With both cores under 100% load, core 1 on my E7300 gets up to 7 degrees warmer than core 2. Under no load they are the same. I have read that the sensors can be misleading as in they are not both seated at the hottest part of the core and micrometres can make a big difference, but surely the readings would then be consistantly different. I'm using the stock cooler at the mo and don't really want to do a reseating of the H/S until i have a new decent one. Has anyone else noticed this with their multicore CPUs?
yep mine usually 2 to 4'C out depending on the load, could be that one core is doing more work than the other? I wouldn't worry much unless its causing your computer to restart.
this happens when the 2 cores are under full load with a small amount of kernel activity. the thing is that the temps are the same when idle. the reason i want to sort this out is because i'm contemplating overclocking my machine. i may have to do this to sort out my memory problems (different theread) cheers for the response tho
ah overclocking, not done that for years, hehe. what application are you using to to push the cpus to full load?
i've used prime95 for a sustained cpu workout and the usage of both cores is 100% flatline. I've also used DVDshrink (backup purposes onlyof course) and with 100% usage again the cores read about 7 degrees different. I stress that the reason for my post is because the cores are only different when under load. If they were different all the time i could explain it - bad seating of the heatsink or sensor inconsistencies. If the processor is faulty i can still get it replaced but if i OC it and it dies then it may void my warranty (not sure if you can tell if a CPUs been OCed or not) Apparently these new 45nm chips are excellent for OCing. My 2.6 rated chip should clock to 3.5-4GHz without really breaking into a sweat. Thanks again for your interest in this thread
