Repeated BSODs (PAGE_FAULT_IN_NONPAGED_AREA & HYPERVISOR_ERROR)

Question

I am experiencing recurring Blue Screen crashes on a brand-new custom-built PC running Windows 11. The system was built approximately three weeks ago, and these issues have been present since early use.

I have been using XMP I since day one, and the system has shown instability since the beginning. I attempted numerous fixes, including a full fresh reinstall of Windows 11, but the issue persists.

I also switched from XMP I (ASUS-optimized) to XMP II (manufacturer-recommended settings) to rule out motherboard-level tuning differences. Despite this change, the same BSODs continue to occur.

Crashes occur:

Sometimes under load (After Effects work)

Sometimes during light tasks

Occasionally while idle

---

Observed Bugchecks

1️⃣ PAGE_FAULT_IN_NONPAGED_AREA (0x50)

Faulting module reported: bthhfenum.sys

Indicates invalid memory access in nonpaged memory

Commonly associated with:

Memory corruption

  Faulty or unstable drivers interacting with memory
  
     Marginal RAM stability
     

2️⃣ HYPERVISOR_ERROR (0x20001)

Indicates a fatal error in the Windows hypervisor

Often associated with:

Memory corruption

  DDR5 instability
  
     Interaction between virtualization-based features and hardware
     
     Occurs even without intentional virtualization workloads
     

---

System Specifications

OS: Windows 11 x64 (fully updated)

CPU: Intel Core Ultra 7 265KF (20 threads)

Motherboard: ASUS ROG STRIX Z890-H Gaming WiFi

RAM:

TeamGroup T-Force DDR5

  96 GB total (2 × 48 GB)
  
     Rated at 6800 MT/s
     

• GPU: ASUS PRIME GeForce RTX 5070 Ti
• Cooling: ASUS ProArt LC 360 AIO
• Storage: Samsung 9100 PRO 2TB NVMe SSD (OS installed on clean drive)
• Power Supply: ASUS TUF Gaming 1000W Gold

---

Memory Configuration & BIOS Settings

XMP enabled from first boot

Initially running at DDR5-6800

Retested at DDR5-6600

DRAM VDD: 1.40 V

DRAM VDDQ: 1.40 V

System Agent Voltage: Auto

Memory Controller Voltage: Auto

VRM settings: Auto

No manual overclocking or undervolting has been applied beyond enabling XMP.

---

Memory Stress Testing

Tool used: TestMem5 (AMTA777 Extreme config)

Tested memory size: ~84 GB

Multiple cycles completed

0 reported memory errors

Despite passing memory stress tests, BSODs continue to occur.

---

Thermals & Power

• CPU package temperature peaks around ~80°C under stress, averaging waay lower
• GPU temperatures within normal operating range
• RAM temperatures peaked around ~59°C

No evidence of thermal throttling

• CSV hardware info logs reviewed at time of crashes show no thermal spikes

---

Key Observations

The PC is brand new

Crashes began with XMP enabled and persist across frequency reductions

No single third-party driver consistently identified as root cause

Bugchecks are consistent with memory corruption or memory controller instability

Hypervisor-related crashes suggest sensitivity to memory stability rather than pure software faults

---

Suspected Root Cause

Based on all diagnostics performed so far, the issue appears most consistent with:

Marginal DDR5 XMP stability

Memory controller / system agent sensitivity with high-capacity DDR5 (2 × 48 GB)

Interaction between Windows 11 hypervisor features and high-frequency DDR5 memory

---

What I’m Looking For

I would appreciate guidance on:

Whether these bugchecks align with known DDR5 XMP / IMC stability issues

Recommended DDR5 frequencies or voltage strategies for 96 GB configurations

Whether Windows hypervisor features are especially sensitive to borderline memory stability

Additional diagnostics Microsoft recommends to conclusively isolate hardware vs OS interaction

---

Additional Notes

System was assembled recently

Clean Windows installation

BIOS updated

No third-party antivirus installed

• Crashes persist across different workloads I am experiencing recurring Blue Screen crashes on a brand-new custom-built PC running Windows 11. The system was built approximately three weeks ago, and these issues have been present since early use. The memory XMP profile has been enabled since day one, using the manufacturer’s recommended settings. Despite passing multiple memory stress tests and having safe operating temperatures, the system continues to crash unpredictably. Crashes occur:
  • Sometimes under load (After Effects work)
  • Sometimes during light tasks
  • Occasionally while idle

  Observed Bugchecks

  1️⃣ PAGE_FAULT_IN_NONPAGED_AREA (0x50)

  • Faulting module reported: bthhfenum.sys
  • Indicates invalid memory access in nonpaged memory
  • Commonly associated with:
    • Memory corruption
    • Faulty or unstable drivers interacting with memory
    • Marginal RAM stability

  2️⃣ HYPERVISOR_ERROR (0x20001)

  • Indicates a fatal error in the Windows hypervisor
  • Often associated with:
    • Memory corruption
    • DDR5 instability
    • Interaction between virtualization-based features and hardware
  • Occurs even without intentional virtualization workloads

  System Specifications

  • OS: Windows 11 x64 (fully updated)
  • CPU: Intel Core Ultra 7 265KF (20 threads)
  • Motherboard: ASUS ROG STRIX Z890-H Gaming WiFi
  • RAM:
    • TeamGroup T-Force DDR5
    • 96 GB total (2 × 48 GB)
    • Rated at 6800 MT/s
  • GPU: AMD Radeon RX 6800
  • Cooling: High-end liquid CPU cooler
  • Storage: NVMe SSD (OS installed on clean drive)
  • Power Supply: High-capacity, sufficient for full load

  Memory Configuration & BIOS Settings

  • XMP enabled from first boot
  • Initially running at DDR5-6800
  • Retested at DDR5-6600
  • DRAM VDD: 1.40 V
  • DRAM VDDQ: 1.40 V
  • System Agent Voltage: Auto
  • Memory Controller Voltage: Auto
  • VRM settings: Auto
  No manual overclocking or undervolting has been applied beyond enabling XMP.

  Memory Stress Testing

  • Tool used: TestMem5 (AMTA777 Extreme config)
  • Tested memory size: ~84 GB
  • Multiple cycles completed
  • 0 reported memory errors
  Despite passing memory stress tests, BSODs continue to occur.

  Thermals & Power

  • CPU package temperature peaks around ~90°C under stress, averaging lower
  • GPU core and junction temperatures within normal operating range
  • RAM temperatures peaked around ~70°C
  • No evidence of thermal throttling
  • CSV hardware logs reviewed at time of crashes show no thermal spikes

  Key Observations

  • The PC is brand new
  • Crashes began with XMP enabled and persist across frequency reductions
  • No single third-party driver consistently identified as root cause
  • Bugchecks are consistent with memory corruption or memory controller instability
  • Hypervisor-related crashes suggest sensitivity to memory stability rather than pure software faults

  Suspected Root Cause

  Based on all diagnostics performed so far, the issue appears most consistent with:
  • Marginal DDR5 XMP stability
  • Memory controller / system agent sensitivity with high-capacity DDR5 (2 × 48 GB)
  • Interaction between Windows 11 hypervisor features and high-frequency DDR5 memory

  What I’m Looking For

  I would appreciate guidance on:
  • Whether these bugchecks align with known DDR5 XMP / IMC stability issues
  • Recommended DDR5 frequencies or voltage strategies for 96 GB configurations
  • Whether Windows hypervisor features are especially sensitive to borderline memory stability
  • Additional diagnostics Microsoft recommends to conclusively isolate hardware vs OS interaction

  Additional Notes

  • System was assembled recently
  • Clean Windows installation
  • BIOS updated
  • No third-party antivirus installed
  • Crashes persist across different workloads

Answer 1

Hi,

Well, the first minidump reports a failure with the bluetooth driver Sys.

Did you install all the chipset drivers that could influence the issue of using XMP?

Do you have Update KB5074109 installed on your PC?

Thanks

Answer 2

It sounds like you're experiencing complex issues with your new custom-built PC, particularly with the recurring Blue Screen of Death (BSOD) errors related to PAGE_FAULT_IN_NONPAGED_AREA and HYPERVISOR_ERROR. Here are some insights and recommendations based on the information you've provided:

Observations on Bugchecks

1. PAGE_FAULT_IN_NONPAGED_AREA (0x50): This error often indicates memory corruption or issues with drivers. Given that the faulting module is bthhfenum.sys, which is related to Bluetooth functionality, it may be worth checking for driver updates or disabling Bluetooth temporarily to see if it resolves the issue.
2. HYPERVISOR_ERROR (0x20001): This error suggests a problem with the Windows hypervisor, which could be linked to memory stability issues, especially with high-capacity DDR5 RAM. The interaction between virtualization features and memory can be sensitive, particularly if the memory is not fully stable.

Recommendations

• Memory Configuration: Since you've already tried different XMP profiles, consider manually adjusting the memory voltage or frequency. Sometimes, slightly lowering the frequency (e.g., to 6400 MT/s) or increasing the voltage can improve stability. Ensure that your memory controller settings are also optimized for the RAM configuration you are using.
• Driver Updates: Ensure that all drivers, especially for the motherboard and any peripherals, are up to date. This includes checking for updates from the ASUS website for your motherboard and any other components.
• BIOS Settings: Since you mentioned that the BIOS is updated, double-check the settings related to memory and virtualization. Sometimes, disabling certain features in the BIOS related to virtualization can help if you're not using them actively.
• Testing with Minimal Configuration: Try running the system with only one stick of RAM and see if the crashes persist. This can help identify if one of the RAM sticks is faulty or if there are compatibility issues.
• Windows Hypervisor Features: If you are not using any virtualization features, consider disabling them in Windows to see if it reduces the frequency of the hypervisor-related crashes.
• Further Diagnostics: You might want to run additional diagnostics using tools like Windows Memory Diagnostic or MemTest86 to check for any underlying memory issues that might not have been caught by TestMem5.

Conclusion

Given the complexity of your setup and the nature of the errors, it may take some trial and error to isolate the issue fully. The symptoms you're experiencing do align with known issues related to high-capacity DDR5 memory and potential instability with XMP settings. If the problem persists after trying the above recommendations, you may want to consult with the hardware vendors or consider professional diagnostics to further investigate the hardware interactions.

---

References:

• Understand bug check system halts
• Common Windows stop errors
• Troubleshoot stop errors: Best practices and dump configuration recommendations