Short answer
Run PCHealthLab with the RAM & Memory and System Integrity & Logs areas. The program shows how the memory is installed and configured and pulls the memory error history and Windows Memory Diagnostic results from the event log. The memory test itself is started with the Open Windows Memory Diagnostic fix in the Fix Engine.
Step by step
-
Run the analysis
In PCHealthLab tick RAM & Memory and System Integrity & Logs and press ANALYZE. The program runs as administrator; without that it could not read the event logs.
-
Check what is installed
The RAM area shows slot population, memory type and speed, channels, ECC, usage and the paging file. It also flags disabled XMP/EXPO, where the memory runs slower than it is built for.
-
Look at the error history
From the event log the program pulls memory errors and the result of the last Windows Memory Diagnostic run. The System Integrity area shows hardware errors (WHEA), blue screens with their stop code, and unclean shutdowns.
-
Run the Windows memory test
In the Fix Engine click Run ▸ next to Open Windows Memory Diagnostic. The Windows memory test opens and you schedule the restart there. The test runs outside Windows and writes its result to the event log, where PCHealthLab finds it on the next run.
-
Evaluate and save the report
After the second analysis open Open report. A report with the errors and test results is useful for a memory module warranty claim.
How to read the result
- Windows Memory Diagnostic found an error — the memory is faulty. Test the modules one at a time to find which.
- Blue screens with a different code every time — the typical fingerprint of bad RAM. One code over and over points more to a driver.
- WHEA errors — the hardware itself reported an error. It can be the memory, the processor or the motherboard.
- Modules of different types or speeds — mixed modules can cause instability even when each is fine on its own.
Things to keep in mind
- Restarts are not only caused by memory: overheating, a failing power supply or a failing drive look similar. That is why it makes sense to run the CPU, GPU & Thermal and Storage areas too.
- What preceded each crash is analysed by EventAnalyzer: for crashes and restarts it gathers every event from the ten minutes before and reads the stop code from minidumps.
- PCHealthLab does not have its own memory stress test yet; it is planned. Until then, Windows Memory Diagnostic does the testing.