• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Temps

Suscept

n00b
Joined
May 20, 2005
Messages
9
Hi,

What temp should an Athlon64 3000 run at idle?

Mines 45c (60c load) but i do have a Zalman on the low setting.

I like it to be quiet as its in the living room, would this do any long term damage?
 
never, never, go above 50c

for each 10c you go above 50c, you reduce the life of the CPU by 50%
 
i did not know that...is it really true?...
i hate being anywehre near 50c

edit: yay...im not a noobie nemore...im a [H]lite :)
 
He's exagerating...
I say max 55c should be fine.
I think 60 is a scary number for temperatures.
My cpu idles at around 45c.
In the BIOS I have it set to put full load on the fan if the temperature goes above 50c
So basically the CPU hits 51c the fan goes all out and usually drops the temperature to 49c then it works back up to 51c and turns to max again...
Stock AMD fan is VERY quiet.
Maybe your motherboard is reporting the wrong temperature...
Seems high for such a nice cooler.
 
Borgschulze said:
He's exagerating...
I say max 55c should be fine.

sorry, but your wrong. im not exaderating... there was a study done by IBM.

the cpu rating in hours is how long it will last at 30c. if this is 30,000 (for example), running it at 45c will reduce the life expectency to 15,000 hours. now this is not an apsolute, because of how manufactures mesure life expectency, but it does have a significant impact on how long it will last.
 
theshadow27 said:
never, never, go above 50c

pure paranoia.
60C full load is toerable for A64, though lower would be good.
At 60C, you have no worries of getting a new key chain, and replacing a CPU anytime soon.
Though at 1.4-1.5V on a 90 or 130nm process (especially one usign SOI) I'm not as comfortable with them in the 70's as I would be with Prescott.

for each 10c you go above 50c, you reduce the life of the CPU by 50%

that's a reasonable swag, though it will vary from process to process, and is also signifigantly influenced by Eox (Electric field through the di-oxide insulator).
What you should also mention is that even at 80-85C (the actual temps for the transistors, temp readings can easily be 10-20C lower than a given hotspot like an ALU) a 130 or 90nm chip will last well beyond obsolecence with no trouble.
With an Eox around 1V/nm and a temp of about 100C (actually probably more than that) you can expect a silicon chip to last a decade or more.
Eox is more like 1.2-1.3V/nm in more agressive process (like those used for cutting edge CPUs) - higher Eox means more wear and tear - adjusting temps accordingly means 80C or so for the same decade of life. And this, with the inherent inaccuracy of temp monitoring is often where we get the '60C' rule.
Of course you probably only run your chip at full load (and max temps) a few hours a day, so there really is no concern that 60C will kill a chip in it's usefull lifespan, at least not on a current Copper process and reasonable voltages.
(Wait's for the inevitable ancedotal evidance about how temperature killed some one's chip...)

If you'd like this to be a smoothwall router box in 2020, lower the temps. If you're going to toss it in 3 or 4 years (or less) like most people around here, don't sweat the details, as long as it's stable, it's good.
 
FreiDOg said:
If you're going to toss it in 3 or 4 years (or less) like most people around here, don't sweat the details, as long as it's stable, it's good.
err, i like to at least add to my pharm when im done with a computer/mobo... i dont know about the rest of you but i dont have $ comin out of my ass and it isnt THAT much harder to 40 rather then 70 unless you forget thermal compound or something. just get a big cpu cooler ... not hard.

i have a P4 2.6 overclocked to 3.0 with a volcano 7 (remember thos... back in the day) and its been running for 24/7 3 years w/o breaking 40c. you dont gain anything by running hotter, and there is no real way to mesure what you loose. i guess if you simpathize with smokers "im gonna die anyway" thats all good.
 
My winchester 939 3000+ at stock speeds with stock cooling barely ever went over 35.. and only went to MAYBE 37 under EXTREME LOAD with case fans all turned all the way down.. I have it overclocked 400 mhz on stock cooling and it never goes over 45..
 
There are soooooo many variables when it comes to cpu temps. Don't compare your specific situation to everyone else's. Some guys have a dozen case fans running at 12v's. Some have one case fan at 5v. Different motherboards will read different temps, or can be completely inaccurate. Some people have ambient room temps of 27c, some have 20c. Everyone is going to report different temps.

However...37c full load with the stock cooler and all fans turned down sounds a bit too good. Do you live in an igloo in Alaska JonnyB?

I've been building systems for 10 years, and have only seen ONE processor die. And it was because the heatsink fell off during a rough transportation (very rough). The chip fried within minutes. RIP Athlon 1500 :(
 
Not alaska, but even with all fans turned down, there are still 10 fans... (including hsf) :rolleyes:
 
Back
Top