Intel 13th and 14th gen crashes in games: the fix that matters

An Unreal Engine game closes while loading with a shader decompression error. Another one says it ran out of video memory on a graphics card with plenty to spare. Programs that work the processor hard on many threads, such as a benchmark or a video encoder, can crash too, and the whole PC may hang. If the processor is a 13th or 14th Gen Intel Core desktop chip, that picture has a documented hardware cause, and Intel's fix is a firmware update that prevents the damage rather than repairing it.

What the crashes look like

Intel describes the instability as "commonly experienced as consistent application crashes and repeated hangs". RAD Game Tools, the maker of the Oodle data compression software, reports crashes in games built with Unreal, and when an Unreal Engine game starts, failures of this type:

DecompressShader(): Could not decompress shader (GetShaderCompressionFormat=Oodle)

RAD adds that the problem "can also show up as a GPU error message, such as spurious 'out of video memory' errors, even though it is caused by the CPU."

None of these messages proves on its own that the processor is at fault. A game can run out of video memory for real, and a crash can have other causes; why did my game crash? covers where Windows records each kind.

What Intel found

Intel calls the problem Vmin Shift Instability. Vmin is the minimum operating voltage a core needs. Intel's August 2024 update describes "a significant increase to the minimum operating voltage (Vmin) across multiple cores on affected processors due to elevated voltages", and adds that elevated voltage events "can accumulate over time".

In its root-cause statement of September 2024, Intel traced the problem to "a clock tree circuit within the IA core which is particularly vulnerable to reliability aging under elevated voltage and temperature", and named four conditions that drive it, each with its own mitigation:

In May 2025 Intel released microcode 0x12F, "supplementing the 0x12B microcode update", after investigating "a limited number of reports regarding systems continuously running for multiple days with low-activity and lightly-threaded workloads". Intel says 0x12F "does not alter the root cause determination", and its own testing against 0x12B, games included, came out within run-to-run variation.

As of October 2026, Intel's support article on the issue (000102331, last reviewed in July 2026) asks for two things: "the relevant Intel Default Settings in BIOS" and "the latest BIOS version containing Microcode 0x12F or later".

Which processors it applies to

Only desktop processors: Intel says its 13th and 14th Gen mobile processors are unaffected. Intel's warranty extension covers these desktop models:

Intel lists as not affected: 12th Gen Core desktop and mobile processors, 13th and 14th Gen Core i5 (non-K) and i3 desktop processors, 13th and 14th Gen mobile processors including the HX series, Xeon processors, and Core Ultra (Series 1). The Arrow Lake and Lunar Lake families that came after are not affected either. Intel also says most 13th and 14th Gen desktop processors are not impacted, and still recommends the latest BIOS and Intel Default Settings to every owner of one.

Getting the fix onto your machine

The fix is delivered in the BIOS. Intel said 0x129 "will not be available through operating system updates", and that 0x12B "must be loaded via BIOS update".

1. Find out which board you have. Intel's BIOS guide suggests opening System Information (msinfo32) and reading the BaseBoard Manufacturer and BaseBoard Product lines. On a prebuilt desktop, the BIOS comes from the system manufacturer.
2. Get the latest BIOS for that exact model from the board or system maker's support page, or look the board up in Intel's Product Compatibility Tool at compatibleproducts.intel.com. As of October 2026, Intel's requirement is a version containing microcode 0x12F or later.
3. Before you flash, write down anything you changed in the BIOS. Intel warns that updating "resets BIOS settings to their default values", so a memory profile you set earlier will need setting again. Intel also lists the risk of "corruption if power is lost during the minutes-long update or if a crash occurs".
4. If BitLocker or device encryption is on, suspend protection first or make sure you have the recovery key. Microsoft lists "a BIOS or UEFI firmware upgrade" among the changes that can trigger BitLocker recovery; Secure Boot and TPM for anti-cheat explains both steps.
5. After the update, select the Intel Default Settings profile in the BIOS, which Intel recommends to all 13th and 14th Gen desktop owners.

Running above those defaults is your decision, but Intel notes that overclocking "may void their warranty and/or affect system health", and power delivery above Intel's guidance is the first of the four conditions in the table.

If the crashes started before the update

Install the update anyway, and know its limit. Intel describes the voltage cap in 0x129 as "a preventative mitigation for processors not experiencing instability symptoms". RAD is blunter about chips that are already unstable: "once this deterioration has occurred, it is sadly irreversible", and "Once the crashes start, it's too late for a software solution!"

If the crashes continue on the new BIOS at Intel Default Settings, and the other causes have been ruled out, the processor is a candidate for an exchange. As of October 2026, Intel extends the warranty by two years for eligible processors, "providing coverage for up to five (5) years from the original date of purchase". The warranty details say who to contact:

• A boxed processor you bought yourself: Intel Customer Support.
• A tray processor: the place you bought it.
• A prebuilt desktop from an OEM or system integrator: the system manufacturer.

The extension applies to new and previously purchased processors from the list above, to customers globally, and Intel's standard warranty process and terms apply. If an earlier exchange request for a boxed processor was turned down, Intel asks you to contact Intel Customer Support again.

What does not help

• Waiting for Windows Update. Intel's route for these microcode updates is the BIOS.
• Replacing the graphics card, or reinstalling its driver, because of an "out of video memory" message. RAD says that message can come from the processor. If you might really be short of video memory, running out of VRAM shows how to check before you buy anything.
• Reinstalling Windows or the game on a processor that has already degraded. By RAD's account, no software reverses the damage.
• Adding voltage to make the crashes stop. Intel says the affected circuit is vulnerable to aging under elevated voltage, and that elevated voltage events contribute to the rise in Vmin.

Related

• Why did my game crash? Reliability Monitor and Event Viewer
• CPU and GPU temperatures while gaming
• Is your RAM at its rated speed?
• Running out of VRAM
• Sudden FPS drops in games