High pitch noise detected. CPU status?
High pitch noise detected. CPU status?
I’ve noted the patterns I’ve seen. A sudden restart last week. Last night my display turned black, then shifted to magenta before shutting down completely. (1) I tried to boot without video output. (2) Placed my 2060 in a different slot—nothing happened. (3) Swapped my GPU for an older one, but it didn’t respond in the main PCIe lane. (4) Moved it again and got a weak signal. One SATA SSD was disconnected. During startup 1-4 I heard a high-pitched whine, centered around the CPU. This morning there was another whine when I powered on, showing a “generic PNP monitor” in Windows. I attempted driver updates, but NVIDIA claimed my Windows version didn’t exist—Windows 10 21H1 said otherwise. A driver update was offered, but Windows didn’t recognize my setup. My GPU provided H/W details for an older model, yet Windows couldn’t interpret them. This afternoon, after talking to others, we identified it as a capacitor whine issue. I replaced the parts and got a new motherboard. In the evening everything worked smoothly except for RAM issues—removing both sticks resolved the problem. After installing the new board, everything launched properly, except when using both RAM modules simultaneously. I updated drivers for the new motherboard via MSI Center, checked memory settings, and rebooted. Windows started normally. I opened Firefox, watched videos, and encountered a white screen during startup. After rebooting, everything stabilized. I launched MSI Center again to verify updates; everything appeared fine. I also tried changing settings like data collection and minimized the interface. The CPU performed stress tests, but I plan to run GPU tests next. Edit: I forgot that the issue doesn’t resolve long-term—the video plays as long as it runs. I removed all drives except the boot drive, sound card, and WiFi card; I also swapped the PSU last night to narrow down the cause. Update 01/30: ran Prime95 for two hours without issues. Ran Cinebench 2024 with both CPU and GPU, no problems. Watched videos for an hour—only my own actions caused crashes. In the morning, after booting, I saw a 5-minute grey screen, then a brief tan screen, followed by a desktop again in 30 seconds. 01/31: The system crashed shortly after the last update. I tested another PC to check the PSU; it was still dead. I swapped the GPU and managed to boot. After installing the new board, everything functioned correctly. CPU, RAM, PSU, motherboard—everything is back to normal a few days ago. This afternoon, I reloaded the Parts Cannon with a B580 and it worked! I returned to the new motherboard, MSI B550 M-VOC. The 2060 didn’t boot in the donor system. I reused the old board (B450 Strix) with the 5700X, 32GB RAM, GT 460, and P5 850. The system is now operating as expected. CPU, RAM, PSU, motherboard—all stable. This afternoon, I refreshed the Parts Cannon using a B580 and it worked perfectly! I removed the CPU, inspected the pins, everything seemed fine. I reinserted it, but it still didn’t post. I installed my old 2600 and placed it back. Fortunately, I have an extended warranty, and it’s only been about four months since purchase. So far, the new motherboard, GPU, and CPU are functioning properly. It might be worth trying a BIOS update to fix the 5700X issue, though the board was marketed as compatible with the 5000 series and had previously passed stress tests. BIOS updates aren’t always automatic, unless there’s a pattern of failure. Unless I’m dealing with two consecutive defective boards, I’ll keep going.
The system focuses on the CPU. Most coil whine recordings I've seen match this pattern. It seems the issue might be the CPU drawing excessive current through the board's inductors, which could affect both boards. A faulty PSU is another possibility, though I haven't tested it yet. The parts I haven't changed are the CPU and RAM, and I don’t have spare components for testing.
Unless the power was unstable, that caused my previous board to fail. Checking the old PSU was meant solely to check for noise, but I didn’t let the system run long enough to identify instability, because the whine kept happening—I concluded it was the board issue, not the PSU.
Updates provided here are detailed. Our discussion remains unclear. The old board still complains like a Cap is missing or about to leave. If this was for the PCIe lanes and damaged my CPU and GPU, that could explain it. I’m really hoping it’s not due to the PSU power supply. I’ve been lucky with my ST750F, it’s been over ten years and only developed coil whine recently during testing. Using Seasonic parts in this PSU with a 10-year warranty gives me hope it will last beyond my Silverstone. I haven’t checked the EVGA warranty policy to find out if lost components are covered. I haven’t tested my old CPU on the old board to see if the whine persists. I’ll do that once I resolve the new CPU issue. If the old system starts up without any whine and no parts, I’ll reset the BIOS and reboot to recover at least the cost of the new Mobo. If my 2600 fails, I won’t be able to sell it.
Just a short time ago the mobo arrived, so I tried it out. It didn’t work properly, so I ordered a replacement. I’ll keep an eye on it and update you once things are sorted.
Just finished an OCP/OVP update on RE:RE:3. Not sure if it's the PSU—it only has one 12V rail, so I can't split GPU power. Yet, given the quality of the PSU, I wouldn’t expect much from this setup. I ran Stalker2 at medium-high settings on my 2060 and it ran smoothly until the capacitor problem on my board triggered this thread. I’m keeping this discussion going because the PSU seems to be the main culprit here, unlike other parts which were already replaced. Quick tests for booting showed other components were faulty. This OCP/OVP could be related to RE engine. RE8 worked fine with similar settings; maybe ray tracing on this card is the cause? Also, yesterday I returned from the bathroom and saw an unexpected shutdown in Windows Event Viewer, plus a Pagefile BSOD later. It might be Windows too.
I received an RMA from EVGA and returned just the PSU without any cables. After waiting another week, only three-quarters of the cables arrived. I haven’t tested it with the new EVGA PSU yet. Recently, I installed my 2008 Silverstone ST750f, which has worked perfectly for three weeks except for a slight coil whine when the system is off—just the normal buzz from those cheap USB chargers at 60Hz. I’m still unsure if the PSU or the EVGA testing affected it. There was a possibility that the old GPU drawing too much power caused instability, leading to power draw issues on the PCIe board and eventually damaging the last GPU before it failed. Now my Silverstone is functioning normally. I haven’t run the same stress tests to see if it would crash with the Strider because I don’t have the original specs or an article confirming OVCP support. I’m confident it has OVCP, but I didn’t want to risk losing the PC.
04/19 - I'm sure the VRM Capacitor failed on my motherboard. It's not clear if the PSU or the power supply unit was at fault. The new Evga PSU seems stable so far, and all other parts are fine. More testing is needed. For now, I'll close this to preserve it for future helpers.