FAULTY_HARDWARE_CORRUPTED_PAGE

FAULTY_HARDWARE_CORRUPTED_PAGE means Windows detected that your RAM returned corrupted data — a bit went in as 1 and came back as 0 (or vice versa). The fastest diagnostic path: run memtest86 overnight. If it shows errors, you have a bad RAM stick. If memtest passes and you have an Intel 13th or 14th gen i7 or i9, your CPU's memory controller is likely degrading — update your BIOS to the latest version with microcode 0x129+ and contact Intel for an RMA.

What's actually happening

Windows maintains a checksum for critical memory pages. Periodically, and especially during page faults (when data is swapped between RAM and disk), Windows verifies that the data in physical memory matches what was originally written. Bug check 0x0000012B fires when this verification fails — the data in RAM has been silently corrupted without any write operation to explain it.

This is one of the few BSODs that almost certainly points to hardware. Software bugs can corrupt data in many ways, but they can't change the physical contents of a RAM address that nothing has written to. When Windows reads a memory page that hasn't been modified since the last checksum and the data doesn't match, the only explanation is that the physical memory cells flipped a bit on their own.

The scary part: by the time you see this BSOD, corruption has already happened. Windows caught one instance, but there may have been many others that went undetected — corrupted data that happened to pass checksum checks, or corruption in non-checksummed memory regions. Silent data corruption can cause bizarre symptoms long before the BSOD appears: random application crashes, file corruption, save game corruption, game desyncs, and system instability that seems impossible to diagnose.

The most common causes (in order of likelihood)

Failing RAM module. The most straightforward cause. DRAM cells degrade over time, and a failing cell can flip bits unpredictably. This is more common in older RAM (3+ years), RAM that has been run at high temperatures (poor airflow over DIMMs), or RAM that was pushed beyond its rated specs with aggressive XMP/EXPO profiles. The corruption is usually concentrated in specific address ranges corresponding to the failing cells, which is why memtest86 catches it by writing and reading every address.

Intel 13th/14th gen CPU memory controller degradation. This is the elephant in the room for 2024-2026 PC builders. Intel's 13th gen (Raptor Lake) and 14th gen (Raptor Lake Refresh) Core i7 and i9 processors have a well-documented issue where elevated voltage causes the CPU's memory controller to degrade over time. The degradation is permanent and progressive — the memory controller becomes less reliable at handling data transfers, causing bit flips that look exactly like bad RAM. Intel acknowledged this publicly and released microcode update 0x129 to limit the voltage and prevent further degradation, but it cannot reverse damage already done. If your CPU is already degraded, the BIOS update stops it from getting worse, but the existing instability remains.

The telltale signs of Intel CPU degradation vs bad RAM: memtest86 passes (because memtest doesn't stress the CPU's memory controller the same way real workloads do), but you still get FAULTY_HARDWARE_CORRUPTED_PAGE during gaming or heavy multitasking. Event Viewer shows WHEA-Logger errors (Event ID 18 or
19) with Machine Check exceptions. The BSODs started months after building the PC and gradually became more frequent.

XMP/EXPO instability. XMP (Intel) and EXPO (AMD) profiles overclock your RAM beyond its base JEDEC specification. The base spec for DDR4 is 2133MHz at 1.20V; XMP might push it to 3600MHz at 1.35V. The base spec for DDR5 is 4800MHz at 1.10V; EXPO might push it to 6000MHz at 1.35V. These profiles are tested by the RAM manufacturer but not guaranteed to work on every motherboard and CPU combination. If the memory controller on your specific CPU can't handle the speed or the motherboard's signal integrity is marginal, you get intermittent bit flips under load.

Insufficient RAM voltage. RAM needs a precise voltage to operate reliably. DDR4 JEDEC spec is 1.20V; XMP profiles typically need 1.35V. DDR5 JEDEC is 1.10V; EXPO profiles need 1.25-1.40V depending on speed. If the voltage delivered to the DIMMs is too low — because the BIOS default is wrong, because you manually set it incorrectly, or because the motherboard's voltage regulators can't deliver stable power — RAM operations become unreliable and bits flip.

Overheating RAM. RAM modules generate heat, and high temperatures reduce the reliability of DRAM cell retention. This is more common with DDR5 (which runs hotter than DDR4) and in systems where the RAM DIMMs are sandwiched between a large CPU cooler and the top of the case with no airflow. RAM temperatures above 50C reduce stability; above 60C, errors become significantly more likely.

How to fix it

1. Run memtest86 overnight. Download memtest86 from memtest86.com (the free version is fine). Create a bootable USB drive with the tool. Boot your PC from the USB and let memtest run for at least 8 hours — ideally overnight. Even a single error means you have bad RAM. Note which test and which address range shows errors — this helps identify which specific DIMM is failing. If memtest shows zero errors after 8+ hours, skip to step 3 for the Intel CPU check.
2. Test each RAM stick individually. If memtest shows errors, shut down your PC, remove all but one RAM stick, and place it in the first DIMM slot (usually the slot labeled A2 or DIMM_1 — check your motherboard manual). Run memtest again for at least 4 hours. If it passes, swap to the next stick and test again. The stick that produces errors is the faulty one. Replace it — don't try to "fix" bad RAM by underclocking or adding voltage. Once DRAM cells start failing, the degradation accelerates.
3. Check for Intel 13th/14th gen CPU degradation. If you have an Intel Core i7-13700K/KF, i9-13900K/KF/KS, i7-14700K/KF, or i9-14900K/KF/KS (or their non-K variants), your CPU's memory controller may be degrading. First, update your BIOS to the latest version from your motherboard manufacturer — it must include Intel microcode 0x129 or newer. Check the BIOS changelog on the download page; it will mention the microcode version. After updating, open Event Viewer > Windows Logs > System and filter for Source: "WHEA-Logger" with Event ID 18 or
4. If you see Machine Check Exception (MCE) errors, especially with MCI Status values indicating cache or memory controller errors, your CPU is degraded. Contact Intel for an RMA — they extended the warranty for affected processors.
5. Disable XMP/EXPO and test at JEDEC spec. Enter BIOS and disable XMP (Intel) or EXPO/DOCP (AMD). This returns your RAM to its base JEDEC speed and voltage. For DDR4, that's 2133MHz at 1.20V. For DDR5, that's 4800MHz at 1.10V. Run your system for several days of normal use. If the BSOD stops, the XMP/EXPO profile is pushing your RAM or memory controller beyond what's stable. You can try re-enabling XMP at a lower speed (e.g., 3200MHz instead of 3600MHz) or manually increasing RAM voltage by 0.02-0.05V to improve stability.
6. Check RAM voltage in BIOS. Go into BIOS and find the DRAM Voltage setting (sometimes called Memory Voltage or VDIMM). For DDR4 with XMP: should be 1.35V (or whatever the XMP profile specifies — check your RAM's rated voltage on the label). For DDR5 with EXPO: should be the voltage specified by the EXPO profile, typically 1.25-1.40V. If the voltage is set to "Auto" and you're having issues, try setting it manually to the rated value. Some motherboards under-deliver voltage on Auto.
7. Run Windows Memory Diagnostic as a quick check. Open the Start Menu, search "Windows Memory Diagnostic," and run it. It restarts your PC and runs a basic memory test. It's less thorough than memtest86 but it catches severe issues quickly. Check results after reboot: open Event Viewer > Windows Logs > System, look for "MemoryDiagnostics-Results."
8. Check for WHEA errors in Event Viewer. Open Event Viewer > Windows Logs > System. Filter by Source: "WHEA-Logger." Look for Event ID 18 (Machine Check Exception) and Event ID 19 (Corrected Hardware Error). MCE events (ID
9. with memory controller or cache bank codes indicate CPU memory controller degradation. Corrected errors (ID
10. with high frequency indicate hardware that's barely holding on — currently correctable but heading toward uncorrectable failures.

Is this a hardware or software problem?

This is almost always hardware. FAULTY_HARDWARE_CORRUPTED_PAGE specifically means physical memory returned wrong data, which is a hardware failure by definition. The question is which hardware: RAM modules, CPU memory controller, or (rarely) motherboard signal integrity.

Here's the diagnostic tree: Run memtest86 overnight. If it shows errors, it's a bad RAM stick — test individually to find which one. If memtest passes, check for Intel 13th/14th gen CPU degradation (BIOS update + WHEA errors in Event Viewer). If you're on AMD or older Intel and memtest passes, try disabling XMP/EXPO — if the BSOD stops, the memory overclock was pushing beyond what your system can handle.

The one exception where this could be software: a severely corrupted Windows installation can, in very rare cases, trigger false FAULTY_HARDWARE_CORRUPTED_PAGE errors due to page file corruption on a failing SSD. If memtest passes and no WHEA errors exist, check your SSD health with CrystalDiskInfo.

If you're not sure, Crashless can check your drivers, temps, VRAM, and 400+ known patterns automatically — just use the chat above.

Games commonly affected

This isn't game-specific — it's a system-level hardware error. However, gaming triggers it more than light desktop use because games stress both the CPU and RAM heavily. Memory-intensive games like Cities: Skylines II, ARK: Survival Ascended, Hogwarts Legacy, Star Citizen, and heavily modded Skyrim/Minecraft are the most common triggers because they fill RAM to capacity, exposing bad cells that lighter workloads would never touch.

Frequently asked questions

Q: Memtest86 passed with zero errors but I still get this BSOD. What now?
A: Two possibilities. First, check for Intel 13th/14th gen CPU degradation — memtest doesn't stress the memory controller the same way real workloads do. Look for WHEA-Logger Event ID 18 in Event Viewer. Second, try disabling XMP/EXPO — the BSOD might only happen at overclocked speeds that memtest doesn't push hard enough to trigger.

Q: Can I just replace one RAM stick or do I need a whole new kit?
A: If memtest identifies one faulty stick, you can replace just that one. But for best compatibility, buy the same brand, model, and speed. Mismatched RAM kits can introduce subtle instability even if each stick tests fine individually. If your RAM is still under warranty (most kits have lifetime warranty), RMA the faulty stick.

Q: I have an Intel 13th gen i9. Is my CPU going to die?
A: If you update your BIOS to include microcode 0x129+, the voltage issue that causes degradation is mitigated going forward. If your CPU is already degraded (WHEA errors, instability that RAM replacement doesn't fix), Intel is honoring RMA requests for affected SKUs with an extended warranty. Don't wait — the degradation is progressive, so an RMA sooner is better than later.