Red Screen of Death on Windows: Causes and Fixes

A red screen crash is rare, and it is not the red-coloured cousin of the Blue Screen of Death. There is no stop code, no QR code and no crash dump to read — which is exactly why it is harder to diagnose. What you get instead is a red screen, usually a frozen machine, and no explanation.

The good news: the causes are a short list, and most of them are fixable without replacing anything.

A red screen of death error displayed on a Windows 10 computer — Unlike a blue screen, a red screen usually gives you nothing to search for.

Checked August 2026. Applies to Windows 10 and Windows 11 — the causes and fixes are the same on both. Note that Windows 10 left support on 14 October 2025, and consumer Extended Security Updates end on 13 October 2026, so a Windows 10 machine crashing today is worth moving to Windows 11 as part of the fix.

What a red screen actually is

Windows itself has no “red screen of death”. The colour comes from whatever crashed, and there are three distinct things people call by this name:

A graphics driver crash. By far the most common. The GPU stops responding, and what you see is the last thing the display pipeline managed to put on screen — sometimes a red or magenta field, sometimes a corrupted image. Often paired with the fans spinning up to full.
A firmware or bootloader error. Some UEFI implementations use a red screen for boot-time errors — a failed Secure Boot check, a corrupt boot entry, a hardware initialisation failure. These appear before Windows loads at all.
A hardware fault under load. Unstable memory, an overheating or failing GPU, or a power supply that cannot hold up under load. These typically strike during games or rendering rather than at idle.

Which of the three you have is mostly determined by when it happens, so start there.

Narrow it down by when it happens

When the red screen appears	Most likely cause	Start with
Before the Windows logo, every time	Firmware, boot configuration or storage	Reset UEFI to defaults; check boot order
Only in games or under 3D load	GPU driver, overheating, power supply	Temperatures, then a clean driver reinstall
Randomly, including at idle	Memory, overclocking, or storage	Memory test; remove any overclock
Started after a specific change	That change	Roll it back or restore
When connecting a second display	GPU driver or the cable/port	Different cable and port, then driver

1. Remove any overclock

Do this first if it applies, because it is free and it is a genuinely common cause. Any overclock — CPU, GPU, or an XMP/EXPO memory profile — should come off while you diagnose.

In the UEFI/BIOS, load Optimized Defaults (or Setup Defaults) and save.
In vendor tools like MSI Afterburner or the AMD/NVIDIA control panels, reset the GPU clock and voltage curve to stock.

Memory profiles deserve a specific mention: XMP and EXPO are factory overclocks. They are advertised as one-click and mostly work, but an unstable memory profile produces exactly this kind of intermittent, code-free crash. Run at stock speed until you have ruled it out.

2. Check temperatures

Use HWiNFO, HWMonitor or your motherboard vendor’s utility, and watch under load rather than at idle. Rough guidance:

CPU: sustained above 95 °C means throttling and instability.
GPU: above roughly 85 °C for the core is worth investigating; hotspot readings run higher.

If temperatures are high, the fix is physical: clear the dust out of the heatsinks and fans, confirm every fan is actually spinning, check the case has an intake and exhaust path, and on a machine more than about five years old consider replacing the thermal paste.

3. Reinstall the graphics driver cleanly

An update over a damaged driver installation keeps the damage. Do it properly:

Download the current driver for your card from NVIDIA, AMD or Intel before you start. On a laptop, prefer the system manufacturer’s driver for your exact model.
Boot into Safe Mode: Settings → System → Recovery → Restart now under Advanced startup, then Troubleshoot → Advanced options → Startup Settings → Restart → press 4.
Remove the existing driver completely. Display Driver Uninstaller (DDU) is the tool most people use for this; Device Manager‘s Uninstall device with Attempt to remove the driver ticked is the built-in equivalent.
Restart normally and install the driver you downloaded.

Avoid “driver updater” and “PC repair” programs. Advice to fix a red screen with a third-party repair tool is common and worth ignoring. They install generic drivers over vendor-tested ones, and the category has a long history of bundling unwanted software. Everything you need is on the hardware maker’s own download page.

4. Test your memory

Faulty RAM produces crashes with no pattern and no error code — the signature of a red screen at idle.

The built-in tool:

Press Win + R, type mdsched.exe, press Enter.
Choose Restart now and check for problems.

It is a light test. For anything conclusive, run MemTest86 from a USB stick for at least four full passes. If you have more than one stick, test them one at a time — that identifies which one is faulty rather than just telling you one of them is.

5. Check the storage and system files

From an elevated Command Prompt or Terminal:

DISM /Online /Cleanup-Image /RestoreHealth
sfc /scannow
chkdsk C: /scan

DISM before SFC, because SFC repairs system files using the component store that DISM repairs. Also check the drive’s own health — a manufacturer utility (Samsung Magician, Crucial Storage Executive, WD Dashboard) reports remaining life and reallocated sectors more usefully than Windows does.

6. Update the UEFI/BIOS — carefully

If the red screen appears before Windows loads, firmware is a real candidate. Firmware updates fix genuine hardware initialisation bugs, particularly on boards launched with a new CPU generation.

A failed firmware update can leave the machine unbootable. Only use the file for your exact board model and revision from the manufacturer’s own site, never update from Windows if a UEFI-based flashing tool is available, and never update on battery power or during a storm.

Before flashing, try the free option: load Optimized Defaults in the UEFI. That clears misconfiguration without any risk, and it resolves a good share of boot-time red screens on its own.

7. Reseat the hardware

On a desktop, with the machine unplugged and after touching bare metal to discharge static:

Remove and refit the RAM sticks until they click at both ends.
Remove and refit the graphics card, and confirm its PCIe power connectors are fully seated.
Check every power connector at both the drive and the power supply end.

Then test with the minimum that boots: one memory stick, no extra drives, integrated graphics if the CPU has it. If the crash stops, add components back one at a time.

8. Roll back what changed

If the crashes started at a definite point — after an update, a new driver, a new application, a hardware change — undo that change before doing anything else.

System Restore: pick a point from before the trouble. See how to create and use a system restore point.
Driver rollback: Device Manager → the device → Properties → Driver tab → Roll Back Driver.
Uninstall a recent update: Settings → Windows Update → Update history → Uninstall updates.

When it is the hardware

If you have removed the overclock, cleanly reinstalled drivers, passed a memory test, repaired the system files and it still crashes, you are most likely looking at a failing component. In rough order of likelihood:

Graphics card — especially if crashes only happen under 3D load. Test with a different card, or with integrated graphics if available.
Power supply — crashes under load on a machine that is stable at idle, particularly with an ageing or underspecified unit.
Memory — if MemTest86 reports errors, the stick is faulty. RAM usually carries a long warranty.
Motherboard — the diagnosis of last resort, once everything else has been swapped out.

Frequently asked questions

Is a red screen worse than a blue screen?

Not inherently, but it is harder to diagnose. A blue screen gives you a stop code that names the failing component or driver. A red screen gives you nothing, which is why this guide works by elimination.

Can I recover my files after a red screen?

Yes, in almost every case. A crash does not delete data. If Windows will not boot at all, a live USB (a Windows installer’s repair environment, or a Linux live image) will let you copy files off the drive.

Does a red screen mean my graphics card is dying?

Not on its own. Driver problems and overclocking cause it far more often than failing hardware. Suspect the card when the crash only ever happens under 3D load, and when a clean driver reinstall at stock clocks does not stop it.

Why is there no error code?

Because Windows is often not the thing that crashed. A GPU driver hang or a firmware fault happens outside the path that writes a stop code and a crash dump, so there is nothing to record.

Will reinstalling Windows fix it?

It fixes software causes — corrupt drivers, a damaged system image, conflicting software — and does nothing for overheating, an unstable overclock, or failing memory. Work through the hardware checks first; a reinstall that does not fix it costs you a day.