When temperatures are favorable, what maximum voltage can be achieved?
When temperatures are favorable, what maximum voltage can be achieved?
I'm currently operating at 4.99GHz with 1.515v on my 6600k. Temperatures are stable and at 65c under load. I'm wondering when it's safe enough to reduce the frequency even though temperatures are fine. Based on my tests, around 1.53v seems necessary for stable 5GHz performance, but I'm not sure if that's safe at all temperatures. No clear guidance seems available here.
As discussed earlier, with contemporary CPUs 1.6V becomes a threshold where harm might occur without LN2 cooling, even though most CPUs would likely be affected by that point regardless of cooling.
I'm not sure if you'll achieve 65C under load at 1.5V without LN2 cooling. Just keep pushing the clock until the benchmark fails.
Mr Kagouris :
I really doubt you're getting 65C under load with 1.5V unless you have LN2 cooling. Anyways push the clock up until the benchmark crashes.
That is about what I am getting. I found that a bit strange too, but I'm guessing it just is some combination of things working together to help (its a test bed with the cpu on a 360 rad and the vrms/mosfets on water too).
At the moment this is as high as I can get of a clock at 1.515v, I want to try and get that bit extra to run 5GHz but don't know what kind of stopping point I should find for the voltage.
As long as you stay below 1.6V and avoid excessive heat, it should work. Start with 1.52V at 5GHz.
In theory, you may apply as many volts as desired provided you use good components and maintain a stable temperature. A rise in voltage itself doesn't harm parts; it's the resulting and uncontrolled heat buildup that causes damage.
I track your CPU temperature and the specific application running. They advise keeping it below 1.4V to preserve its lifespan.
Can't seem to find a quote or reply button on mobile so here comes a really poorly written reply!
Mr kagouris: I'll try that once I get home from work. I think the most I will try going is 1.54 or 1.55 if my temps allow it still. Unfortunately I have to work today ;-; so once I get back I'll jump on that!
Bycicle: From what I have researched there is some point where even with good temps the voltage can still do some pretty good damage to the lifespan of the cpu. What point that is and how much it decreases it I don't know and that's the tough part xD
Max: I have seen 1.45 as a max, 1.52, theoretically no max also, depends who you talk to. Most say if your temps are still good you can push it further, but they never really give a "this is just too far point". I have used the bios temps, msi command center, intel xtu, and that system temps program every youtuber uses that I forget the name of lol. It seems they are all reporting it just fine, and when I put my hand over different areas of my computer it doesn't feel like anywhere has more heat being outputted than what is being reported.
As discussed earlier, with contemporary CPUs 1.6V becomes a threshold where harm might occur without LN2 cooling, though most CPUs would likely be affected by that point even without it. After some trials, even at 1.55v, I wasn't able to achieve the additional 7MHz required to reach 5GHz ;-;