checks; if the problem persists, consider technician intervention for thermal or PSU diagnosis; device replacement is the last option for severe and uneconomical hardware failures.

If after installation you notice a metallic click coming from the Corsair RM750x when starting or shutting down the PC, do not be alarmed. This is the normal operation of the internal overcurrent protection relay, a standard behavior for this model.