The short answer
Write down the stop code, since it is the only useful piece of information on the screen and it narrows the cause immediately. Most blue screens are drivers, and most of the rest are memory or storage. Boot into safe mode to test whether the machine is stable without third-party drivers, which splits the problem in half.
A blue screen means Windows encountered a condition it could not safely continue from, usually caused by a driver operating in kernel space. The screen itself is unhelpfully friendly now, and the stop code at the bottom is the part that matters.
Capturing the information
Photograph the screen
Including the stop code and the file name if one is shown. A file name ending in .sys names the driver responsible, which is the single most useful clue you can get.
Check Event Viewer afterwards
Windows Logs, System, and look for the error at the time of the crash. It records the bug check code even if you missed it.
Use Reliability Monitor
Search for it in the Start menu. It presents a timeline of crashes alongside installations and updates, which often shows the cause visually.
Install a dump reader if crashes repeat
WhoCrashed or BlueScreenView read the memory dump files Windows writes and name the driver involved in plain language.
Note what you were doing
Crashes only while gaming, only on waking from sleep, or only when a particular device is connected are all strong signals.
What the common stop codes point at
- DRIVER_IRQL_NOT_LESS_OR_EQUAL: a driver accessed memory it should not have. The .sys file named is the culprit. Network and graphics drivers are the most common.
- PAGE_FAULT_IN_NONPAGED_AREA: often faulty memory, sometimes a driver or a failing drive.
- SYSTEM_SERVICE_EXCEPTION: usually a driver, frequently graphics or antivirus.
- CRITICAL_PROCESS_DIED: corrupted system files or a failing drive.
- KERNEL_SECURITY_CHECK_FAILURE: commonly memory or a driver incompatibility after an update.
- WHEA_UNCORRECTABLE_ERROR: genuine hardware. Often the processor, memory or an unstable overclock.
- INACCESSIBLE_BOOT_DEVICE: storage driver or controller mode change, common after a firmware or BIOS change.
If the same .sys file appears in several crashes, you have found it. Search the file name to identify which product it belongs to, then update, roll back or remove that software. This single step resolves a large share of recurring blue screens.
The order to work through
1. Undo the most recent change
If crashes began after installing hardware, a driver, an update or a new program, that is where to look first. Windows keeps a list of installed updates that can be uninstalled individually, and Device Manager can roll a driver back to the previous version.
2. Test in safe mode
Safe mode loads a minimal set of drivers. If the machine is stable there for an extended period, the problem is a third-party driver or service rather than hardware. A clean boot, disabling all non-Microsoft services and startup items, then re-enabling them in batches, identifies which.
3. Update the important drivers deliberately
Graphics, chipset, network and storage drivers, obtained from the component manufacturer or the laptop maker rather than a driver updater utility. Automatic driver updater tools are a frequent cause of the problem rather than a solution, because they install mismatched versions.
4. Test the memory
Run MemTest86 from a USB stick for at least four passes. Windows has its own memory diagnostic and it is far less thorough. If you run an XMP or EXPO profile, disable it and test at default speed, because memory overclocking that was stable for a year can become unstable as components age.
5. Check the drive
Run a SMART health check with a tool such as CrystalDiskInfo. Then run chkdsk and, separately, sfc /scannow followed by DISM to repair system files. A failing boot drive produces varied and confusing stop codes.
6. Look at temperatures and power
If crashes happen under load, monitor temperatures during gaming or rendering. Also consider the power supply, particularly on an older machine or after a graphics card upgrade, since unstable power produces crashes that look like anything.
Disable every overclock, including memory profiles and manufacturer performance modes, before spending hours on other diagnosis. It costs nothing to test at stock settings, and an ageing overclock is a common cause of crashes that appear suddenly after a long stable period.
Patterns worth recognising
- Crashes on waking from sleep: almost always a driver, often network or storage. Check power management settings for those devices.
- Crashes only in games: graphics driver, temperature or power supply.
- Crashes at random with no pattern: memory or a failing drive.
- Crashes when a specific device is plugged in: that device’s driver.
- Crashes immediately after a Windows update: a driver that update installed. Roll it back.
- Crashes during large file transfers: storage or memory.
The last resorts
An in-place repair install reinstalls Windows while keeping files and programs, which fixes corrupted system components without the disruption of a clean install. It is a genuinely useful middle step that many people do not know exists.
A clean install is definitive for software causes and useless for hardware ones. If crashes continue after a clean install with only essential drivers, the problem is hardware and further software effort is wasted.
Where do I find the stop code if the PC restarts too fast?
In Event Viewer under Windows Logs, System, at the time of the crash. You can also disable automatic restart in System Properties, Advanced, Startup and Recovery, so the screen stays visible.
Are blue screens always a hardware problem?
No. The majority are drivers. Hardware causes exist, particularly memory and storage, and software is the more likely starting point unless the stop code specifically indicates a hardware error.
Should I use a driver updater program?
Generally no. They install versions that do not match your exact hardware and are a recurring cause of instability. Get drivers from the component or laptop manufacturer.
How long should I run a memory test?
At least four full passes, which usually takes several hours. Short tests miss the intermittent faults that cause occasional crashes.
Will reinstalling Windows fix it?
It fixes software and driver causes. If the cause is failing memory, a dying drive or an inadequate power supply, a reinstall changes nothing, which is why diagnosis first saves time.
In summary
Capture the stop code and any .sys file name, because that points directly at the cause. Undo recent changes, test in safe mode, and update the graphics, chipset and storage drivers from the manufacturer. Then test memory properly and check drive health. Disable overclocks before anything else. Use an in-place repair install before resorting to a clean one.
