are theese temps ok???

im running a barton 2500+ @ 2.3ghz 11.5x200 with a vcore of 1.80 using a vantec aeroflow have 2 intake 2 exhaus plus the psu fan my mobo is an a7n8x deluxe rev2 and asus probe tells me the cpu temp is 52c-54c is that an ok temp ? just wanna make sure i dont burn something down thx also i wanted to know if anyone has this speed achieved with a lower voltage because i want to try to get the heat down a bit but i dont want instability

If it is stable...then it is fine...the bartons will go up to 85 or 90c before they fry...

i know the max temp before damage happens is pretty high but i also know corruption starts at around 65c so i just wanted to verify this because if i remember correctly the asus probe measures from the socket probe so the on die temp is probably higher right?

55C is getting hot there,
put in a better HSF and you'll drop under 50C.
as a CPU gets hotter it will perform worse then it would if it wasnt OC'ed, I have noticed when a cpu passes 50C it starts to perform a little better but not muc, once it passes 60C it seems to perform worse then that last OC.
unless you can keep it cool.
I sugest just getting a better HSF and you should be all good.
-L_P

Check www.amd.com in their tech specs area. That will give you the range that people with lots of money and an intrest in it running well think it should have. I would suspect that the actual temp it runs reliably at will be a little higher.

AMD tells you the temps where damage will occur(85 or 90c depending on model: here )...you shouldn't be anywhere near that. You shouldn't aim for that, especially when overclocking. You will start getting errors and significantly reduce the life of the processor as temps increase. AMD is in the business of selling processors, I'm sure they'd be happy to have everyone running at 80c and buying a new processor every 10-12 months.

52-54 is alittle warm for a idle temp, I would shoot for below 50 idle and around 50c for load - but lower is always better! More than that and I think you will probably run into stability problems. Either reduce your OC and voltage, or improve your cooling (water )

hey, what are you ram timings at? I have HyperX 3500 its WinBond -5... can't get it to run at any good speeds, like cas 2 @ 200 FSB, I have the same mobo as you also.. and does yours allow you to change the chipset voltage? and what voltage are you running your ram at... Also, I have a Aero 7+ and after running at 1.8 for a few days, I bumped it down to 1.75 and it runs fine, could probably run it at stock, or slightly above bump im trying to find out what the hell is causing my ram to not run at its 2 -3 -3 -8 timings..

kct2 said:

AMD tells you the temps where damage will occur(85 or 90c depending on model: here )...you shouldn't be anywhere near that. You shouldn't aim for that, especially when overclocking. You will start getting errors and significantly reduce the life of the processor as temps increase. AMD is in the business of selling processors, I'm sure they'd be happy to have everyone running at 80c and buying a new processor every 10-12 months.

Click to expand...

Good point.
Isn't the point to get reliabilty not low temps? So theoretically if you had a chip that ran reliably at 120C it would be ok and you could cook with it.(I'm not saying this is at all possible and if I'm wrong please correct me)

kct2 said:

52-54 is alittle warm for a idle temp, I would shoot for below 50 idle and around 50c for load - but lower is always better! More than that and I think you will probably run into stability problems. Either reduce your OC and voltage, or improve your cooling (water )

Click to expand...

Just wanted to point out that lower is not always better, if you were cooling to -273C you wouldn't be running, but thats not really physically possible. But the point were you start running into those kind of problems is a probably possible just not feasable. Also the colder you get the more of a hassle condensation probably is, and expansion and contraction forces may be a problem at some point.

A friend of mine has a Barton 3200+ and he idles in the high 40's. He will hit 50 - 53 under full load. So i would maybe reseat the HS and check to see if it needs lapping or not.

Isn't the point to get reliabilty not low temps?

Click to expand...

With current technology those things go hand in hand. When overclocking you are always looking for a higher reliable speed, you achieve that mainly by keeping everything cool.

So theoretically if you had a chip that ran reliably at 120C it would be ok and you could cook with it

Click to expand...

Just because a chip CAN run at 120c doesn't mean it will run OPTIMALLY at 120c. If you could run it at 80c, 60c, or even 20c, it would run better. It is just the physics of how chips work. The eletrical resistance of a metal goes up proportionally with temperature increase so the lower the temp the more easily current will pass through the interconnects. Also, electromigration would be a large problem at temperatures that high for any high powered chip.

Just wanted to point out that lower is not always better, if you were cooling to -273C you wouldn't be running, but thats not really physically possible.

Click to expand...

I'm very impressed that you know where absolute zero is. But its relevence to this persons quesiton is non-existant.

Also the colder you get the more of a hassle condensation probably is, and expansion and contraction forces may be a problem at some point.

Click to expand...

Just because condensation may be a problem doesn't mean temperatures below ambient aren't good for the processor or overclocking. Many people run phase change and peltier (thermoelectric) coolers which can cool WELL below ambient. Are there potential condensation problems? Of course, but with some simple insulation and some good planning these people easily get past them.