OCCT Power Test and PSU Diagnosis What the Result Really Proves
The OCCT Power test loads CPU and GPU together to stress the power supply at maximum demand. Used honestly, its result is powerful - but it is not a simple pass or fail certificate. Here is how to read it correctly.
✍️ Technical Anshu🧩 Windows 10 / 11⏱️ 9 minute read
🎥 Watch the full video - every OCCT test shown live, step by step
OCCT's Power test is the closest thing to an exam for your power supply: it runs the CPU and GPU loads at the same time so the PSU faces its highest realistic demand, while OCCT graphs the voltage rails live. Many users treat the result as a simple pass-or-fail certificate. It is not. This page explains what a failure strongly proves, what a pass does not prove, and how to read the voltage graphs honestly.
OCCT Power test - example voltage rails under full combined load
+12V rail11.35 V
+5V rail4.98 V
+3.3V rail3.28 V
CPU powerMAX
GPU powerMAX
Example rails during a Power test. The +12V line feeds the CPU and GPU, so a sagging or jumping 12V line under load is the reading that matters most.
1
What the Power Test Actually Does
The Power test starts OCCT's CPU load and GPU 3D load together. Almost no real program does this so completely - a game comes close, most apps do not. The idea is simple: if the power supply has a weakness, the highest combined demand is where it will show, as a restart, a shutdown, or a voltage rail that cannot stay steady.
Because the load is extreme by design, treat the test as a diagnostic, not a routine: 10 to 15 minutes, supervised, is enough for an answer.
2
Reading the Voltage Rails
The +12V rail powers the CPU and graphics card. Under load it should stay close to 12V. The ATX tolerance is plus or minus 5 percent, so below roughly 11.4V is out of spec.
A line that slowly sinks as the test runs suggests a PSU struggling with heat or age.
A line that jumps up and down suggests poor regulation - common in very cheap or dying units.
Software sensor readings are approximate. They are excellent for spotting a sick trend, but only a multimeter gives laboratory accuracy. Act on clear sagging or a restart, not on a 0.1V difference.
3
What a Failure Proves - Strong Evidence
If the PC restarts or shuts down during the Power test while temperatures were normal, and the same PC passes the CPU and GPU tests run separately, the conclusion is strong: the power supply cannot deliver the combined load. This pattern - fine alone, dead together - is the classic fingerprint of a weak PSU.
A collapsed 12V rail at the moment of failure tells the same story. At that point, further testing of other parts is mostly wasted time until the PSU question is settled with a known-good replacement.
4
What a Pass Does Not Prove - The Honest Part
A pass means the PSU survived one session, at today's room temperature, with today's dust level. It does not promise next month.
Intermittent PSU faults are real: a unit can pass an hour and still drop out during a long gaming session on a hot day.
The Power test also cannot judge long-term quality - ripple and ageing inside a cheap PSU are invisible to this test.
So use the result correctly: a fail is a strong signal to replace or swap-test the PSU; a pass only lowers the suspicion, it does not erase it.
🔴 Read this before any stress test: a stress test pushes hardware harder than any game. Stay with the PC while it runs, keep the temperature graph on screen, and stop the test at once if you smell burning, hear crackling, or see temperatures racing past 100°C. Always test at stock settings first (no overclock), and never run a long Power test on a power supply you already suspect is faulty — on a suspect PSU, a short supervised run is the most you should attempt.
5
If You Suspect the PSU, Test Differently
If the PSU is already a prime suspect - random restarts, a burning smell once, visible age and dust - do not run repeated or long Power tests on it. A failing PSU under maximum load can take other parts with it when it finally dies.
Run one short supervised Power test (about 5 to 10 minutes) purely to confirm the pattern.
Stop immediately on any smell, noise, or sharp voltage drop.
The safest confirmation is a swap test: a known-good PSU of equal or better quality. If the restarts vanish, the diagnosis is done.
Never open a PSU yourself - the capacitors inside can hold a dangerous charge long after unplugging.
🎥 Watch the Complete Video for Full Knowledge and Installation
Every test in this guide is shown live in the video - download, setup, CPU, GPU, RAM and PSU testing with real readings.
My +12V rail shows 11.5V under load. Is my PSU dying?
Not automatically. Software readings are approximate, and the ATX limit is about 11.4V. Watch the behaviour: a stable 11.5V that never moves is very different from a line that keeps sinking or jumps around. Combine the graph with symptoms - restarts under combined load are the real warning sign.
The Power test passed but my PC still restarts in games. Is the PSU cleared?
No. A pass lowers suspicion; it does not clear the PSU. Intermittent and heat-related PSU faults can hide through a short test. The honest final check is a swap test with a known-good power supply.
Is it safe to run the Power test on an old PSU?
Only briefly and supervised, and only if there are no danger signs like a burning smell or crackling. A PSU you already strongly suspect should be swap-tested or replaced rather than torture-tested, because a failing unit can damage other parts when it dies.
This guide is based on the Technical Anshu video on testing a PC with OCCT • Stuck on any step? Ask in the video comments - you will get an answer there 🙏