PC Diagnostic Tools Test Hardware & Find Errors

The Best Hardware Diagnostic Utilities at a Glance

When your computer starts behaving erratically - random crashes, sluggish performance, or ominous clicking sounds - the cause is often a failing hardware component. Before you rush to replace parts or take the machine to a repair shop, a systematic diagnostic approach can pinpoint the exact problem. The good news is that some of the most powerful hardware analysis utilities available today cost nothing at all.

In this guide, we present ten free diagnostic tools that every PC user should know. Whether you suspect a dying hard drive, unstable RAM, or an overheating processor, these utilities provide the data you need to make informed decisions about repairs, replacements, or when it is time to call in a professional data recovery service.

Storage Diagnostics

Your hard drive or SSD is arguably the most critical component in your system - it holds your operating system, applications, and all your personal data. Detecting storage problems early can mean the difference between a planned migration and a catastrophic data loss. For a deeper look at what happens when a drive fails, see our Hard Drive Failure Emergency Guide.

1. CrystalDiskInfo - SMART Monitoring for HDDs and SSDs

What it does: CrystalDiskInfo reads the S.M.A.R.T. (Self-Monitoring, Analysis and Reporting Technology) data embedded in every modern hard drive and SSD. It displays the overall health status of each connected drive along with dozens of individual parameters such as Reallocated Sector Count, Power-On Hours, Temperature, and Wear Leveling Count for SSDs.

When to use it: Run CrystalDiskInfo as a routine health check at least once a month, or immediately if you notice slow file access, disappearing files, or system freezes. It is especially valuable for monitoring aging drives that are approaching or past their typical three-to-five-year lifespan.

How to use it: Download and install the application, then launch it. The main window immediately displays all detected drives with a colour-coded health status: blue for good, yellow for caution, and red for bad. Click on each drive to see the full SMART attribute table. Pay close attention to Reallocated Sector Count, Current Pending Sector Count, and Uncorrectable Sector Count - any non-zero values in these fields indicate physical degradation.

What the results mean: A "Good" status means the drive is operating within normal parameters. A "Caution" status means one or more parameters have crossed a warning threshold - you should back up your data immediately and plan a replacement. A "Bad" status means imminent failure is likely. In this case, stop using the drive and consider professional data recovery if the data has not been backed up.

2. CrystalDiskMark - Storage Speed Benchmarking

What it does: CrystalDiskMark measures the sequential and random read/write speeds of your storage devices. It runs a series of standardized tests and presents the results in megabytes per second, allowing you to compare your drive's current performance against its rated specifications.

When to use it: Use CrystalDiskMark when you suspect your drive is underperforming - for example, after noticing unusually long boot times or slow file transfers. It is also helpful for verifying that a new SSD is delivering the speeds promised by the manufacturer, or for comparing performance before and after a firmware update.

How to use it: Launch the application, select the target drive from the dropdown menu, choose a test size (1 GiB is standard), and click "All" to run the complete benchmark suite. The test takes a few minutes. Sequential read and write speeds reflect large-file transfer performance, while random 4K speeds indicate how responsive the drive feels during everyday use.

What the results mean: If your SSD's sequential read speed is significantly below its rated specification - for instance, 200 MB/s instead of 550 MB/s on a SATA SSD - this could indicate a firmware issue, thermal throttling, or early-stage failure. For HDDs, consistently low speeds combined with SMART warnings are a strong signal to replace the drive. Our SSD Data Recovery Guide covers what to do when an SSD begins to degrade.

3. HD Tune - HDD Health and Error Scanning

What it does: HD Tune provides a comprehensive overview of hard drive health, including temperature monitoring, a detailed error scan that checks every sector on the disk surface, and a benchmark graph showing read speed across the entire platter area. The free version covers the most essential features.

When to use it: HD Tune is particularly useful when you suspect bad sectors on a traditional hard drive. The full surface scan takes time but reveals exactly which areas of the disk are damaged. Use it after recovering from a system crash, when files become corrupted, or when SMART data shows increasing reallocated sectors.

How to use it: Install HD Tune, select your drive, and navigate to the "Error Scan" tab. Click "Start" to begin scanning the entire disk surface. Green blocks indicate healthy sectors, while red blocks mark damaged areas. The "Health" tab displays SMART data similar to CrystalDiskInfo, and the "Benchmark" tab shows a performance curve across the disk.

What the results mean: A few red blocks in the error scan typically indicate the early stages of surface degradation. If red blocks are concentrated in one area, the damage may be localized and the drive could continue functioning temporarily - but a backup and replacement should be your immediate priority. Widespread red blocks indicate severe platter damage, and continued use risks total data loss.

Memory Diagnostics

Faulty RAM is one of the most commonly overlooked causes of system instability. Defective memory can produce blue screens, random application crashes, and data corruption that mimics storage or software problems.

4. MemTest86 - Comprehensive RAM Testing

What it does: MemTest86 is the gold standard for memory testing. It boots from a USB drive independently of your operating system and runs a battery of pattern-based tests across every address in your RAM modules. This isolation ensures that the operating system itself does not interfere with the test results.

When to use it: Run MemTest86 whenever you experience unexplained blue screens (BSODs), random reboots, or application crashes that do not correlate with a specific program. It is also essential when building a new PC or after installing new RAM modules to verify stability.

How to use it: Download the MemTest86 ISO from the official website, create a bootable USB drive using the included tool, and boot your computer from that USB. The test runs automatically and cycles through multiple test algorithms. A single complete pass takes roughly 30 to 60 minutes depending on the amount of RAM installed. For thorough testing, let it run for at least four complete passes.

What the results mean: Any errors detected during the test indicate defective RAM. Even a single error means the module is unreliable and should be replaced. If you have multiple RAM sticks, test them individually to isolate the faulty module. Defective RAM can cause silent data corruption - files may be written incorrectly to disk without any visible error at the time.

5. Windows Memory Diagnostic - Built-In RAM Check

What it does: Windows Memory Diagnostic is Microsoft's built-in memory testing tool, available in Windows 10 and Windows 11. While less thorough than MemTest86, it offers a convenient first-line check without requiring a bootable USB drive.

When to use it: Use it as a quick initial test when you suspect memory problems but do not have a MemTest86 USB drive ready. It is also useful when you want to verify RAM stability without leaving the Windows ecosystem.

How to use it: Press the Windows key, type "Windows Memory Diagnostic," and select the tool. Choose "Restart now and check for problems." The computer reboots into the diagnostic environment, runs a series of tests, and then restarts into Windows. The results appear as a notification after login - check the Windows Event Viewer under "System" events if you miss the notification.

What the results mean: If the tool reports errors, your RAM is likely defective. However, a clean result does not guarantee perfect RAM - the test is less exhaustive than MemTest86. If problems persist after a clean Windows Memory Diagnostic result, follow up with MemTest86 for a definitive answer.

System Information and Monitoring

Understanding exactly what hardware is inside your computer - and how it is behaving under load - is essential for effective troubleshooting.

6. HWiNFO - Complete Hardware Information and Sensor Monitoring

What it does: HWiNFO is one of the most comprehensive system information tools available. It provides detailed specifications for every hardware component - CPU, GPU, motherboard, RAM, storage, network adapters - and includes a real-time sensor monitoring dashboard that tracks temperatures, voltages, fan speeds, and power consumption.

When to use it: HWiNFO is invaluable for diagnosing overheating issues, verifying hardware specifications before upgrades, and monitoring system stability during stress tests. Run the sensor panel alongside Prime95 or FurMark to observe thermal behaviour under maximum load.

How to use it: Download and launch HWiNFO. You can choose "Sensors-only" mode for monitoring or "Summary" for a quick hardware overview. The sensor panel displays current, minimum, maximum, and average values for every sensor in your system. Look for CPU or GPU temperatures exceeding safe limits (typically above 90 degrees Celsius under sustained load) and voltage readings that deviate from their expected values.

What the results mean: Consistently high temperatures indicate inadequate cooling - a clogged heatsink, failed fan, or dried-out thermal paste. Voltage fluctuations can point to a failing power supply. Abnormal sensor readings under idle conditions are particularly concerning and warrant immediate investigation.

7. CPU-Z - Processor and Motherboard Details

What it does: CPU-Z provides precise, real-time information about your processor, motherboard, and memory configuration. It displays the CPU model, core stepping, cache sizes, clock speeds (including real-time multiplier and bus speed), motherboard chipset, BIOS version, and detailed RAM timings and frequency.

When to use it: CPU-Z is essential when planning hardware upgrades - it tells you exactly which CPU socket, RAM generation, and chipset your system uses. It is also useful for verifying that your processor is running at its rated speed and that RAM is operating in the correct dual-channel or quad-channel configuration.

How to use it: Download and run CPU-Z. The "CPU" tab shows processor details, the "Mainboard" tab reveals the motherboard manufacturer, model, and BIOS version, and the "Memory" tab displays the current RAM frequency and timings. The "SPD" tab shows the specifications programmed into each individual RAM module.

What the results mean: If your RAM is rated for 3200 MHz but CPU-Z shows it running at 2133 MHz, the XMP/DOCP profile has not been enabled in the BIOS. If the CPU clock speed is lower than expected, power-saving features may be active, or thermal throttling could be limiting performance.

8. GPU-Z - Graphics Card Analysis

What it does: GPU-Z provides detailed information about your graphics card, including the GPU model, revision, manufacturing process, driver version, clock speeds, memory type and size, and real-time monitoring of temperature, fan speed, GPU load, and memory usage.

When to use it: Use GPU-Z when diagnosing graphics-related issues such as visual artefacts, crashes during gaming or rendering, or suspected overheating. It is also useful for verifying that a used or second-hand graphics card matches its advertised specifications - counterfeit GPUs are a known problem in the second-hand market.

How to use it: Download and launch GPU-Z. The "Graphics Card" tab immediately shows all hardware details. Click the "Sensors" tab to monitor real-time performance data. Watch for GPU temperatures exceeding 85 to 95 degrees Celsius under load, memory clock speeds that differ from specifications, or a high "PerfCap Reason" indicating that the card is being throttled.

What the results mean: If GPU-Z shows a different GPU chip than what the card's label claims, you may have a counterfeit product. High idle temperatures suggest failed fans or degraded thermal interface material. Frequent thermal throttling under load indicates that the cooling solution is insufficient for the card's power consumption.

Stress Testing

Stress tests push your hardware to its absolute limits, revealing instabilities that may not appear during normal use. They are essential after building a new system, overclocking, or when intermittent crashes suggest borderline hardware stability.

9. Prime95 - CPU Stress Testing

What it does: Prime95 was originally designed to search for Mersenne prime numbers, but its intense computational workload has made it the de facto standard for CPU stress testing. It pushes all CPU cores to 100 percent utilisation and can also stress the memory subsystem, making it effective at revealing instabilities in both the processor and RAM.

When to use it: Run Prime95 after installing a new CPU, applying a new overclock, or replacing thermal paste. It is also the go-to tool when you experience random blue screens or crashes under heavy workloads and need to determine whether the CPU is stable.

How to use it: Download Prime95 and select "Just Stress Testing" when prompted. Choose "Small FFTs" to focus maximum heat and stress on the CPU, or "Blend" to test both CPU and RAM stability. Let the test run for at least one hour - ideally overnight for thorough validation. Monitor temperatures with HWiNFO running simultaneously.

What the results mean: If a worker thread reports an error or stops, your CPU is not stable at its current settings. For overclocked systems, this means you need to reduce the overclock or increase voltage (within safe limits). On a stock-clocked system, Prime95 failures can indicate a cooling problem, a failing CPU, or unstable RAM. If temperatures exceed the CPU's maximum rated temperature (Tj Max), stop the test immediately and address the cooling before proceeding.

10. FurMark - GPU Stress Testing

What it does: FurMark renders an extremely demanding 3D scene - a fur-covered doughnut - designed to push your graphics card to its absolute thermal and power limit. It is the most widely used GPU stress test for verifying cooling performance and overall stability.

When to use it: Use FurMark after installing a new graphics card, replacing thermal paste on the GPU, or adjusting GPU overclocks. It is also useful when diagnosing graphics crashes, artefacts, or suspected overheating under gaming or rendering workloads.

How to use it: Download and launch FurMark. Set the resolution to match your display and click "GPU Stress Test." The application renders the test scene while displaying real-time temperature and frame rate data. Run the test for at least 15 to 30 minutes. Watch the temperature curve - it should rise and then plateau at a stable level.

What the results mean: A stable temperature plateau below the GPU's rated maximum (usually 85 to 95 degrees Celsius depending on the model) indicates healthy cooling. If the temperature continues to rise without levelling off, or if the system crashes during the test, the GPU's cooling is inadequate. Visual artefacts - flickering pixels, colour distortion, or geometric glitches - during the stress test indicate GPU instability or a hardware defect.

Building a Diagnostic Workflow

Rather than running these tools randomly, a systematic approach yields the best results. Start with information-gathering tools like HWiNFO, CPU-Z, and GPU-Z to understand your hardware baseline. Next, check storage health with CrystalDiskInfo and HD Tune. Then verify RAM integrity with MemTest86 or Windows Memory Diagnostic. Finally, run stress tests with Prime95 and FurMark to validate overall system stability under load. Use CrystalDiskMark and GPU-Z sensors to track performance metrics throughout.

This structured approach ensures that you do not waste time replacing the wrong component. A system that crashes during Prime95 but passes MemTest86 likely has a CPU or power delivery issue, not a memory problem. Conversely, random errors across multiple applications that correlate with MemTest86 failures clearly point to defective RAM.

When to Stop DIY Diagnostics and Contact Professionals

Free diagnostic tools are powerful, but they have clear limitations. They can identify a problem, but they cannot always fix it - and in some cases, continued testing on failing hardware can make things worse.

Stop self-diagnosing and contact a professional immediately if:

• CrystalDiskInfo reports a "Bad" health status and you have irreplaceable data on the drive
• Your hard drive is making clicking, grinding, or buzzing noises - physical damage requires a cleanroom environment
• Diagnostic tools detect problems but your data is already inaccessible
• Stress tests cause immediate shutdowns or produce burning smells
• You have experienced data loss and need to recover files from a damaged drive

For storage failures involving critical data, a TÜV-certified data recovery laboratory offers the safest path to retrieving your files. DATA REVERSE operates ISO-certified cleanrooms staffed by engineers with decades of experience across all drive manufacturers and storage technologies. Their professional analysis determines the exact cause of failure and the recovery options available for your specific situation.

Your data is irreplaceable - when free tools reveal a serious problem, trust certified professionals to handle the recovery.