Inuyasha1771

I have to agree a bit, my laptop runs an E6600, and it consistently is at around 80c, and has been for the last few months..it runs strong anyways. I still don't like my CPU being above 65 though, no reason for it to be above that hot.

__________________
Originally Posted by VCheeZ
That mix of Zotac cooler and blue pcb is the ugliest thing I have ever witnessed that was not made by Gigabyte.[/quote]

http://www.overclock.net/mobile-comp...-2000-obo.html Alienware M17x $2000 OBO!!!

awdrifter

If you don't plan to keep your CPU long then I agree, the temp is not going to be a big factor. But in the long terms, I do believe high temps (over the max temp in the manufacturer's spec) will kill the chip. A better example would probably be found in GPUs. While people say their GPUs can take 100c (or higher), a lot of people that had dead GPUs are overclocked on stock (reference) cooler. That leads me to believe in time, high temp will still kill your chip.

__________________
99.99% of the world listens to everything else. If you are part of the 0.01% who listen to Eurobeat, copy and paste this into your sig.
This is the Secret Signature Organization (SSO) add me to your sig to help me take over the world

Nenkitsune

I don't agree at all. Temperatures DO effect stability WAY before they hit the max temps where the chip throttles. I decided to test this, and shut my cpu fans off. normally, it's 24h stable, rock solid, and never crashes. Cruises at 45c load. shut the fans off, test crashes at around 57-60c load (which is still within the chips safe limit of 55-78).
I would try with it at stock voltage and such, but the chip doesn't produce enough heat in that configuration hahaha.

My s754 newcastle was the same way. i could never get it stable at 2.5ghz because of temperatures. it would get up about 60c (max of 70c) got a better cooler, dropped the temps down to 55c, and now it's stable. I was pumping a ton of voltage into that poor chip...stock is 1.5v I had it up to 1.77 loaded

sure, the temps are a bit high in both cases, but they were well within the limits.

I do think that high temps can cause longevity issues, but not nearly as soon as high voltages will

__________________
BE-2400 3.105ghz 8 hours orthos stable
BE-2400 3.484ghz on AIR PROOF!
7750 BE 3.3ghz Validation

M0RT3GU1R3

When overclocking, performance is the prerogative right? If an integrated circuit runs stable at high temps that's great.. more power to you. I do believe however, that regardless of longevity, integrated circuits run better in the 30 - 60 degrees C range.

Mr. Mojo

I know it is kind of goofy...but even though I post regularly on overclock.net, and OC every piece of hardware I can, I tend to err on the side of caution. Sure, I could more than likely get away with 65c on my CPU, but part of the challenge to me is to get the best performance without unnecessarily straining the components. If I wanted a suicide machine I'd have a ln/di pot. Right now my CPU is running 3.0Ghz...not the fastest e2160 out there, not even the max OC for my particular chip (I've had it up to 3.33 fully stable and 3.4 unstable while finding the max OC). At 3.0, it does everything I need it to do and max temp is always under 55c.

In over 10 years of building computers and 6 years of OCing, I have never had a component fail due to heat issues, or the OC in general. I damaged my 7900GS with a volt mod and an insane OC, but it still works...I just had to lower the OC a bit.

__________________

Auld

I have always thought you were suppose to keep your Vcore and load temps under what intel or amd recommends. for an example my chip is 1.5v max vcore and under 70C load temps. are you suggesting it is ok to let the chip run hotter than 70C? i have not personally pushed any chip far enough to throttle down and i dont feel comfortable doing so. what temperature range do processors actually throttle down?

Nenkitsune

It seems to depend on intels, but the C2D's are supposed to throttle at 85-100c depending on the chip.

I don't know when an AMD chip will throttle. I've never wanted to try.

__________________
BE-2400 3.105ghz 8 hours orthos stable
BE-2400 3.484ghz on AIR PROOF!
7750 BE 3.3ghz Validation

bomfunk

Okay, now it's time for bomfunk to pop in...

It is true that temperatures will have very little effect on the chip's lifespan, but it will still take its toll. Why you probably shouldn't care is because by the moment the heating kills your chip you'd upgraded anyways. I believe that the line goes around 80-90C, having the chip constantly that high (or higher) will begin significantly affecting the chip's lifespan. Well, I'll let you argue whether it does or doesn't; also, remember that as the chips manufacturing node gets smaller (45nm, 32nm,...), they become more sensitive to heat as well as voltage. So while you may have nothing to worry about with your old 130nm or 90nm Pentium 4 / Athlon 64, or even a 65nm CPU, even when constantly licking 80's, this may be very different with a 45nm chip (45nm apparently being a kind of a "magical point" where the chip seems to get more fragile because of different materials & manufacturing technologies used).

However, as for stability, temps do affect it. You see, the switching behaviour or transistors changes (degrades) as temperatures go higher; cooler transistors = better switching operation. So while having load temps of 45 during a cold day and 65 during a hot day makes no difference to you, when the same chip hits 80C-90C, it may well result in an unstable system with the exact same OC.

__________________
Statistix show that 92% of teenagers have moved to real music . If you are one of the sane 8% that still listen to funk, hip-hop or jazz, copy and past this into your sig, for thou art now offically awesome!DISCLAIMER:
By reading the above post, you accept that the poster ("bomfunk") is now the rightful owner of all your present possessions.

Nenkitsune

like last night I improved my best overclock by quite a bit, by dropping the idle temps down to 17c with 1.68v previous best is in my sig, new best is 3.484ghz

also, if temps didn't matter, then why is it that people use watercooling, phase cooling, and even more extreme forms of cooling to get better overclocks?

in my case an 8c drop in temps lead to a pretty nice jump in my overclock.

__________________
BE-2400 3.105ghz 8 hours orthos stable
BE-2400 3.484ghz on AIR PROOF!
7750 BE 3.3ghz Validation

Retrospekt

Those are idle temps. Either way, your overclock didn't improve because of the temperatures. They improved because you saw that your idle temp was lower, and you allowed yourself to raise it more.

I'm sort of making an assumption that you aren't loading at 80c, since you didn't state your load temps.

__________________