I've been reading some alarming reports about AM5 CPUs, specifically the 7000 and 9000 series, experiencing burning issues. AMD has even cautioned their motherboard partners to keep voltage levels safe, yet I've noticed that some manufacturers are still pushing voltages like CPU core, SoC, and Vcore beyond AMD's recommended limits. This seems to stem from aggressive factory overclocking practices that aim to market their motherboards as being up to 20% faster than competitors. Personally, I'm using a Ryzen 9 7950X with an MSI Carbon B650 motherboard, set to 105W Eco Mode and a thermal limit of 75°C with a negative -15 PBO curve. I'm concerned that my motherboard may be delivering too much voltage. Can someone help clarify which voltages I should be monitoring in HWMonitor? There are so many different sensors listed, and I want to know which ones are the most important!
4 Answers
I think you're right to keep an eye on Vsoc, especially since AMD has set some AGESA updates that limit it to 1.3V maximum. However, I've heard of cases where CPUs have failed even at lower Vsoc, like 1.1V, which is worrying. I don't believe VDD is the issue since some users have managed to prolong their CPU's life with positive CO adjustments. You might want to check VDDIO as well, just to cover all your bases.
Voltage issues with the 7000 series were pretty big news a year or two back. What’s the age of your BIOS? Watch for CPU VDDCR_SOC Voltage (SVI3 TFN) being less than or equal to 1.30V. Personally, when using my 7950X3D and running EXPO at 5200 MT/s, my SoC voltage sits at 1.25V. When I tried 5600 MT/s, it climbed to 1.3V, so just be careful with those settings!
Just curious, how up-to-date is your BIOS? The reports are mostly older, and you should make sure everything's patched. Keep an eye on that specific SoC voltage, as that seems to be where some of the failures stemmed from!
From what I've seen, Vsoc tends to be the main culprit when it comes to issues like this. For a stock setup, you should be around 1.1V with JEDEC memory settings or about 1.2V if you're using EXPO settings for higher RAM speeds. Unfortunately, it seems some motherboard manufacturers aren't setting the accurate voltages or they might have faulty voltage regulators, which can cause boards to hit 1.3V and beyond, leading to overheating and serious damage. Ideally, you should aim for less than 1.2V with EXPO or less than 1.1V with JEDEC to avoid these risks. That was a common fix back when issues first started popping up with the 7000x3Ds.
That makes sense! I've heard that sticking to those voltage levels can definitely help.
Good to know! I’ll check that out, thanks for the heads up!