Core Temps

ubern00b · 7 Nov 2006 at 15:56

Hi All

If you could be so kind, those of you with an X2 or Opty, could you fire up coretemp (http://www.persianpcshop.com/CoreTemp.zip) and let me know the temps of Core0 and Core1 either load or idle, or both.

I just want to compare the temps with mine. I am seeing core0 being 26C and core1 being 33C. Under load i can see as much as a 10C difference. Just want to compare with others results. I have tried 3 different heatsinks now and they all seem to be about the same difference in temps so I dont think its that much to do with an uneven base of an hsf. Could be the IHS isnt making good contact with one of the cores? I dont know, pulling at straws here. It is worrying me tho, hence the post.

mikeymike · 7 Nov 2006 at 16:03

could well be the IHS, although most DC setups, the cores will always have slightly differing temps, just search this forum, theres many threads about this, if its really doing your head in, or really holding back an overclock, check out the sticky on removing the IHS, i recommend a double bladed razor blade, selotape over one side, insert the blade 3-4mm and apply firm pressure as you push it all the way round, nerve-wracking the 1st time, my IHSless opteron 144 @3.1ghz runs at 45 degrees at full load

Mr Paul · 7 Nov 2006 at 18:47

Many are like this including mine. If its stable don't worry about it

I've been told that AMD only calibrated one core and thats the lower one. Also even the "core temp" programme isn't that accurate, AMD tell system builders NOT to use the core temp reading for thermal management.

ubern00b · 7 Nov 2006 at 22:14

Mr Paul said:
Many are like this including mine. If its stable don't worry about it

I've been told that AMD only calibrated one core and thats the lower one. Also even the "core temp" programme isn't that accurate, AMD tell system builders NOT to use the core temp reading for thermal management.

Do you have any linkage for the above info? Sounds very interesting. It just sounds strange that AMD would only calibrate one core (and its the lower one...) just sounds a bit strange, but if you have any links to this sort of information I would appreciate it. Im on a mission to find out as much as I can now.

I tried to search these forums for info on this too but didnt get very far, if someone could top my search skills I would appreciate those links too

ubern00b · 7 Nov 2006 at 22:27

Found some interesting info written by the author of CoreTemp:

"In Rev F chips from AMD, the reported temperature also seems to be quite accurate, but from different reports and white papers I've seen, the CPU leaves the factory without having the DTS properly calibrated. AMD claims it could have an accuracy range of ±14ºC. The only thing I've noticed is some older AMD CPUs either have a very large delta between two cores or sometimes give some really low temperature readings. I guess this is understandable as this feature was unofficial in those CPUs."

Well, I guess I have my answer on that note, but the question still remains as to which core is the calibrated one? The cooler one (because I would really like that to be the case) or the hotter one? /me continues his quest for knowledge on the subject.