AIDA64 Engineer system stability test
AIDA64 Engineer system stability test
Hello everyone, I have a few questions about the AIDA64 Engineer system stability test.
I’m using a Windows 10 PC that was built specifically for me by a local IT firm in Melbourne, Australia. The hardware includes an AMD Ryzen 9 3900X at 3.8 GHz with 12 cores and 24 threads, a Gigabyte X570 Aorus Elite WiFi motherboard, 64 GB DDR4 3200 Team T-Force Vulcan RAM, a 1 TB PCIe Gen4 NVMe SSD, a Gigabyte Nvidia GeForce GTX-1050Ti 4GB graphics card, and a case powered by a 750-watt supply.
Michael, the hardware specialist at my local IT company, recommended running a “System stability test” in AIDA64 Engineer. He advised selecting ‘Stress GPU(s)’ in the top left corner and then starting it from the bottom left. This window lets you track system temperatures, CPU usage, and any throttling.
In a follow-up message, Michael clarified what the test does: it simulates a load that puts stress on the CPU, motherboard, RAM, GPU, and drives—all critical components for failure testing.
I hadn’t run the stress test for months. I finally did yesterday. After the test, I contacted Ryan G. at the IT company, who admitted he wasn’t familiar with AIDA64.
I was unsure whether to enable “stress local disks” before starting the test. After re-reading Michael’s email, I chose to activate that option. Then I ran the test for three hours and 32 minutes, ticking the following boxes: Stress CPU, Stress FPU, Stress cache, Stress system memory, Stress local disks, and Stress GPU(s).
The results were stopped early because an AIDA64 message board mentioned not to use write tests on SSDs due to wear on flash cells. I then ran a shorter test for one hour, 11 minutes, and 57 seconds with the same boxes selected.
I have several questions:
- Were my first and second system stability test results satisfactory?
- How much wear did I cause to the flash memory in my 1 TB PCIe Gen4 NVMe SSD during the three-hour test with “Stress local disks” enabled?
- What should I tick next when performing a system stability test in AIDA64?
- How long should I continue running the test on my PC?
- How can I interpret the results and what constitutes a good outcome?
Aida64 is a top stress tester and completely free. I rely on it for every build.
Both tests show very high CPU temperatures near 90°C, indicating your cooling system is barely holding. You should keep temperatures under 80°C during load. What cooling system do you have?
3. It's hard to judge the wear on your SSD, but I think it's not significant.
The Aida64 forum admin explained that the System Stability Test checks all local disks together, but only if they have a partition table and files on them. This is because the test focuses on file system performance, depending on the presence of files on the drives being tested.
They advise against running write tests on SSDs, as this can damage the flash memory cells.
Aida64 is a top-tier stress evaluator that comes free. It works well with all my projects. The first two results show very high CPU temperatures near 90°C, indicating your cooling solution is barely holding. You should keep the temperature under 80°C during operation. Please note the details of your cooling setup.?
The SSD wear assessment is difficult to determine, but it seems unlikely to be significant. The forum administrator explains that the System Stability Test evaluates all local disks together, provided they have a partition table and files on them. This is because the test focuses on file system integrity. They advise against writing tests on SSDs since they can degrade flash memory cells. However, you can run the Linear Write + Verify benchmark once per new drive to verify data integrity—this only requires one write cycle per cell, which isn’t excessive for memory longevity.?
The initial stress test should include CPU, FPU, and cache checks, with other tests performed separately. Start with a 10-minute duration and halt if temperatures rise above 80°C. After confirming stability, continue with a full hour of stress testing.?
Using AIDA64 alongside HWinfo64 is recommended for monitoring temperature on the die under load and checking rail voltages for deviations beyond ±5%. I don’t run a RAM test; instead, I use Memtest86 from a USB drive for DIMM evaluation.
Hi, thank you for your question. Your understanding is mostly accurate. You should perform a stress test using AIDA64 with only the CPU, FPU, and Cache check boxes for 10 minutes. Check if the CPU temperature goes above 80°C during that first test. If it stays below, repeat the test for an hour and verify again.
Regarding "other tests run separately," you should follow the same procedure for each additional test type.
For the GPU stress test, use the appropriate option and focus on the relevant metrics in the results.
For the system memory stress test, apply the correct settings and monitor the memory usage indicators during the test.
The setup shown accurately reflects the configuration.
Checking the CPU, FPU, and Cache mainly evaluates thermal conditions.
I don’t perform full initial tests at once; instead, I run them individually. The GPU stress exceeds what AIDA64 provides, so I rely on benchmarks like Heaven and Cinebench.
Aida64 is just one of many stress testers in your collection.
AIDA64 with Just Memory is suitable for assessing memory stability during overclocking, either alone or alongside other workloads.
The Memory Latency benchmark tracks the usual delay when the CPU fetches data from system memory—specifically the time between issuing a read command and receiving the data in the CPU’s registers.
I view Realbench as the most realistic performance evaluator; passing it confirms full stability.