How do I run a full diagnostic on my PC?

How to Run a Full Diagnostic on PC: Disk Tab & SSD Failure Rate

How to run a full diagnostic on my pc is essential for detecting early signs of drive failure before data loss occurs. Ignoring performance issues like sluggish response or high disk activity leads to costly repairs. Understanding these diagnostic steps helps you maintain system health and avoid unexpected downtime.

A Quick Guide to Running a Full Diagnostic on Your PC

To run a full diagnostic on your PC, start with the built-in Windows Memory Diagnostic for RAM, use Resource Monitor for real-time performance tracking, and execute the SFC and DISM commands in Command Prompt to repair system files. These native tools can identify and help address common system instabilities without requiring third-party software installation. ^[1]

Running these diagnostics is essential because software bugs and driver issues are frequent causes of system crashes, while hardware failures (such as failing drives or components) account for a significant but varying portion - often surfacing as unpredictable blue screens or sudden freezes. Identifying the root cause early prevents minor component issues from escalating into total system failure. But there is one specific, hidden diagnostic step that most people skip, even though it catches many hidden hardware failures - I will reveal that secret in the hardware health section below. ^[2]

I have been there. Staring at a frozen screen while a deadline looms is a special kind of stress. One time, my hands were actually shaking as I tried to save a document before the next crash. I spent an entire weekend reinstalling Windows, only to find out it was a simple hardware conflict. Dont be like me. Use the systems own brain to tell you what is wrong before you start deleting your files.

Testing Your RAM with Windows Memory Diagnostic

Memory issues are notoriously difficult to pin down because they often look like software errors. RAM failures or data corruption can cause Blue Screen of Death (BSOD) incidents.^[3] If your PC restarts randomly or apps crash without warning, your memory is one important place to check. Windows has a dedicated tool for this that runs outside the operating system environment to ensure total accuracy.

To start, type Windows Memory Diagnostic into your search bar. You will be prompted to restart your computer immediately or schedule the test for the next boot. Once your PC restarts, it enters a blue-screen environment (not the scary kind) and begins a multi-pass scan of every sector of your RAM. This process typically takes 15-30 minutes depending on your total memory capacity. It is a slow process. Be patient.

Lets be honest: waiting for a progress bar to move for 20 minutes is boring. But it is necessary. In my experience, if the tool finds an error in the first two minutes, your RAM stick is likely dead and needs immediate replacement. If it passes the basic test, you might want to try the Extended mode by pressing F1 during the test, though this can take several hours. It is thorough.

Analyzing System Health with Resource and Performance Monitors

If your computer feels sluggish but isnt crashing, you need to see what is happening under the hood in real-time. Resource Monitor provides a much deeper look than the standard Task Manager. It shows exactly how your CPU, Disk, Network, and Memory are being utilized by every single process. This is the best way to catch ghost processes that hog resources in the background.

You can open this by searching for Resource Monitor or clicking the link at the bottom of the Performance tab in Task Manager. Look specifically at the Disk tab. If your Highest Active Time is consistently at 100% even when you arent doing anything, your drive may be failing or deeply fragmented. Modern SSDs have an annual failure rate of about 0.5% to 1.2%^[4], which sounds low until it happens to your primary drive.

I used to ignore these graphs. I thought they were just for IT pros. Then I realized my PC was slow because a single browser extension was eating 4GB of RAM in the background. Seeing it visually in a graph made the solution obvious. Sometimes the diagnostic isnt about finding a broken part, but finding a broken habit or a bad piece of software. It makes sense.

Command Line Diagnostics: SFC, DISM, and CHKDSK

The Command Prompt is your most powerful ally for fixing software-based corruption. Many hardware problems are actually just corrupted system files. The System File Checker (SFC) tool scans all protected system files and replaces corrupted ones with a cached copy. It is the first thing you should run if Windows feels glitchy or icons are missing.

Open Command Prompt as an administrator and type sfc /scannow. If SFC finds errors it cannot fix, you follow up with DISM (Deployment Image Servicing and Management). DISM actually connects to Windows Update to download fresh, healthy versions of system files. Using these two in tandem can be effective in fixing many cases of operating system instability without a full reset. ^[5]

Wait - dont forget the hard drive. Running chkdsk /f is the classic way to repair file system errors on your storage drive. It requires a restart and can take a while if you have a large HDD. On modern NVMe drives, it finishes in seconds. It is a quick win. Ive seen CHKDSK save a dying laptop simply by fixing a corrupted index that was preventing the OS from finding essential boot files.

The Overlooked Diagnostic: BIOS and Physical Health

Here is the secret I mentioned earlier: nearly 45% of hardware-related system crashes are caused by thermal throttling or outdated firmware, not broken components. ^[6] If your PC passes all the software tests but still shuts down during heavy tasks, it is likely overheating. Dust accumulation in a laptop or desktop can increase internal temperatures significantly in just a few months.

Restart your computer and tap F2 or Del to enter the BIOS/UEFI. Almost every motherboard has a System Health or Monitor tab. Check your CPU temperature here. If it is idling above 50 degrees Celsius, your cooling system is failing. Additionally, verify that your BIOS version is current. Manufacturers release updates that fix power delivery issues and hardware compatibility which can stop random reboots instantly.

I once had a PC that crashed every time I played a game. I ran every software test known to man. Nothing. Then I opened the case and found a literal wall of cat hair blocking the GPU intake. A 5-minute cleaning with compressed air fixed a problem I had been fighting for a month. Sometimes the best diagnostic tool is a flashlight and a can of air. Truly.

Built-in Windows Tools vs. Manufacturer Diagnostic Software

Windows Built-in Tools

Always available; requires no downloads or internet connection to start.

Limited; cannot always test specific sensors or proprietary hardware features.

General purpose; excellent for OS and system file corruption.

Manufacturer Software (e.g., Dell SupportAssist, HP PC Hardware Diagnostics)

Usually pre-installed; requires manufacturer-specific drivers to function properly.

Deep; can pinpoint exactly which hardware component is failing for warranty claims.

Highly specific; tests motherboard circuits, fan speeds, and battery health with high precision.

Minh's Laptop: The Mystery of the Slow Startup

Minh, an IT student in Ho Chi Minh City, noticed his laptop taking 5 minutes to boot. He assumed his 3-year-old SSD was dying and was ready to spend 2 million VND on a replacement.

He first ran the manufacturer's disk check, but it came back 'Healthy.' Frustrated, he tried to reinstall his heavy IDEs, but the installation kept hanging at 90%, wasting his entire evening.

Instead of buying a new drive, Minh ran Resource Monitor and noticed 'System' was reading thousands of tiny corrupted log files. He realized the issue was logical, not physical.

After running 'chkdsk /f' and 'sfc /scannow', the corrupted file index was repaired. His boot time dropped to 15 seconds, saving him the cost of a new drive and a week of stress.

The 2 AM Server Scare

A small design studio faced random reboots on their primary workstation every time they rendered a 4K video. The team lead feared the expensive GPU had failed.

They initially tried updating drivers, but the crashes continued. They were convinced a hardware replacement was the only path forward, which would have stalled their project for a week.

They decided to check the BIOS thermal monitor during a render. They saw the CPU temp spike to 95 degrees Celsius before the shutdown occurred, indicating a thermal trip.

It turned out the thermal paste had dried out. A $10 tube of paste and 20 minutes of work restored full stability, improving rendering speeds by 20% due to the removal of thermal throttling.