I'm on the ground dropping bombs. I've got three beers, and I'm about to pour one into my shelter.
I'm on the ground dropping bombs. I've got three beers, and I'm about to pour one into my shelter.
Before and after images of temperature data in HWinfo show similar spikes during p95 or other demanding CPU tasks. This occurred when loading a game into the menu and closing it. Of course, it didn’t reach full speed—still throttled some cores, but never hit 100%. The third screenshot is from p95 blending and was taken in under a minute. Pay attention to the errors as well as the temperature and core throttling patterns. Just capturing a screenshot with the snip tool adds about 2,000 PCIe bus errors each time.
Fractal Celsius+ S36 Prisma AIO set to manual and maximum speed. Three fans on Rad connected via splitters 2-1, then 2-1 into the CPUFAN (not CPUFANOPT). I’d like a custom loop, but it seems too complex. This setup is mainly flexible. I don’t render videos; I mostly work in an office and play games at full power. My 10700k worked with the 4090, though it was a bit slow. I might think about a 420 AIO, but three Rad units? That’s extra for nothing?
The motherboard is attempting to damage the CPU using voltage settings. Maximum core voltages are around 1.45, which should be near 1.35 volts. Access BIOS to adjust core voltage offset to -0.10 and observe the effect. If that doesn’t resolve the issue, try -0.075v. It appears the CPU is bent, with cores close together—typical for a bent unit. An uneven cooling contact causes this problem. The locking mechanism on the LGA1700 chip can warp the CPU. Replace it with a proper contact frame. Thermalright offers an affordable replacement. If the CPU is severely bent, temperatures could drop by about 10°C. This should still not fully explain the freezing behavior. More information on the WHEA errors would be helpful.
Just installed a brand new CPU and memory, only been using them for a few days. I saw similar reviews and thought about whether it’s necessary. Before deciding, I’ll check my paste settings first, just in case the installation method changed the process. If needed, I’ll try the offset now and observe. Alternatively, I can review the BIOS settings to find the right configuration.
I prefer working with BIOS carefully and having solid understanding. This BIOS comes from an MSI board, and it seems I can't tweak it easily. It feels like it's stuck at 1.447V, but I'm wondering if lowering it to around 1.347V would help cooling things down. I'm not sure if that voltage is correct. I have two ATX12V headers connected, and since I'm not overclocking, should I stick with the default? I ran a test yesterday and noticed more RAM and CPU errors.
As the process continues, I gather more occurrences. These happen within the past hour. I’m examining the bios options to adjust and modify settings. Yet I’ll wait until I grasp the situation fully. I need to step outside for a while but will follow up later. Are these issues connected to overheating or unrelated? If I should prioritize voltage first, I’ll focus on that. I recall when the store refreshed the bios, they kept the CPU cooler during the process, possibly to prevent damage from heat. It seems like a common practice, so I didn’t notice any warning signs.
Check this out from khovi's post about the middle of the way down: https://social.technet.microsoft.com/For...itprosetup
It seems there might have been some unexpected changes near the cooler, but we did manage to apply. I'm returning the CPU since it's still within the 30-day window. Keep an eye on what the lottery has in store next. I'm thinking about upgrading the motherboard to something that offers more memory capacity—would an 8000MHz board suit my 7200MHz RAM? Or am I misunderstanding stability requirements? It feels better not to push too hard right now.