Intel 13th/14th Gen Crashes & BSODs? Here’s the Fix

November 6, 2026November 6, 2026 0 Comments

Rate this post

A PC that built fine, ran cool, and stayed quiet for months — then starts crashing under load, throwing a WHEA_UNCORRECTABLE_ERROR blue screen, or dying in the middle of a game. If the CPU is an Intel 13th or 14th gen chip, this is probably not your fault.

This is the Intel degradation issue, and it landed hard enough that Intel extended warranty coverage for affected processors. I've had three customer machines with it this year — two i5-13600K builds, one i9-14900K — and the pattern was the same every time: fine for a while, then a slow slide into crashes, and the fixes below are the ones that actually stopped it.

Table of Contents

What's actually going on

In 2024 Intel confirmed that certain 13th and 14th gen desktop CPUs can degrade over time due to voltage stress in the silicon. Symptoms match what you're seeing:

Crashes, freezes, or blue screens during games or heavy load
WHEA_UNCORRECTABLE_ERROR, MEMORY_MANAGEMENT, or generic crash-on-load
Failures getting more frequent over weeks
Sometimes: system won't boot at all after a period of working fine

The degradation is progressive — the chip slowly gets worse, which is why a PC can seem "fine" for months and then collapse. It's a silicon/voltage problem, not a Windows problem, though a Windows update is sometimes the thing that finally tips a marginal chip over.

Which chips are affected: Intel Core i9, i7, and i5 from the 13th gen (Raptor Lake) and 14th gen (Raptor Lake Refresh) desktop families — roughly the i5-13400 and i7-12700-class and up within those gens. Laptops (mobile parts) are generally not part of this. If you have a 12th gen or older chip, a very small number were also later added to the covered list — Intel's checker (below) tells you exactly.

Step 1: Run Intel's official health check (5 minutes)

Intel ships a free tool that tells you whether your specific CPU is covered and whether degradation is likely:

– Whether your SKU is in the affected/covered list – Whether the chip shows signs consistent with the degradation issue

Go to Intel's support site and search "Intel Processor Identification and Health Check" (or "DCH tool").
Download and run it. It reads the processor ID and reports:

This single step separates "my CPU is one of the affected ones" from "my crashes have another cause" — and it decides which branch of the fix tree you're on.

Step 2: Apply the Intel microcode update (the core fix)

Intel's actual remedy is a microcode update — a firmware patch to the CPU itself that adjusts voltage behavior to slow or stop the degradation. It's free, and it's the thing that saves an affected chip.

How to get it on:

Update your BIOS/UEFI first. Go to your motherboard maker's support page (ASUS, MSI, Gigabyte, ASRock, or your PC brand), find the latest BIOS for your exact board, and flash it. The current BIOS builds include Intel's latest microcode.
Check the microcode version in Windows: Task Manager > Performance > CPU — the "Baseboard" area, or use the CPU-Z / HWInfo tools which show the microcode revision. Your board's BIOS release notes will state which microcode it carries.
If your board can't get the new microcode (very old BIOS, or a board the maker stopped updating), the microcode can also be delivered via Windows as a driver update — Intel and Microsoft push it through Windows Update for supported systems. Run Settings > Windows Update > Check for updates and look under Advanced options > Update history for an Intel microcode entry.

After the update, restart. This doesn't reverse damage already done, but it stops the progression on most still-working chips — which is why it's step two, right after the diagnosis.

Step 3: Check your warranty / RMA window

Because Intel acknowledged the defect, it extended warranty coverage for affected 13th/14th gen desktop processors — the window has been extended through end of 2025 (and Intel has signaled continued coverage beyond for affected SKUs; check their current statement for the exact date, since it has moved more than once).

What that means practically:

If your PC is still under its original warranty (most 2023–2024 builds), request an RMA now. A dead or dying i9-13900K is exactly what this covers.
If it's a prebuilt (Dell, HP, Lenovo, ASUS System), the manufacturer handles the CPU swap under the same extended program — open a ticket citing the "Intel 13th/14th gen degradation" issue and your Processor Health Check result.
Keep the Health Check report screenshot. It's your evidence.

Don't wait for the chip to die completely — a chip that still boots is easier to RMA and easier to recover data from.

Step 4: Stabilize a still-usable chip (power limits)

If the chip is degrading but you need the machine stable now (RMA takes weeks), you can reduce the voltage stress that's driving the degradation. This buys stability, not a cure:

Flash to the latest BIOS (Step 2) — newer BIOS builds include more conservative power defaults.
In the BIOS, if your board exposes them, lower the multi-core power limit (PL1/PL2) by 10–20%, or enable an "Intel default" / "balanced" power profile if offered. You lose a little peak performance; you gain a chip that stops cooking itself.
Keep temperatures sane — repaste if the last paste is years old, make sure the cooler is seated, and don't run sustained all-core loads at stock if you can avoid it.

This is the bridge, not the destination. The destination is the microcode update holding the line and the RMA replacing the chip.

Step 5: Rule out the usual suspects too

Not every crash on a 13th/14th gen box is the silicon issue — these chips run hot and hungry, so the ordinary culprits show up at higher rates:

RAM: XMP profiles pushed too far. Disable XMP, test at stock, and run a memory check if you see memory errors.
Thermals: a cooling solution undersized for a 14900K will throttle and crash. Check temps under load (HWInfo); sustained above ~95°C is the problem.
Power supply: a marginal PSU under a heavy CPU+GPU load throws exactly these crashes. If the board and CPU check out and the PSU is cheap or old, it's on the list.
Drivers: a fresh post-update blue screen with a named driver file is a driver issue, not the CPU — the two coexist often.

FAQ

How do I know my specific CPU is affected? The Processor Health Check tool is authoritative for your exact serial/SKU. If it says the chip is in the covered population and shows degradation signs, treat it as affected.

Does the microcode update fix a chip that already blue-screens weekly? It stops further degradation, but damage already accumulated may remain — a chip that's already crashing hard might stay flaky. That's the RMA case. The update is still worth applying; it's free and it's the basis of the warranty claim.

Are 12th gen chips affected? Mostly no — the big affected population is 13th and 14th gen desktop. Intel later added some 12th gen SKUs to coverage; run the Health Check to be sure about your part.

My laptop has a 13th gen Intel chip. Same issue? The degradation program targets desktop processors. Mobile (laptop) 13th/14th gen parts are largely outside it — a crashing laptop is more likely thermals, RAM, or a driver.

Is it safe to keep using the PC while waiting for an RMA? Yes, with care: apply the microcode update, ease the power limits, back up regularly, and avoid running it hot at full load daily. The risk is it degrades further — not that it takes your data down with it, though back up anyway.

Will Windows 11 updates make it worse? An update won't cause the degradation, but a heavier feature update can push a marginal chip past the point where it holds — which is why "it started after an update" is such a common report on these chips.

One last thing

If you're on a 13th or 14th gen desktop and the crashes came on slowly over weeks, run the Health Check today and get the microcode on — both are free, both take an evening, and the RMA window is the part with a clock on it.

Three of my customer machines this year: two resolved with the BIOS microcode plus a power-limit trim, one went to RMA when the chip refused to hold a game session. All three are running fine now. The silicon issue is real, but it has a straight-line fix if you catch it while the chip still boots.

For the other post-update crash patterns, the blue screen guide covers the non-CPU causes, and the 0x800F0922 update error guide handles the updates that won't install in the first place.