A Blue Screen of Death (BSOD) on Windows 10 is your PC's emergency stop — a kernel-level crash that halts everything to prevent data corruption or hardware damage. Every BSOD carries a stop code, and every stop code points somewhere specific: bad RAM, a failing drive, a corrupted driver, overheating hardware. The problem isn't random, and it isn't hopeless.
This guide walks through every major stop code you'll encounter on Windows 10, what each one actually means, and — most importantly — the step-by-step fixes that resolve the underlying cause rather than just rebooting and hoping the problem disappears. We've organized fixes from simplest to most involved, so you can work through them efficiently.
Back up your data before anything else. Recurring BSODs — especially those involving drive-related stop codes — can escalate to total drive failure with no warning. Copy your important files to an external drive or cloud storage now, before you start troubleshooting. It takes minutes and could save everything.
Step 1 — Read the Stop Code
Every Windows 10 BSOD shows a stop code in plain text at the bottom of the blue screen. This is your primary diagnostic tool. If you missed it during the crash, find it here after Windows restarts:
- Event Viewer: Press Win+R, type
eventvwr.msc, navigate to Windows Logs → System, and look for Critical events at the time of the crash. The stop code and faulting module appear here. - Minidump files: Navigate to
C:\Windows\Minidump\. The free tool WhoCrashed reads these files and produces a plain-language crash summary including the responsible driver or module. - Reliability Monitor: Search "reliability history" in the Start menu for a visual timeline of all crashes with dates and descriptions.
Tip — make the stop code visible longer: Windows 10 reboots automatically after a BSOD by default. To disable this: right-click This PC → Properties → Advanced system settings → Startup and Recovery → Settings, then uncheck "Automatically restart." The blue screen will stay on screen until you manually restart, giving you time to photograph the stop code.
Most Common Windows 10 Stop Codes — What They Mean
Stop codes aren't random strings — each one maps to a specific class of failure. Here's what the most frequently encountered codes mean and where to look first:
Faulty or Unstable RAM
Windows detected a memory management error — the most common BSOD on Windows 10. Caused by defective RAM, mismatched sticks, or RAM running at incorrect voltage. Run MemTest86 overnight to confirm.
Driver Accessing Illegal Memory
A driver attempted to access memory at an interrupt level it isn't permitted to use. Almost always caused by a recently updated or newly installed driver — GPU, network, and audio drivers are the most common culprits.
Memory Reference to Invalid Address
Windows tried to access a memory location that doesn't exist. Possible causes: defective RAM, a corrupt or outdated driver, or a failing drive providing bad data reads to system processes.
System Driver or Kernel Error
A privileged kernel operation failed. Often caused by outdated GPU drivers (look for nvlddmkm.sys or atikmpag.sys in the crash dump), corrupted system files, or antivirus software with deep kernel hooks.
Core Windows Process Terminated
A process essential to Windows operation (csrss.exe, winlogon.exe, lsass.exe) exited unexpectedly. Causes include corrupted system files, a failing storage drive, or malware that has tampered with system processes.
Kernel Data Integrity Violation
Windows detected corruption in kernel-mode data structures. Frequently triggered by a Windows 10 feature update hitting an incompatible driver, or by RAM returning corrupted data under load.
SSD Firmware or Storage Driver Timeout
A Deferred Procedure Call ran too long without completing. Very common on Windows 10 machines upgraded from Windows 7/8 without reinstalling storage drivers. Also caused by outdated SSD firmware on NVMe drives.
Hardware-Level Failure
The Windows Hardware Error Architecture (WHEA) detected a hardware fault it couldn't correct. Points to CPU, RAM, or power delivery issues. Can also indicate a failing motherboard or an aggressive overclock. One of the most serious BSOD codes.
Memory Pool Corruption
A driver wrote outside its allocated memory boundaries, corrupting the heap. Commonly caused by faulty or incompatible RAM, a buggy driver, or recently installed software that hooks into kernel memory.
Drive Failure or Filesystem Corruption
The NTFS driver encountered errors it couldn't recover from. A very strong indicator of drive hardware failure — bad sectors, a failing controller, or catastrophic filesystem corruption. Back up immediately.
Driver or Kernel Exception
A kernel-mode driver or Windows component threw an exception with no handler. Usually caused by a specific incompatible driver — the crash dump typically names the offending .sys file directly.
GPU Driver Timeout / Display Driver Crash
The display driver failed to respond within the allowed time and couldn't recover. Almost always fixed by uninstalling the current GPU driver completely (using DDU) and installing the latest stable version.
Best Solutions — Step by Step
Work through these in order. Each fix takes only minutes if it works, and confirms or rules out a category of cause so you can narrow down quickly.
Disable Automatic Restart — Then Capture the Stop Code
Before troubleshooting, make sure you can see the full stop code. Right-click This PC → Properties → Advanced system settings → Startup and Recovery → Settings. Uncheck Automatically restart. Now the blue screen stays visible until you restart manually, giving you time to photograph or write down the stop code.
Once you have the code, search for it specifically. The pattern section above tells you exactly what hardware or software category each code targets — this single step can eliminate 80% of blind troubleshooting.
Roll Back a Recent Windows Update or Driver
If BSODs started immediately after a Windows Update or a driver install, a rollback is the fastest fix:
For Windows Updates: Settings → Windows Update → View Update History → Uninstall updates → remove the most recent one. Restart and test.
For drivers: Open Device Manager (Win+X → Device Manager), expand the relevant category (Display adapters, Network adapters, etc.), right-click the device → Properties → Driver tab → Roll Back Driver. If this option is greyed out, the driver has no previous version saved — proceed to Fix 5 instead.
Repair Windows System Files (SFC + DISM)
Corrupted Windows system files can trigger BSODs with CRITICAL_PROCESS_DIED, SYSTEM_SERVICE_EXCEPTION, and similar codes. Windows has built-in repair tools — run them in this order:
SFC scans and repairs protected system files. DISM repairs the Windows component store that SFC uses as a reference — run DISM if SFC finds unrepairable errors. Both commands take 20–30 minutes; don't interrupt them.
Test Your RAM (Windows Diagnostic + MemTest86)
RAM failures cause a wide range of BSOD codes — not just MEMORY_MANAGEMENT. Test in two stages:
Quick test — Windows Memory Diagnostic: Press Win+R, type mdsched.exe, choose "Restart now and check for problems." This runs before Windows loads. It catches obvious failures in about 15 minutes.
Thorough test — MemTest86: Download MemTest86 from memtest86.com, write it to a USB drive using the provided tool, and boot from it. Run it overnight — at minimum two complete passes. Any errors at any pass count mean the RAM needs to be replaced. If you have two sticks, test each one alone to identify which is faulty.
Windows Memory Diagnostic misses a significant percentage of real RAM failures. MemTest86 is the standard used by professional technicians and is far more thorough.
Update or Clean-Install GPU Drivers
Outdated or corrupted GPU drivers are the single most common driver-related BSOD cause on Windows 10. Don't update through Device Manager — it rarely installs the latest version. Go directly to the manufacturer:
- NVIDIA: NVIDIA.com → Drivers
- AMD: AMD.com → Drivers & Support
- Intel integrated: Intel.com/content/www/us/en/download-center
For a clean install (recommended if you've had recurring GPU-related crashes): download Display Driver Uninstaller (DDU), boot into Safe Mode, run DDU to fully remove the existing driver, reboot normally, then install the fresh driver. This eliminates any leftover corrupt driver files that a standard installer won't touch.
Check Drive Health and Run CHKDSK
Check S.M.A.R.T. data first: Download CrystalDiskInfo (free) and open it. Any drive showing "Caution" or "Bad" status — especially with Reallocated Sectors, Pending Sectors, or Uncorrectable Sectors above zero — needs to be cloned and replaced immediately. Don't wait for more crashes.
Run CHKDSK for filesystem errors: Open Command Prompt as Administrator:
Windows schedules this for the next restart (can't check a live drive). Restart and let it run — it can take 60–90 minutes on a large drive. If CHKDSK finds and fixes many bad sectors, treat the drive as suspect and start planning for a replacement.
Use Safe Mode to Isolate a Software Cause
Safe Mode loads only the essential Windows drivers and services. If your PC runs stably in Safe Mode but crashes in normal mode, a third-party driver or startup program is the culprit — not hardware.
To enter Safe Mode in Windows 10: hold Shift while clicking Restart → Troubleshoot → Advanced Options → Startup Settings → Restart → press 4 (Enable Safe Mode) or 5 (Safe Mode with Networking).
From Safe Mode: uninstall any recently added software, roll back or uninstall suspect drivers in Device Manager, then restart normally to test. If the crash returns only in specific applications, the app itself or its background services are the issue.
Check for Overheating
If BSODs happen consistently during gaming, video encoding, or other CPU/GPU-intensive tasks — but the PC runs fine at idle — heat is a strong suspect. Download HWiNFO64 or Core Temp (both free) and monitor temperatures during a load test.
Safe maximums: CPU below 90°C, GPU below 95°C under sustained load. If you're exceeding these:
- Shut down and blow out all vents with compressed air — dust buildup is the most common thermal cause and takes minutes to fix.
- Make sure all case fans are spinning and the heatsink fan is running.
- If the machine is several years old and cleaning doesn't help, the thermal paste between the CPU and heatsink may have dried out — reapplication by a technician is a straightforward repair with dramatic results.
Update Chipset, Storage, and NVMe Drivers
The DPC_WATCHDOG_VIOLATION code is almost exclusively caused by outdated storage controller or NVMe drivers — especially on PCs upgraded to Windows 10 from an older OS without a clean install. Update these from your motherboard manufacturer's support page (not Device Manager):
- AMD: Look for "AMD Chipset Drivers" on AMD.com
- Intel: Look for "Intel Chipset Device Software" and "Intel Rapid Storage Technology" on Intel.com
- NVMe drives: Check your SSD manufacturer's site for a firmware update (Samsung Magician, Crucial Storage Executive, etc.)
In-Place Windows 10 Repair Upgrade
When SFC and DISM can't fix corrupted system files — or when BSODs persist after all driver and hardware checks pass — an in-place repair upgrade replaces every Windows system file while keeping all your apps, settings, and data intact. This is more thorough than SFC and less drastic than a clean reinstall.
Download the Windows 10 Media Creation Tool from Microsoft.com, run it, choose "Upgrade this PC now," and follow the prompts. Select "Keep personal files and apps" when offered. The process takes 1–2 hours and leaves Windows completely fresh at the system level while preserving everything you've installed.
When the DIY Fixes Aren't Enough
The fixes above resolve the large majority of Windows 10 BSODs. But some situations require professional hardware testing and tools that aren't available at home. Come to us if:
WHEA_UNCORRECTABLE_ERROR
What Our Technicians Do That You Can't at Home
Our Roswell shop uses professional-grade diagnostic tools that go significantly further than the Windows built-in utilities:
Professional Crash Dump Analysis
Microsoft's kernel debugger reads the full memory dump — not just the stop code — and extracts the exact function call stack, faulting driver, and memory address at the moment of the crash. This often identifies the root cause in under five minutes for driver-related crashes, and tells us precisely which hardware to test for hardware-related ones.
Extended MemTest86 Overnight Runs
We run MemTest86 across multiple complete passes (often 6–8+ hours) using all available test patterns. We test individual sticks in each slot to identify both which module is bad and whether the issue is the slot itself — a motherboard problem that a RAM swap alone won't fix.
Full S.M.A.R.T. + Sector Surface Scan
Beyond reading S.M.A.R.T. data, we run a full surface scan of every sector on every drive — a process that finds bad areas that S.M.A.R.T. doesn't yet reflect. For drives showing early failure indicators, we clone the drive to a new one before it progresses to complete failure.
Power Supply Load Testing
An unstable PSU delivering dirty or insufficient power causes BSODs that look like RAM or CPU problems. We test PSU output under realistic load conditions — something no software tool can do — to rule out power delivery as the cause before blaming more expensive components.
How to Prevent BSODs on Windows 10
Once you've resolved the current problem, a few habits keep Windows 10 stable long-term:
- Keep Windows and drivers updated — Enable Windows Update, but know that rolling back a bad update is always an option if it causes problems. Update GPU drivers directly from NVIDIA/AMD/Intel, not through Windows Update.
- Clean your PC annually — Dust blocks airflow and causes thermal crashes. A can of compressed air through the vents takes five minutes and makes a real difference.
- Use a UPS or quality surge protector — Power spikes and sudden outages corrupt file systems and wear on storage hardware. A battery backup costs less than a repair and protects against both.
- Monitor drive health proactively — Run CrystalDiskInfo every few months. Catching a drive at "Caution" before it reaches "Bad" gives you time to back up and replace it without data loss.
- Don't disable Windows Defender — Malware is a real BSOD cause, especially for
CRITICAL_PROCESS_DIED. Defender is surprisingly effective when left running and updated. - Back up regularly — Windows 10 includes File History and Backup for a reason. A current backup turns a worst-case BSOD scenario into a minor inconvenience.
Getting Blue Screens? Computer Repair Roswell Can Fix It
If you've worked through the solutions above and the crashes continue — or if you'd rather have a professional confirm the diagnosis before spending money on replacement hardware — our certified technicians in Roswell are here to help. We use professional crash dump analysis, extended hardware testing, and years of Windows diagnostics experience to identify the root cause and fix it correctly the first time.
We serve Roswell, Alpharetta, Sandy Springs, Marietta, Johns Creek, Milton, Dunwoody, Cumming, and the greater North Atlanta area. Walk-ins welcome. Most BSOD diagnostics are completed same-day.
Free diagnostics, no fix no fee. We assess every BSOD at no charge and give you a plain-language explanation of what's wrong and a firm quote — before we start any work. Our 90-day workmanship warranty covers every repair we perform.
Still Getting Blue Screens After Trying These Fixes?
Our certified technicians read crash dumps, test hardware, and find the root cause — not just the symptom. Free diagnostics, no fix no fee.