Quote:
Originally Posted by XTK Mike 
In terms of overclocking and heat generation, the general rule of thumb is the lower the better. But I understand what you're saying about having a fixation on 60C. Generally speaking you want to keep it between Tcase and Tjunction and as close to Tcase as possible. Excessive heat, even if it is within Tjunction will, over time, reduce the life of the CPU. Semiconductors are extremely sensitive to heat.
Having said that, you could consider going to a higher RPM fan with greater CFM, but the tradeoff here would be noise. How much you can stand is down to personal choice. Some can stand a lot to get the maxium overclock for their gaming rigs. Others who use their PCs most of the time for more mundane tasks may want to keep the noise levels down. If you have the space, you could check to see if your heatsink supports a push-pull fan configuration. That would allow you to maximize airflow across the cooling fins to keep temps down while possibly sticking with lower RPM fans.
Hope this helps.
Michael
In terms of overclocking and heat generation, the general rule of thumb is the lower the better. But I understand what you're saying about having a fixation on 60C. Generally speaking you want to keep it between Tcase and Tjunction and as close to Tcase as possible. Excessive heat, even if it is within Tjunction will, over time, reduce the life of the CPU. Semiconductors are extremely sensitive to heat.
Having said that, you could consider going to a higher RPM fan with greater CFM, but the tradeoff here would be noise. How much you can stand is down to personal choice. Some can stand a lot to get the maxium overclock for their gaming rigs. Others who use their PCs most of the time for more mundane tasks may want to keep the noise levels down. If you have the space, you could check to see if your heatsink supports a push-pull fan configuration. That would allow you to maximize airflow across the cooling fins to keep temps down while possibly sticking with lower RPM fans.
Hope this helps.
Michael
My heatsink does support push/pull, but from what I've read you only gain 1 or 2 degrees by doing it. Better fans might yield more, but probably not by a lot. I use this rig for both gaming and day-to-day, I don't mind higher CFM fans as long as I can have the BIOS spin them down when they aren't needed.
Quote:
Originally Posted by dragonsamus
Your thermal paste needs 200 hours to burn in. Once it does, temps will be better. How much better? I don't know. 1.3 core voltage is too high for 4.0 GHz. My 2500K runs 4.5 GHz with core voltage at 1.330 stable over 8 hours in prime95. I've never tested longer than that. Max temps were an average of 73.5C using core temp to monitor. This is the Hyper 212 Plus with two blade master fans in push pull. All chips will vary with core voltage but I'm sure yours will do 4.5 with aroundthe same core voltage as mine. Less volts if its a better overclocker and more volts if its worse.
Your thermal paste needs 200 hours to burn in. Once it does, temps will be better. How much better? I don't know. 1.3 core voltage is too high for 4.0 GHz. My 2500K runs 4.5 GHz with core voltage at 1.330 stable over 8 hours in prime95. I've never tested longer than that. Max temps were an average of 73.5C using core temp to monitor. This is the Hyper 212 Plus with two blade master fans in push pull. All chips will vary with core voltage but I'm sure yours will do 4.5 with aroundthe same core voltage as mine. Less volts if its a better overclocker and more volts if its worse.
Yeah, I figured 1.3 vcore was too high, but I just wanted a quick test to see what kind of temps I'd get with an overclock. Even a minute and a half of that, according to RealTemp, caused me to exceed what you're saying with your higher overclock and vcore.
Quote:
Originally Posted by TurboMach1
alternatively right from that same guide, troubleshooting high temps lol
alternatively right from that same guide, troubleshooting high temps lol
I took a look at their troubleshooting steps, thanks!
I've pulled the heatsink off and reapplied the thermal paste. I used the same isopropyl alcohol (90%) and Q-tips that I've been using. I managed to get all of it off (that is, an alcohol soaked q-tip came away clean), but on further inspection I noticed either a...build up of paste, or maybe fibers from the Q-tips stuck in the little Vs where the heatpipes meet. I don't think that's over the CPU itself, but it might just be enough to cause the heatsink to rise a tiny bit and effect cooling? I wasn't able to get it to come out, either. I didn't want to take anything harder to it in case I caused damage though. Even with the re-application, I'm still seeing roughly the same temps that I was before, at the same ambient.
I've been considering just switching to a self-contained water cooling system, something like the H100. It's something I've always wanted to try, and I'm pretty sure whatever is causing the high temps in this case is related to the heatsink. I mean, I have had to reapply it 7 or 8 times recently (from RMAs of the motherboard, mostly). It could be not seating properly, it could be the build-up that I noticed, or it could be that I just don't use the right amount of grease.
One thing that I'm not sure about comes from the Arctic Silver link: "The measurement probe moved when the chip was removed to clean off the previous compound." Is there a quick way to check for that? This is the first time I've seated this chip in this motherboard, but this chip has been in a couple others beforehand. I never noticed anything that seemed able to move on the CPU, but I was never quite looking for it either.
Stormputer
(12 items) |
|
Stormputer
(12 items) |
