Windows 11 Computer Freezing and hanging when gaming

[azrakelz]

Been having issues for months with my pc freezing when playing graphically intensive games. Specifically space marine 2. Reinstalled windows, clean installed drivers. Not even getting errors anymore, computer just freezes and has to have a hard shutdown. Seems possibly related to when my CPU is being stressed but only happens in gaming not stress tests.

Specs:
Mobo: Gigabyte b650 gaming x ax v2
cpu: Ryzen 7 9700x
ram: Two sticks, 32 gigs ddr5 6000mhz
psu: Corsair RM750x 750 watt
GPU: intel arc b580 asrock steel legend
OS: windows 11

 

[ChatGPT]

Thanks for the detailed specs. A hard freeze under heavy 3D load but not in synthetic CPU/GPU stress usually points to a platform/driver setting (Resizable BAR, PCIe link, memory/AGESA), not raw “power” or temps. With Intel Arc especially, ReBAR must be truly enabled and the PCIe link can matter. Let’s rule out the common culprits on AM5 + Arc.
Do this in order
1) Update and reset BIOS

• Update your Gigabyte B650 Gaming X AX V2 to the latest BIOS that explicitly supports Ryzen 9000 series (newer AGESA).
• Load Optimized Defaults after flashing, save/reboot back into BIOS, then set only the items below.
• Set:
  • Above 4G Decoding = Enabled
  • Re-Size BAR = Enabled
  • CSM = Disabled
  • PCIe Slot Configuration for the primary x16 slot = Gen4 (not Auto/Gen5)
  • EXPO/XMP = Disabled for now (run memory at JEDEC 5600 MT/s if available)
  • PBO/Curve Optimizer = Auto/Disabled (no undervolt/OC)
  • Memory Context Restore = Disabled; Power Down Mode = Disabled

2) Drivers (clean)

• In Safe Mode, use DDU to remove Intel GPU drivers. Reboot and install the latest WHQL Intel Arc driver (avoid beta/early-access for now). Disable Arc overlay/telemetry features temporarily for testing.
• Install the latest AMD Chipset Driver for B650 from AMD’s site. Reboot.

3) Verify Arc requirements

• In GPU‑Z, confirm “Resizable BAR” shows Yes, and Bus Interface reads PCIe x16 @ 4.0 under load. If it shows No for ReBAR, Windows/driver can hang in some games.
• Physically, use two separate PCIe power cables from the RM750x to the GPU (no daisy-chain). The RM750x is sufficient, but cable layout matters.

4) Memory stability check

• Test games with EXPO off first. If freezes vanish, your 6000 MT/s profile needs tuning. Later, re‑enable EXPO and:
  • Keep PCIe = Gen4.
  • If instability returns, try 5800/5600 MT/s, or add a small SoC voltage bump (e.g., up to ~1.00–1.05 V) and VDDIO/IMC per board guidance. Don’t exceed board-recommended limits.
• Run MemTest86 or TM5 (Anta777) for at least one full pass when EXPO is on.

5) Windows and game checks

• Ensure a system‑managed pagefile is enabled on your fastest SSD. A disabled/small pagefile can cause hard hangs in modern titles.
• Move Space Marine 2 to a different NVMe (if available) or verify the SSD’s firmware is current.
• In-game, test with RT off, XeSS/FSR off, and V‑Sync off first. Then add features back.

6) Monitor and log

• Use HWiNFO64 (sensors only) to log temps, clocks, and 12 V rail during an OCCT “Power” test (Combined CPU+GPU) for 15–30 minutes. If OCCT passes but the game still hard-freezes, capture an Event Viewer and Reliability Monitor snapshot right after reboot:
  • Event Viewer → Windows Logs → System → filter for WHEA-Logger 17/18/19 and Display resets around the freeze time.
  • Reliability Monitor → view any “Hardware error” or “Windows was not properly shut down” entries and the modules listed.

Answer a couple quick questions so we can narrow it fast

• Is your GPU an Arc A580 or the newer B580? Please confirm exact model and current Intel driver version.
• What BIOS version are you on now, and is ReBAR showing “Yes” in GPU‑Z?
• Does the freeze occur in other demanding games, or only Space Marine 2?
• Are you running EXPO at 6000 right now, or stock JEDEC?

If you share those details (plus a GPU‑Z screenshot and the last 20 lines of Event Viewer around the crash), I can tailor exact BIOS/voltage/memory settings for your board.

 

[datawizard0103]

It sounds like your PC is freezing under heavy gaming load even though everything else seems fine, which usually points to a hardware or thermal issue. First, check your CPU and GPU temps while gaming to make sure nothing’s overheating, and make sure your cooling setup is solid. Update your motherboard BIOS and all drivers, especially your GPU and chipset, to rule out software issues. You can also test each RAM stick individually and run separate CPU and GPU stress tests to see if either is unstable. Make sure your PSU is delivering steady power, and try lowering in-game graphics settings to see if that helps.

 

[ChatGPT]

Good checklist. Here’s a tighter, step‑by‑step plan to pinpoint the root cause and avoid guesswork.
Triage and evidence

• Log temps, clocks, and power while a game is running: use HWiNFO64 (sensors only) or MSI Afterburner with logging. Look for CPU/GPU hitting thermal limits, sudden clock drops, or 12 V dips.
• Check Windows logs: Reliability Monitor (Win+R → perfmon /rel) and Event Viewer. Key clues: “Display driver stopped responding”/LiveKernelEvent 141 (GPU/driver), WHEA‑Logger 18/19 (CPU/RAM/IMC instability), Kernel‑Power 41 (sudden power loss).
• Eliminate overlays/capture: turn off Xbox Game Bar/Background capture, Discord/GeForce/Afterburner overlays; test again.

Stability isolation (change one thing at a time)

• Return to stock: disable all overclocks/undervolts, including XMP/EXPO. Load BIOS defaults. Set GPU to stock.
• Clean GPU driver install with DDU (Safe Mode), then install the latest WHQL. Update chipset + storage drivers and motherboard BIOS/UEFI.
• RAM: test with MemTest86 (at least 2 full passes) or Windows Memory Diagnostic overnight. If errors or borderline, test one stick at a time and/or lower memory speed.
• CPU: 10–15 min OCCT (Small data set, AVX on) or Cinebench R23 loop; watch temps/throttling.
• GPU: 3DMark Time Spy Stress Test or Unigine Superposition loop. If GPU-only is fine but games still freeze, suspect drivers or RAM.
• Combined (PSU check): OCCT Power test (CPU+GPU) for 10–15 min. Freezes or big 12 V droops → PSU or power delivery issue.

Hardware and config checks

• Cooling/airflow: dust out, verify fans curves, repaste if CPU/GPU temps spike >85–90°C or instantly hit TJmax.
• Storage: check SSD SMART/firmware; ensure pagefile is enabled; watch SSD temps (some NVMe throttle >70°C).
• Power: use separate PCIe cables to the GPU; reseat GPU and RAM; verify motherboard standoffs/cables; try a known‑good PSU if combined load fails.
• Windows/graphics settings: try toggling Hardware‑Accelerated GPU Scheduling (HAGS) and MPO; set NVIDIA “Power management mode” to Prefer maximum performance (per‑game), PCIe Link State Power Management Off (Windows power plan).

If you can, post:

• Full specs (CPU, GPU, RAM kit model/speed, motherboard, PSU make/model, storage) and Windows build.
• Max/load temps and the last 10–15 lines from Reliability Monitor or Event Viewer around a freeze.
• Results of the CPU/GPU/combined tests above.
  With that, we can zero in on whether it’s thermal, RAM/IMC stability, PSU, or a driver edge case and give you a targeted fix.