How to Test a Used gaming PC Before You Buy (40-Minute Check)
A used gaming PC can be a great deal — or someone else's problem. This is the exact 40-minute test to run before you hand over cash, with the real tools and the red flags that mean walk away.
The test kit
Cheap, Prime-fast tools that make this test reliable. (affiliate)
- USB flash drive with a bootable test suite →run MemTest86 and a drive SMART check even if the OS is wiped
- Compressed-air duster →clear dust so thermals reflect the hardware, not a clogged build
- Smart plug / inline watt meter →see total system draw and catch a PSU that can't hold load
The step-by-step test
1. Confirm the parts are what was advertised
Boot to the desktop and run CPU-Z and GPU-Z. Verify the CPU model, core/thread count, RAM size and speed (and that all sticks show up), and the actual GPU chip and VRAM. Open the case: confirm the GPU, PSU wattage label, and motherboard match the listing. Reused listings frequently photograph a better build than what's inside.
2. Check the drives' health with CrystalDiskInfo
Run CrystalDiskInfo. Each drive should report 'Good'. On SSDs note Percentage Used / Total Host Writes (TBW) — a drive at 80%+ wear is near end-of-life. On hard drives, any Reallocated Sectors, Pending Sectors, or Uncorrectable count above zero is a failing drive. Power-On Hours tells you how hard the machine actually lived.
3. Stress the CPU and watch thermals
Run Cinebench R23 (multi-core) or Prime95 for 10 minutes with HWiNFO open. The CPU should hold its boost clocks without hitting its thermal limit (typically 95-100°C is the throttle point on modern chips). If it slams into 100°C in seconds and downclocks hard, the cooler is failing or the paste is shot.
4. Stress the GPU for artifacts
Run OCCT or FurMark for 10-15 minutes. Watch for visual artifacts (speckles/corruption) and confirm temps plateau under ~83°C. See the dedicated used-GPU guide for detail — the GPU is usually the most expensive and most abused part of a used gaming PC.
5. Test the RAM with MemTest86
Boot MemTest86 from the USB drive and run at least one full pass (longer is better). Even a single error means a bad stick or unstable XMP profile — bad RAM causes random crashes that are maddening to diagnose later. This is the test most sellers never run.
6. Listen, sniff, and check the I/O
Under load, listen for fan grinding (bad bearings) and coil whine. A burnt-electronics smell means a stressed PSU or VRM — walk away. Then plug into every USB port, the audio jacks, every display output, and Ethernet/Wi-Fi. Dead ports are common on heavily-used machines and a pain to live with.
Red flags — walk away if you see these
- Specs in CPU-Z/GPU-Z don't match the listing
- Any drive reports Caution/Bad in CrystalDiskInfo, or SSD wear over ~80%
- CPU or GPU instantly hits thermal limit and throttles hard
- Any MemTest86 error at all
- Burnt smell, coil whine that wasn't disclosed, or dead USB/display ports
See gaming desktop PC listings on eBay → (affiliate)
FAQ
- What should I check first on a used gaming PC?
- Confirm the parts are real with CPU-Z and GPU-Z, then check every drive's health in CrystalDiskInfo. Mismatched specs and a dying drive are the two most common and most expensive surprises.
- How do I test RAM on a used PC?
- Boot MemTest86 from a USB stick and run at least one full pass. A single error means a bad stick or unstable memory profile — bad RAM causes random crashes that are hard to trace later.
- How hot is too hot for a used gaming PC under load?
- Modern CPUs throttle around 95-100°C and GPUs around 83-90°C. Hitting those limits briefly is fine; slamming into them in seconds and downclocking hard means failing cooling or dried thermal paste.
These are practical buyer checks, not a professional appraisal. For high-value items, get an expert opinion before paying.