Detailed analysis
WHEA_UNCORRECTABLE_ERROR means your hardware reported a problem that Windows could not silently fix. The fastest way to stop this BSOD for most people: reset your BIOS to defaults, update to the latest BIOS version, and disable XMP. If you have an Intel 13th or 14th Gen i7 or i9, stop everything and read the Intel section below — your CPU may be physically degrading and every day without the microcode fix makes it worse.
What is actually happening
WHEA stands for Windows Hardware Error Architecture. It is a reporting layer built into your CPU, RAM controller, and chipset that watches for data corruption at the hardware level. When your CPU does a math operation and gets the wrong answer, or your RAM flips a bit that ECC-less memory cannot correct, WHEA catches it. A correctable error gets logged silently. An uncorrectable error — one that would mean corrupted data reaching your OS or files — triggers an immediate BSOD to prevent damage. That is 0x00000124.
The BSOD parameters tell you the source. Parameter 1 = 0x0 means Machine Check Exception, which is a CPU error. This is the most common one. Parameter 1 = 0x2 means NMI (Non-Maskable Interrupt), pointing to a broader hardware fault. You can read your minidump in C:\Windows\Minidump with a free tool called WhoCrashed or WinDbg Preview from the Microsoft Store.
The Intel 13th/14th Gen crisis — read this first if you own a Core i7-13700K/KF, i9-13900K/KF/KS, i7-14700K/KF, or i9-14900K/KF/KS
Intel confirmed that these CPUs have a known degradation issue caused by excessive voltage under load. The CPUs physically degrade over time — transistors oxidize, error rates climb, and eventually the chip becomes unstable at any voltage. This is permanent and irreversible. Intel released microcode update 0x129 (and later 0x12B) to prevent further degradation by capping eTVB behavior and enforcing safe power limits. But here is the critical part: the microcode update only prevents future damage. If your CPU has already degraded, the fix is an RMA replacement.
How to tell if your Intel CPU is affected: First, update your BIOS immediately. Go to your motherboard manufacturer's website (ASUS, MSI, Gigabyte, ASRock), find your exact board model, and download the latest BIOS. It must contain microcode 0x129 or newer — the changelog will say so. Flash it via USB following the instructions in your manual. After the update, look for an "Intel Default Settings" profile in BIOS and enable it. This enforces Intel's recommended power limits instead of the unlimited power profiles that many motherboard makers shipped by default.
If WHEA errors continue after the microcode update at Intel Default Settings, your CPU has likely already degraded. Contact Intel support at intel.com/supporttickets and request an RMA. Intel extended the warranty to cover this issue. You will need your processor's batch number (ATPO/FPO) which is printed on the CPU or can be read from CPU-Z. The RMA process takes about two to three weeks and they ship you a replacement.
The most common causes (in order of likelihood)
Unstable XMP or EXPO profile — This is cause number one for people who are not on affected Intel chips. XMP is technically an overclock, and it pushes your CPU's memory controller beyond its rated speed. If your system was stable for months and suddenly throws WHEA, your silicon may be aging out of the XMP speed it used to handle. Disable XMP in BIOS, run for a few days, and see if WHEA stops. If it does, re-enable XMP but drop one speed tier (for example, 3600 to 3200 for DDR4, or 6000 to 5600 for DDR5).
Manual CPU overclock — Any manual overclock to core ratio, voltage, or power limits can cause WHEA. Reset BIOS to optimized defaults. Run at stock for at least 48 hours of normal use. If WHEA stops, your overclock was the problem.
CPU overheating and VRM overheating — Open HWiNFO64 in sensors-only mode. Look at CPU Package temperature under load. If it hits 100C, your cooler is insufficient or improperly mounted. Also check VRM MOS Temperature — if VRMs exceed 110C, they cannot deliver clean voltage to the CPU and you will get WHEA errors even at stock. VRM overheating is common on budget B660 and B760 boards running power-hungry i7 or i9 chips.
Failing RAM — Bad memory causes WHEA when the CPU's memory controller tries to read corrupted data. Download memtest86 (free from memtest86.com), create a bootable USB, and let it run overnight. Any errors mean the stick is bad. If you have two sticks, test each one individually to isolate the bad one.
Actual CPU failure — If everything above checks out — stock settings, good temps, good RAM, latest BIOS — and WHEA keeps happening, the CPU itself may be dying. Run Intel Processor Diagnostic Tool (free from Intel) or Prime95 small FFTs for 30 minutes. If it crashes during these tests, the CPU needs replacing.
How to fix it
- Reset BIOS to optimized defaults. This undoes every overclock and custom setting. Restart your PC, mash Delete or F2 to enter BIOS, find Load Optimized Defaults (the exact name varies by board), save and exit.
- Update BIOS to the latest version. Download from your motherboard manufacturer's support page (not from random sites). Extract to a FAT32 USB drive. Use your board's BIOS flash utility — ASUS calls it EZ Flash, MSI calls it M-Flash, Gigabyte calls it Q-Flash, ASRock calls it Instant Flash. Never power off during a BIOS update.
- After the BIOS update, re-enable XMP only if you were stable before. If WHEA returns, disable XMP and drop to JEDEC defaults (2133 for DDR4, 4800 for DDR5). You can try a lower XMP speed later.
- Monitor temps under load. Run Cinebench R23 multi-core for 30 minutes while watching HWiNFO64. CPU Package should stay below 90C and VRM MOS below 110C. If either is too hot, fix cooling before doing anything else.
- Test RAM with memtest86 for a minimum of 4 passes (usually overnight). Any single error means bad RAM.
- If on Intel 13th/14th Gen i7 or i9: enable Intel Default Settings in BIOS after the update. If WHEA continues, initiate an RMA with Intel.
Is this a hardware or software problem?
WHEA_UNCORRECTABLE_ERROR is almost always hardware. The word "uncorrectable" means the error happened at the transistor level — software cannot cause this. However, software-controlled settings like voltage, clock speed, and memory timings determine whether the hardware operates within safe margins. So the root cause is hardware, but the trigger is often a BIOS setting pushing that hardware too hard.
If you see this error exactly once and it never comes back after a BIOS reset, it was probably a transient fault — cosmic ray, power glitch, whatever. It happens. If it repeats, something is wrong and you need to find it.
Frequently asked questions
Q: Does WHEA_UNCORRECTABLE_ERROR mean I need a new CPU?
A: Not necessarily. About 60% of the time it is an unstable overclock or XMP profile that just needs to be dialed back. Reset BIOS to defaults and test first. Only suspect actual hardware failure if it crashes at completely stock settings with verified good RAM.
Q: I have an Intel 13700K and just started getting these. Is my CPU dying?
A: It is possible. Update your BIOS to get microcode 0x129+, enable Intel Default Settings, and test. If WHEA continues at stock with default settings and clean RAM, yes, your CPU may have degraded. Start the Intel RMA process now — the sooner you send it, the sooner you get a replacement.
Q: My WHEA errors only happen when XMP is enabled. Is my RAM bad?
A: Probably not. Your CPU's memory controller just cannot handle that specific XMP speed. This is the silicon lottery — some CPUs run DDR4-3600 fine, others max out at
3200. Disable XMP, run memtest86 to confirm the RAM itself is healthy, then re-enable XMP at a lower speed.
Q: Can a Windows update cause WHEA errors?
A: No. WHEA is reported by hardware before Windows even gets involved. A Windows update cannot cause a genuine WHEA error. However, a Windows update that changes power management or driver behavior can shift CPU load in a way that exposes an existing hardware instability you had not triggered before.