Throttling at 105°C indicates overheating.
Throttling at 105°C indicates overheating.
Intel site provides only vague specs without useful details. The board uses a Giga P55A U3, and I need to check if this i3 throttles at different temperatures. My X5650 is running on a P6X58D-E with throttling at 100°C, but I want to see how it behaves on a P7P55D-E. It’s been getting over 100°C before, yet it ran smoothly at around 1.5V. After flashing to the P7P55D-E Deluxe, it dropped to about 1.26V instead of 1.5V. I’m worried about CPU throttling possibly being broken due to crossflash. It’s been running for a few minutes now and still hitting 105°C on one core and 103/4°C on the other.
It's running high because it's processing demanding tasks or handling intensive workloads.
Many Intel processors default to a heat limit of 100°C, though some specified 105°C. The third generation models typically reached 105°C. Check Core Temp or ThrottleStop for verification.
I believe my X5690 might restart between 107-112°C. It didn’t slow down even with static clocks running continuously.
Why 105c though? I see readings below 1.3v that are totally safe, but why the shift to 105c instead of 100c lately? With these high-end 13th gen chips, people using less efficient coolers might be able to push their CPUs to 115c for under 1.2v if the cooling isn't too bad. As long as the voltage doesn’t hit its limits, it’s fine even near the threshold. I’m guessing you turned off some monitoring feature or something similar on your board. That stops throttling but most boards still shut down before reaching 115c, so you won’t damage your CPU just from pushing high loads. It happened once on my P5Q at 107c and it shut down. At 115c I mostly worry about the board itself since those are usually pricier than the parts. Still, if I get a decent build that doesn’t auto-shutdown, I’d like to see how far it goes before it fails.
Absolutely not speedrunning or playing that game. My system really disliked Mother Nature, particularly those polar bears.
Why not 105°C? Most first and second generation models operated at 100°C. Intel increased this to 105°C for most third generation chips, then returned to 100°C starting with the fourth generation and onward. There might have been warranty concerns when they reached 105°C, or perhaps Intel simply preferred a clean even figure like 100°C. Simple to recall. All the Core i CPUs I’ve seen cease updating temperature readings once they hit their upper limit—whether it’s 100°C, 105°C, or 101°C. I recall having a first-generation Xeon W3680 that ran at 101°C. The temperature sensors behave like inverted thermometers. As the core temperature rises, these indicators decrease. When they hit zero, thermal throttling kicks in. If throttling begins at 100°C, the CPU can push up to 110°C or 115°C, but the sensor will just show zero, matching 100°C. The sensors Intel used in earlier Core 2 Duo chips would cycle through values like 3, 2, 1, 0, then jump again—RealTemp employed a method to keep readings more reliable above 100°C. I remember seeing a third-generation 3570K running up to about 105°C during high heat tests; it stayed over 100°C with both CPU cores and the iGPU operating at full capacity and no throttling. I used to enjoy experimenting in science.
It's interesting but not all motherboards shut down automatically at 125°C. The C2Q reached its maximum before failing, not the motherboard itself. As for the temperature, that CPU peaked well above 1.35V—especially since voltage issues become more common at higher voltages. I’m planning similar experiments with my own hardware to verify and share accurate data.
Using Core 2 Duo at 80°C pushed things close to the limit for 100% stability with maximum overclock. Keeping the core temp under 80°C was ideal; anything higher risked random restarts. The Core i line handled heat better, offering solid performance between 80°C and 95°C during overclocking. Some models might still perform well past that point. What I realized is that CPU temperature isn’t the main concern—just whether it stays stable or not. Intel once mentioned their CPUs would shut down about 25°C above the thermal throttling mark, which was separate from motherboard limits. It seems most boards relied on the CPU to manage heat. With a throttling temp set between 100°C and 105°C, shutdowns would likely occur around 125°C to 130°C. It appears some 45nm Core 2 Duo mobile chips were among the first to adopt a 105°C threshold. A bit of margin in temperature helped avoid sudden stops, even if just over 100°C. People seem less willing to push hardware now—old-school methods feel outdated. I still enjoy testing extremes to see how my gear handles it.