- Aug 25, 2001
- 56,587
- 10,225
- 126
Just upgraded a friend to an X2 3800+ (939) CPU. His heatsink (from a 3000+) is an all-aluminum block, no copper, no heatpipes. Currently overclocked to 2.65Ghz, from 2.0Ghz. According to CPU-Z, the vcore is 1.4v, although the VID default is 1.35v, and I have the BIOS set to auto-detect default volts. Perhaps the board (Biostar 939 NF4) overvolts.
Anyways, the temps are reading 70C for Core1 and 79C for Core2, with CoreTemp 0.95.4.
I used Kingwin thermal paste, a dollop in the middle of the core. I didn't spread it around. I wonder if that could be part of the problem.
Amazingly, it's prime-stable in this condition, so I'm thinking just leave it alone. What do you think?
Oh, CoreTemp reports a TCaseMax of 73C too. So if AMD's also have an ~15C delta between TCore and TCase, then I'm probably just barely alright.
Edit: Ugh, core2 just hit 83C. That seems too hot to me.
The heatsink is almost too hot to touch, so I guess I'm making good thermal contact with the CPU. It seems that the stock heatsink/fan is not quite up to the task. Should there be an 11C difference between the cores?
Anyways, the temps are reading 70C for Core1 and 79C for Core2, with CoreTemp 0.95.4.
I used Kingwin thermal paste, a dollop in the middle of the core. I didn't spread it around. I wonder if that could be part of the problem.
Amazingly, it's prime-stable in this condition, so I'm thinking just leave it alone. What do you think?
Oh, CoreTemp reports a TCaseMax of 73C too. So if AMD's also have an ~15C delta between TCore and TCase, then I'm probably just barely alright.
Edit: Ugh, core2 just hit 83C. That seems too hot to me.
The heatsink is almost too hot to touch, so I guess I'm making good thermal contact with the CPU. It seems that the stock heatsink/fan is not quite up to the task. Should there be an 11C difference between the cores?
