CPU temp difference

[fatherdougal]

Hi all,

just rebuilt my system with a new mobo and gfx card and I am having a big discrepancy between different temperature monitoring software. I am using a p35 Gigabyte ds3l and a e4300 with a stock cooler.

On my old board, a ASRock Dual-vsta, I consistently had temps on both cores in the high 30's, monitored using speedfan.

With a similar overclock on the new board I am seeing temperatures under speedfan of 70c under heavy load and 45c at idle.

The mobo itself has temperatures over 10c lower all the time as does cpu cool and real temp but PC wizard has the higher temperatures, the same as speedfan.

I have cleaned and reseated the cooler and cpu a couple of times with no real difference.

I'm tempted to nip out and get a new cooler this lunchtime to see if it makes any difference, though I would rather not! Any ideas why there is such a big difference?

 

[PaulyD]

What thermal paste did you use?

Some thermal pastes, like Artic Silver 5, have a curing time before they are most effective. In Artic Silvers case this is around 200 hours.

Also - most monitoring software incorrectly detects the TJ Max of the processor so this is where you will find the differences between different software. The one to use would be Real Temp.

 

[fatherdougal]

Thanks Pauly. It was Arctic Silver that I used. Re-seated it yesterday, after a week of the high-temps, but only monitored using Speedfan. I just find it weird that 3 of the 5 programs that I have used are giving me roughly the same figure, within a degree, and 2 a higher figure, that is again roughly the same.

System is stable though under stress testing and seems to be alright. Doesn't really seem to matter what overclock, within reason, that I am running and the temp is the same.

I didn't bother buying a new cooler yet, my case is really well ventilated and I'm not pushing for any kind of crazy clock on the thing.

However if the higher temperatures are accurate then I may have to. I will download real temp tonight.