Strange—why is it written in an "Asian" script above? In any case, here is the AI's response, incorporating the new information received: "These new details pinpoint the exact cause of your crashes. The issue is most likely driven by instability related to transient power spikes or power delivery management combined with the graphics driver, rather than a pure hardware defect in the GPU die itself. Here is why this happens and how to resolve it. Technical Breakdown Why only single-player AAA games. Unlike competitive esports titles—which are often CPU-bound or frame-rate capped—an AAA game like Far Cry 5 saturates the GPU at 99% load. This triggers abrupt, millisecond-level power draw spikes that a steady 111 W synthetic stress test does not replicate. The role of the degraded battery (67% health). On gaming laptops (Tongfang/MECHREVO chassis), the battery acts as an electrical buffer (Hybrid Power) to absorb peak current draws when the AC adapter reaches its limit. At 67% health, internal resistance increases, making it difficult for the battery to smooth out these sudden power demands. The 0x133 error (Parameter 1 = 1) and nvlddmkm.sys. This bug check indicates that a single DPC routine exceeded its allotted execution time on a CPU core. The NVIDIA driver (nvlddmkm.sys) hangs while waiting for a hardware response from the GPU after a micro-voltage drop. Presence of netwtw14.sys (Intel Wi-Fi). The Intel wireless driver is known to cause high DPC latency. When combined with the heavy workload of the NVIDIA driver, it can push the DPC queue past Windows' maximum threshold. The 0xCE bug checks. These are triggered when a third-party driver (typically fan control software like Control Center, an RGB utility, or an anti-cheat driver) fails to properly unload during sleep or shutdown sequences. Recommended Action Plan Follow these steps in order to isolate and fix the issue: 1. Stabilize the GPU voltage/frequency curve (Mild Undervolting). Use MSI Afterburner to cap the GPU's maximum boost clock around 2400–2500 MHz at 0.925 V or 0.950 V. This eliminates aggressive boost spikes responsible for nvlddmkm.sys timeouts without any noticeable performance loss. 2. Switch the power profile in the MECHREVO software. If you are using "Turbo" or "Beast" mode (which pushes power limits), switch to Balanced or standard Gaming mode. This reduces strain on the AC adapter and battery buffer. 3. Update the Intel Wi-Fi driver. Download the latest Intel Wireless Wi-Fi driver directly from Intel's website to address netwtw14.sys DPC latency, or test a gaming session with Wi-Fi disabled (Airplane Mode). 4. Disable Windows Fast Startup. Go to Control Panel > Power Options > Choose what the power buttons do and uncheck Turn on fast startup. This resolves the vast majority of 0xCE BSODs caused by improperly unloaded drivers. 5. Disable Hardware-Accelerated GPU Scheduling (HAGS). Go to Windows Settings > System > Display > Graphics > Default graphics settings, turn this option off, and restart your PC. HAGS is frequently linked to 0x133 crashes on Windows 11 24H2 (build 26100). These parameters confirm the exact nature of the hardware/driver deadlock: Parameter 1 = 0x1 (Cumulative timeout). Unlike 0x0 (which points to a single stuck DPC routine), a value of 0x1 means the CPU spent too much cumulative time processing interrupts (DPC/ISR) at or above DISPATCH_LEVEL without ever yielding back to normal system threads. Parameter 2 = 0x1E00 (7,680 ticks = ~120 seconds). This is the cumulative watchdog threshold (2 minutes) reached before Windows triggers the bug check. It indicates that the NVIDIA driver (nvlddmkm.sys) is getting caught in a wait loop or failing silent TDR (Timeout Detection and Recovery) resets over an extended period prior to the BSOD, rather than experiencing an instant hard lock. Number of processors = 24 (Intel Hybrid Architecture). This confirms a 13th or 14th Gen Intel Core i7/i9 processor (8 P-cores + 16 E-cores = 24 threads). On these chips, hardware interrupts (such as Intel Wi-Fi netwtw14.sys and GPU nvlddmkm.sys) frequently bottleneck on CPU 0. Specific Adjustments to Add to Your Action Plan Alongside the previous steps (mild GPU undervolting and disabling HAGS), these two settings directly target Parameter 1 = 0x1 on a 24-thread hybrid architecture: 1. Disable PCI Express Link State Power Management (ASPM). Go to Control Panel > Power Options > Change plan settings > Change advanced power settings. Expand PCI Express > Link State Power Management and set it to Off (Plugged in). This prevents the GPU from dropping into unstable low-power states during rapid load fluctuations. 2. Test E-core parking (or Game Mode) if crashes persist. On certain 2024 Tongfang/MECHREVO BIOS versions, DPC interrupt scheduling across P-cores and E-cores under Windows 11 24H2 (build 26100) triggers this cumulative latency (0x1E00). If your MECHREVO Control Center or BIOS includes a Game Mode / Legacy Game Compatibility toggle (which parks E-cores during gaming), enable it for testing."

more info: MECHREVO GM6AQ7C laptop (RTX 4070 Laptop, 8 GB; Windows 11 build 26100; BIOS N.1.04MRO11, 2024-08). Symptom: the machine crashes only during AAA single-player games (confirmed with Far Cry 5); online/esports titles run for hours without problems. Bug check: 0x133 DPC_WATCHDOG_VIOLATION, Parameter 1 = 1 (three occurrences: 2026-10-28, 2026-11-07, 2026-11-09). Analysis of the dumps places nvlddmkm.sys (NVIDIA display driver) in the failing path each time; the most recent dump shows FarCry5.exe running and nvlddmkm.sys+0xD9ED19 as the crash source, with netwtw14.sys/ndis.sys also present in the stack. Four additional 0xCE bug checks (DRIVER_UNLOADED_WITHOUT_CANCELLING_PENDING_OPERATIONS) share an identical faulting offset and appear to be a separate third-party driver issue. Already tried: clean driver update (572.83 → 617.42), VRAM test (memtest_vulkan, 0 errors), 83-minute GPU telemetry session at 111 W / 67 °C with no crash, no thermal or power-cap throttling observed. Findings: battery health is approximately 67 % (40,425 of 60,060 mWh); the GPU power limit is currently 115 W versus a 140 W maximum; BIOS is from 2024.   Bug Check Code 0x00000133 Bug Check String DPC_WATCHDOG_VIOLATION Parameter 1 0x0000000000000001 Parameter 2 0x0000000000001E00 Parameter 3 0xFFFFF802B61C53C8 Parameter 4 0x0000000000000000 Crash date November 9, 2026 10:35 PM Architecture x64 Major version 15 Minor version 26100 Number of processors 24

OK,thanks for the guidence。

You don't need Blue Screen View—DriversCloud handles blue screens. It didn't find the problematic driver, unlike DriversCloud.   https://www.driverscloud.com/en/configuration/p4x25p7lzox-46/key

thank you,i will try this again

I retrieved the crash logs using BlueScreenView. After the issue occurred, I used DDU to remove the old GPU drivers and installed the latest graphics driver. Last night, I played Phantom Blade Zero for 1.5 hours, and the problem did not reappear. This issue has occurred multiple times over the past month. I'm not sure if this solution is permanent, so I'm seeking assistance. Furthermore, I've added a pressure-type cooling pad to improve my laptop's heat dissipation. I'm not sure how effective it will be. If any problems arise again in the future, could I still seek your advice? Thank you. https://d.uguu.se/tkaXbknC.dmp

Hello Both To supplement Charon’s response (which is more precise), here is the AI’s response: "Since the crashes only happen in demanding AAA titles and your PC runs fine during regular desktop usage, this strongly points to a hardware stability or power issue under heavy load rather than a general Windows bug. Here are the most likely culprits to check: Overheating (CPU or GPU): AAA games push your components to 100% usage. If temperatures spike too high, the system will freeze or trigger a BSOD to protect itself. Try monitoring your temps under load with a tool like HWiNFO64. Power Supply (PSU) issue: Modern GPUs have sudden power spikes during heavy gaming. If your PSU is aging, faulty, or slightly underpowered, it can cause instant freezes or crashes under load. RAM or GPU instability: An overclock or a factory RAM profile (XMP / EXPO) can be completely stable on the desktop, yet throw critical errors once the RAM or VRAM gets saturated by a heavy 3D engine. Try disabling XMP/EXPO temporarily in the BIOS to test stability. Corrupted GPU drivers: AAA games heavily rely on DirectX 12 and the latest driver features. Doing a clean driver reinstall using DDU (Display Driver Uninstaller) in Safe Mode often rules out software-level GPU crashes. (as Charon said ?) Could you share the specific BSOD stop code (e.g., VIDEO_TDR_FAILURE, WHEA_UNCORRECTABLE_ERROR, MEMORY_MANAGEMENT) please ? That will help narrow down the exact cause ?"          

Hi James, Welcome to DriversCloud Based on your BSOD, it looks like the graphics card driver is crashing when you play Far Cry. First, try installing the latest graphics driver: https://www.driverscloud.com/en/services/GetInformationDriver/77175-216023/nvidia-62114-desktop-win10-win11-64bit-international-dch-whqlexe   If the problem persists, use the DriversCloud monitoring tool to check if your graphics card is overheating

I get frequent freezes and BSODs, only while playing AAA titles, here is the BSOD crash analysis. Everything works fine under regular desktop usage. https://www.driverscloud.com/en/configuration/sdk35v4rp03-1/key

salut fernand, oui le forum est un peu mort en ce moment :), il  y a de grosses sorties de prévu esperons que ca fasse venir plus de monde