Is a ten degree variation between cores normal?

My turion 64x2 always shows a roughly ten degree difference between the two CPUs. core 0 always being hotter is this normal/acceptable? on the whole dont think temps are an issue highest ive seen is around 80 degrees.

 

this is normal, and for a reason. The dual core processors have two thermal probes in them, one per core. The thermal probe is usually located on the right side of the core, so when you put two processors right next to each other, you get: (core 0, than thermal probe, then small gap between cores, than core 1, than thermal probe 2)
When you think about it, the temperature for the first core will always be hotter, because the probe is between both cores. The probe is basically picking up the temperature of the core, and then some of the heat from the other core, vis heat radiation/transfer. This is normal and happens with all multiple quad core processors.

If your temperatures are high, you may want to check out undervolting, cause that could easily cut 10-15C off your highest core temperatures. Undervolting does not reduce performance in anyway, so try it out and see if you like it.
Use the search toolbar and go to the Undervolting guide, which Flipfire made.

K-TRON

 

I know nothing about AMD chips but I've never, ever seen more than 1 or 2 degree differential on Intel chips and for never more than a few seconds.
In fact I suspect your software is the problem since the two cores are so close together on the same silicon wafer and encased in a thick epoxy shell, that a differential like that couldn't be maintained. Quoting from the Monk TV show theme: Randy Newman sings "I could be wrong.....But I don't think so"

 

I understand your explanation but a TEN degree difference? That doesnt seem normal.

 

scott does bring up a good point. I would need much more info to even theoretically guess. Does this temperature differential remain the same from idle to high demand usage?

K-T's explanation would explain many situations but not all. That is why more input is needed. One K-T's would explain very well is going from heavy usage to idle sensor 1 would be slower to cool.

In XP I can watch core usage can you do the same? Are the cores working about the same? If XP do you have the AMD dual core patch installed? Or the XP equivalent?

 

Lets call AMD and Intel and ask. A 10C (18F) differential over millimeter distances on a common silicon base which is relatively massive compared to the photolithographed (aw hell, painted-on) transistors kind of flies in the face of my 25 years in the lab. And, even more interesting to me, is what kind of operating system would divvy up the workload so unequally as cause such a massive delta T. Now my curiosity won't let go back to bed on this Saturday morning. Shame on whomever. Where are the smilies that match my avatar, anyway.

 

My processor behaves likewise. Core0 and Core1 constantly register 4-6'C difference in temperature. Have been using this machine for a couple of weeks. Machine seems normal.

Processor: Intel.T5450

 

I have a 6-8 degree difference after undervolting, it's normal and running fine. Don't worry about it.

 

It is strange that these differences in temperature occur at such large degrees.
In my dell e1505, when the cpu is running stresses running world community grid for hours, core 0 averages 14F higher than core 1 on my T5300 1.73Ghz core 2 duo.
On the AMD Opteron in my other laptop, using the same software, the temperature difference between core 0 and 1 is usually around 23-28F.

K-TRON

 

It doesn't seem strange. It is probably a combination of the location of the thermistor and the contact of the heatsink to the CPU (most companies do a pretty poor job of this).

 

thanks for replies. i have just downloaded and ran ORTHOS for 30 minutes. max temps core (0)-69c. core(1)-65c. so i think overall temps are not a problem. thanks all