• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

CRAZY high CPU temps.

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

PabloEscobar

Registered
Joined
Jan 4, 2004
Location
Paradise
Ok, I have a major problem. My CPU tempertures are hitting the high 60's at full load using this folding@home program. Just recently i purchased a new case called the X dreamer that has 2 intake fans in the bottom front and 2 outake fans in the back not including the power untit. It also has a fan on the left panel and a fan on the top of the case.

Right now my set up is with a 5000RPM fan on the front intaking and a 3000RPM fan intaking also on the side panel. The fan on the top that is also 3000RPM and is exhausting along with the one on the back thats only running at 1200RPM :( ... not including the power unit of course. Do you think i have proper air flow?

Now i know this has very little with the CPU temps but it does help a little bit. The cases probe right now is just suspended in the air towards the top of the case and it reads 75F. To me thats pretty cool but i dont know.

Anyways now about my CPU...
Yes, as you probly imagined i have stock fan and the thermal grease that came with it. It runs a little below 4000RPM at a constent. Because of these high temps i recently ordered this HSF called the "TT Volcano 7+" and some "Artic Silver 5". I am hoping that will solve this problem...But will it. Is there other things i have to look into about these high temps? Also im not even overclocking.

As of now my temps are at 62C. Now thats what you call HIGH :eek:

ANY HELP?!?!

Thanks,
Pablo :cool:
 
I believe 75F is around 20C which is very low, to compare my pc is 27C case and 40C processor, same proc, overclocked higher. We're on the same cooling setup, except my fan spins at just above 3,000. To me this sounds like a thermometer error, are you on the latest BIOS?
 
Its not a thermometer error. Acually i just recently downloaded the latest BIOS and the temp's didn't change. Also i have this program my MB came with called "Core Center" which sais the temps are higher... but i beleive the BIOS much more. Right now the program sais 65C and my BIOS sais 62C. So im pretty sure the temps are pretty accurate. Also there is something in this Core Center program that monitors the system temp... which is 40C. The 75F temp i was talking about is coming in threw a probe thats just suspended in the air... I think these are totally different things but i dont know. :(
 
yes but they probably both get there temps fromt he same place, so it still could be wrong. but still im sure your new HSF will lower those temps.
 
The temp sensor on nearly any mobo is a good guess at best, they are not a dependable source for real temp readings.

Does your heatsink feel warm to the touch under heavy load? If it's pretty hot, the new HSF should help out. If not, then the HSF is either mounted incorrectly, or the temp sensor is wrong.
 
Yea it feels burning hot to the touch. Not to where is would be painfull to touch...But where it would be unconvertable after leaving your finger on it.

Tekime, how are you running at 2.3GHz with only a 1700+. Thats crazy man. Did you unlock your multiplyer to do this?
Whats the equation you have thats getting these speeds? Also what are your temps of your CPU...A 1700+ running at 2330Mhz must be working very hard.
 
Good, if it's hot at least it's installed right, and the reported temps could be close. Get that new HSF in pronto. :) You could open up your case and get some fresh air in too, see how it affects your CPU temps to gauge your case temps realistically.

I finally got my 1700+ up to about 2.44 actually.. and it runs stable at 2.4GHz @ 1.85V with an SLK-900 and Tornado. It's mighty loud... I've got 120mm YS Tech and seven other fans on top of that.

*edit: Oh, and the best multi I can do is 222x11 or 2442Mhz; I usually run it at 217*11 or 2387Mhz as it seems happiest there and I can even turn my Tornado down a bit. :)
 
Very nice Tekime, Today i was working on my PC switching around the fans in different positions. Well... i found this position where the temps of my CPU is now lowered to 55C! Thats a 5C temp drop and im very happy.
 
The 55C temp isent at load though... :( So i guess its still pretty darn high. Hopefully this will change with the new HSF coming in the mail :)
 
Doh! Well a 5c drop is still worthy, once you get that new HSF in it should help out. The lower case temps will also help contribute to better CPU temps as well.
 
Hmm, if you are not overclocking that machine those temps are fine. I really can't understand the trend that CPU load temp must be "50C or below" or "60C or below" if no overclocking is done. I mean these processors are designed to handle 85C operation temperature! I know CPU does get more unstable at temp x when the processor is run at a higher than stock speed but stock processors will operate stable at around 80C.

So as long as it is stable temperatures do not matter (yes it may reduce it's life from 20 years to 10 years but c'mon...)
 
breez, I have to agree with you at least partially. You really don't need sub-zero temps just to run your system at stock.

IMHO there are a few good reasons to keep your temps in a more reasonable range though. First of all, the on-board temp sensor in 99.9% of motherboards is completely wrong. They can be anywhere from 5-15c off the actually CPU core temperature, so if you're seeing 65c, it could very well be 75c. Secondly, you can't always account for ambient temps; if you get a hot streak one day the 5c rise in ambient could push your comp over the edge. Lastly, while heat does affect the lifespan of your hardware only marginally, the temperature of your PSU directly correlates to the power efficiency of the unit; e.g. how many watts it can handle. Many PSU manufacturer's will advertise a certain wattage but this is under very cool temps, whereas the realistic operating temperature of the unit brings it's capabilities well below its rated maximum. Increased CPU operating temps bring up case temps and in turn PSU temps.

These conditions might be fine for some people to live with, but I would never, ever, sell a computer to one of my clients with operating temps in the 50's or 60's. And I would always recommend keeping your temps in the 50's MAX if you want to avoid long-term issues.

Again, just my humble opinion anyway...
 
Really only case cooling can affect the case temps because the CPU is still dissipating xx number of watts as heat and that number doesn't change if the CPU is just cooled with a better HS&Fan.

I do have temp monitoring (for emergencies, fan stop or something) and am aware of the inaccuracies of the sensors (one thing I rant from time to time, people comparing temperatures taken with different MB, not ideal at all). But I still follow the rule that temps don't really matter if the thing is stable (with various testprograms getting temps up to 5C hotter than usual). With my NF7-S 2.0 and AXP overclocked, this thing starts to behave unstable around 65 so I keep the temps now 10-15C lower (some headroom for hot summer).
 
I can tell you that the Volcano 7+ works great, I use one to cool a 2400+ OC'd to 2700+ (166x13) with AS Ceramique, and run 33C case/37C CPU idle/40C CPU load (24C ambient room temp). And that's only on medium fan speed. Let us know how it works for you!
 
breez said:
Really only case cooling can affect the case temps because the CPU is still dissipating xx number of watts as heat and that number doesn't change if the CPU is just cooled with a better HS&Fan.

Hm, it would still be generating the same amount of heat in either case, but if you aren't dealing with that heat quickly wouldn't it still be contributing more heat to the area around the CPU than if you had been? I'm not honestly sure how exactly to look at it, but having a 65c chip compared to a 45c chip, I would think that 65c chip would be heating up the surrounding components a lot more.
 
Replacing the HS (or changing the TIM in my case) reduces the amount of bottleneck that the HS is. The watts of heat the CPU radiates will be constant, but the temperature will not be. If the HS cannot remove heat as fast as the CPU is putting it out, the temperature will rise. Once the temperature has risen to the point where the HS is accepting and giving off heat at equilibrium the temerature will stop rising. The heat the CPU radiates does not have to go only through the HS, which is why replacing it can and does decrease temps.

...my first post in cooling since forever ago...
JigPu
 
Back