H60 Safe Temps?
H60 Safe Temps?
I completely disagree again; the primary cause of Haswells locking up appears to be temperatures above 75°C. In fact, Haswell is regarded as more problematic than Ivybridge when it comes to thermal concerns. The thermal and voltage specifications are closely related—raising the voltage will raise the temperature, and once you exceed that critical 75°C threshold, crashes become much more frequent. Intel’s maximum rated temperature is around 72.72°C.
Not supported by data. The maximum temperature is 100C at best, but it starts to throttle around 98/99. It might take some time, but if you read this you'll notice it's not happening as described. I've slowed things down and even turned off the fans, reaching temps up to 90C+. I've monitored six temperature points and have spreadsheets with voltage and temperature data. Many people are running high 80s on Hyper 212s.
The maximum temperature is 72.72 throttling happens around 105°C. We observed varying outcomes; I've overclocked several units across different builds and boards, and noticed that at high overclocks (4.8+), issues start around 75°C, causing them to complain.
I'll say.... Asus mentioned they checked more than 1000 Haswell chips and less than 10% succeeded in reaching 4.8C, so if you have several units doing that in a sample of 8, it means you either had an extremely lucky chance or I should probably meet your buddy at Intel.
I've gone through every post in that OCN thread and nobody has mentioned this problem, they joke about setting a 75C / 1.4 v limit when they're running at 1.5 and 85-95 C all the time... out of everyone who reported results, only 11% of those got 4.8 Ghz... and naturally the people getting 4.2 aren't eager to have their names listed.
https://docs.google.com/spreadsheet/ccc?...ring#gid=0
http://www.overclock.net/t/1398276/haswe...emp-limits
I've adjusted every single setting, which has reduced the temperatures by roughly 15 degrees on average. It might be related. I don't really trust the threads you're reading; based on my experience and several other threads I've seen, maintaining low temperatures is crucial for successful overclocking. Voltage is important too, but you eventually hit a limit where you need significantly more than what's comfortable to push the multiplier further. That's why some people go extreme, like using liquid nitrogen in competitions.
I've adjusted every single one of them, which has reduced the temperatures by roughly 15 degrees on average. That might explain it. I don't really trust the threads you're reading; based on my experience and several other threads I've seen, maintaining low temperatures is crucial for successful overclocking. Voltage is important too, but you eventually hit a limit where you need significantly more than what I'm comfortable with to push the multiplier further. Temperature remains the main constraint. That's why some people resort to extreme measures like using liquid nitrogen in competitions.
Maintaining low temperature is essential for good overclocking – it’s not exactly the same as being the key factor, but the goal is to eliminate any obstacles that could hinder a strong overclock. The best approach is to minimize all possible hindrances, allowing your processor to perform at its peak, with cooling being one of those critical elements.