Originally Posted by AMC
I ran Prime and IBT for a few hours. But remember that every chip is different. I have it water cooled. Still runs really warm when stressed so I will be testing out a new setup soon to see if it helps. I might just set it to 4.6ghz soon. 5.0Ghz is nice but no point really for my applications. I want it to run 1.4volts or less. And my chip can with 4.6ghz.
U had 5GHz @ 1.48v yeah? If u don't mind me asking, what were ur ambient temps and what were ur average temps for the cpu under full load.
Also, why do u want to run it at 1.4v? If u manage to find a way to control the temps better at 1.48v would u still prefer to run it at 1.4v? Are u worried about "electromigration" within the die at 1.48v?
Originally Posted by Skiivari
Mine is 12hrs folding stable + 20 mins of 1792 length p95 FFT's.
Haven't done a "proper" long p95 blend run yet but I am quite confident that it's stable.
The accurate speed is 5010mhz and the voltage is 1.35 :O
And no, the IHS wasn't golden or anything
R u saying u have 5GHz running on 1.35v? Wow!!!
Mate I don't mean to preach but confidence should come from empirical test results, not gut feeling. Folding uses ram a lot so even though it shows 100% cpu utilisation it is not a good overall stability test. 20mins Prime is also not enough.
Perhaps a day of each of these using OCCT...
3 different test types
CPU = no RAM tested, alot of heat generated on the CPU, errors are usually CPU-centered
RAM = Maximum error detection, tests the whole chain CPU-Chipset-Memory
Mix = CPU and RAM alternated
If they all pass then u have an awesome chip!