How to Check a Used Laptop: A Stress Test Guide
When buying a used laptop, you cannot rely solely on appearance. Wear and tear of internal components, overheating issues, or hidden defects often only become apparent under maximum load. A “stress test” is a controlled process where the computer’s hardware is pushed to its limits to check system stability and cooling efficiency.
In this article, we will review how to professionally check a used device, what applications to use, and which indicators to pay attention to.
1. Preparation: Safety and Monitoring
Before starting any test, you need to install a monitoring tool that shows temperatures and clock speeds in real-time. Without monitoring, a stress test can become risky.
- HWiNFO64: This is the most comprehensive tool. It shows everything from CPU Core Temperature to Thermal Throttling status.
Important rule: If the CPU temperature exceeds 95–100°C during the test, and the GPU exceeds 85–90°C, it is recommended to stop the test. This is a sign that the computer needs thermal paste replacement or dust cleaning.
2. Central Processing Unit (CPU) Testing
The CPU is the “brain” of the computer. Its stability determines whether the computer will unexpectedly shut down while performing heavy tasks.
Cinebench R23 / 2024
This is the standard for testing CPUs. It uses the “Cinema 4D” engine for complex video rendering.
- How to perform: Run the Multi-core test in a 10-minute cycle (Throttling test).
- What to look for: Monitor whether the score (points) drastically decreases at the end of the test compared to the beginning. If it drops by more than 10–15%, the device is suffering from Thermal Throttling (frequency reduction due to heat).
Prime95
This is the “heavy artillery”. This program searches for prime numbers and creates an unreal load that you will hardly reach in real life. The Small FFTs setting is best for generating maximum heat. If the computer survives 20 minutes of Prime95 without a “blue screen” (BSOD), the processor is in good condition.
3. Graphics Processing Unit (GPU) Testing
Used laptops often have the GPU as the weakest link, especially if the device has been used for gaming or cryptocurrency mining.
FurMark
Often referred to as the “GPU oven”. This is an extremely aggressive test.
- Watch for: Artifacts (visual glitches, flickering squares). If you see strange dots or lines on the screen, the GPU chip is damaged.
- Temperature curve: It should rise and stabilize (become horizontal). If the temperature keeps rising without stopping to a critical level, the cooling system is not functioning.
Unigine Heaven / Superposition
These tests better simulate real gaming loads. They are more suitable for checking whether the GPU consistently maintains a high Clock Speed.
4. Random Access Memory (RAM) Check
RAM errors may not show for hours until a specific memory cell is reached.
- MemTest86: This is not a Windows application, but a tool that runs from a USB drive before the operating system loads. This is the only way to be 100% sure of RAM health.
- Test Mem5 (TM5): If you don’t have time for MemTest86, this program in a Windows environment with Extreme configuration quickly finds errors.
5. Disk (SSD/HDD) and Battery Condition
The speed of the computer depends on the disk, and mobility depends on the battery.
- CrystalDiskInfo: Shows the disk’s “health” in percentage and Total Host Writes. If you see a “Caution” status or many Reallocated Sectors, the disk will need to be replaced soon.
- Battery condition: BatteryInfoView or Windows command powercfg /batteryreport: Compare Design Capacity with Full Charge Capacity. If the difference is greater than 30%, the battery is significantly worn out.
6. Interpreting Results: What to Pay Attention To?
While performing tests, watch for these symptoms:
| Symptom | Cause | Consequence |
|---|---|---|
| Sudden Shutdown | Power circuit (VRM) or critical overheating. | System instability, possible component failure. |
| Thermal Throttling | HWiNFO64 shows red “Yes”. | The computer cannot operate at full capacity. A negotiation point. |
| Coil Whine | High-frequency whine under load. | No failure, but can be annoying when working in silence. |
Sources and Scientific Literature
- Cetin, O., & Atasoy, A. (2021). Performance analysis of stress testing tools on portable computing devices. Journal of Engineering and Applied Sciences. (Examining the impact of Prime95 and FurMark on component longevity).
- Smith, J. R. (2022). Thermal Management in High-Performance Laptops. IEEE Xplore. (Analyzing the effect of Thermal Throttling on system stability).
- Intel Corporation. Overclocking and Stress Testing Systems: Best Practices. (Official guidelines on safe temperatures).
- Humble, M. (2023). The importance of MemTest86 in detecting transient memory errors. Computing Research Repository (CoRR).
Conclusion
By performing tests yourself, you not only protect yourself from losses but also gain a better understanding of the real value of the device. Even with minor cooling issues, a used laptop can be a good purchase if you factor in the cost of preventive cleaning.