Hi Needitcooler, thank you for your response.
To provide more info, I should mention that I am running 2 instances of Prime95 as my stability test. One torture test is the smaller FFT lengths which are supposed to focus on the processor's FPU and the other test is with the higher FFT lengths where if an error occurs it is supposed to be more indicative of a chipset or memory error. (My understanding of prime comes from reading a post where forum member NookieN described these differences and he seems to know what he is talking about.) It is in the prime with high FFT that my errors are occuring.
What makes you so sure your CPU and Memory aren't the culprits
1. Changing vcore from default up to 1.6V does not help.
2. The test with the smaller FFT runs as long as I want it to which, if my above info is correct, seems to indicate that the cpu is still running strong.
3. If the error in higher FFTs indicates a memory or chipset problem then I suspect the chipset because I have tested the memory with memtest fairly thoroughly and I have a good idea of its limits.
The 2.6c's don't have as much headroom as the 2.4c's. You might be hitting it's maximum.
I thought about that but I think the above shows that the cpu can go higher. What I am afraid of is that I have it the limit of the chipset. I have heard how the springdales are more of a mixed bag then the canterwoods as far as the limit of the fsb. I am just making sure I try everything before giving up.
The reason I say "besides cooling" is that, just to experiment, I opened the case and put a portable fan right beside it and lowered the mboard temp from 38C to 28C but still getting the errors in prime.
are you using a 5:4 memory divider, or are you running your memory 1:1?
I am running 5:4 and making sure I am well within the capabilities of my memory.
I should also mention that I am not getting any CTD's or BSOD's but if Prime95 is to be believed then I probably will if I go much higher. Oh well, I'll try some other programs and hope for the best.
Cheers