I replaced my i3-3220 with an i5-2500, and the installation was a bit troublesome. After fixing the cooler, thermal paste, and a RAM-slot issue caused by poor CPU socket contact, the computer worked normally on the desktop and while updating graphics drivers. However, it shuts down after about five minutes of gaming—twice it powered off completely and once it restarted by itself. This only happens under gaming load, with the RX 560D at 100% usage and the CPU around 80%. Temperatures reach roughly 65°C on the GPU and 85–86°C on the CPU using the stock cooler. The power supply is a 400W unit from a reasonably reputable local brand, and I cannot afford replacement parts right now. What should I test first, and could the issue be overheating, the PSU, the motherboard's VRMs, or something related to the CPU installation?
5 Answers
The i5-2500 has a much higher power draw than the i3-3220, so the motherboard’s VRM circuitry may now be under considerably more stress. Very old or low-end boards can shut down when supplying the newer CPU under sustained load. Reinstall the old i3 temporarily and see whether the same games remain stable. Also reset the BIOS and confirm that the board has the correct BIOS support for the i5-2500.
The hardware is quite old, so both the PSU and motherboard capacitors or VRMs could be deteriorating even if they worked with the previous CPU. A sudden power-off under gaming load is more consistent with protection being triggered by power delivery or temperature than with the game simply being too demanding. Check Windows Event Viewer afterward for a Kernel-Power event, but remember that it only confirms an unexpected shutdown—it won’t identify the exact component.
Since the problem began immediately after the CPU swap, inspect the cooler mounting and socket area carefully. A cooler that is slightly loose or uneven can produce short temperature spikes that software monitoring misses. Cleaning the system, applying fresh paste correctly, and ensuring nothing else is overclocked would be good low-cost steps. A basic replacement tower cooler may also help if you can find one compatible with the motherboard.
A better cooler is probably my next purchase. I’m having trouble finding an inexpensive one that fits this older motherboard, though.
An 85–86°C CPU temperature is warm but not automatically high enough to cause an immediate shutdown. Monitor the CPU package and core temperatures more closely while gaming, and check whether they spike toward the processor’s roughly 98°C maximum. Also verify that the cooler is firmly mounted, the fan is spinning, and the thermal paste is spread properly.
I monitored it with Afterburner and it stayed below about 86°C, so overheating may not be the only problem. I’ll still check the cooler and paste again.
The PSU is one of the strongest suspects. A 400W label does not guarantee that an older or budget unit can deliver stable power under load, especially when the GPU and CPU are both being used heavily. Try running a GPU stress test such as FurMark while monitoring for shutdowns, and also try limiting the game’s frame rate. If a frame limiter prevents the shutdown, that points toward power delivery or heat rather than a normal software issue.
I’ll test it with a frame limiter and a GPU load test. The PSU is from a fairly known brand, but it isn’t one of their higher-end models.
I’ll try the old processor again if the other tests don’t help. The CPU socket pins look okay, but I did have a RAM slot problem during installation.