Originally Posted by leopold1985
Thats the effect of the LLC... to compensate for vdroop, it always keeps the vcore a bit high, so the 1.46v is quite fine. It is meant to be this way, so the vcore read will always be a tiny bit above the set value. Trust me, it is fine. This board/chipset has all the OC potentials of a 890FX based board.
Since the phenomenon that you have pointed out is actually not a problem, there is no fix for it.
thanks for the reply, but ive read some posts of others with the same mobo and nobody seems to have an answer to this. you say its normal, but this type of occurrence affects temps for sure. im currently priming, default blend, and multiplier goes up and down when it should stay at a constant x20 (cpu throttling. not good)
this leads me to believe that mosfets are overheating and voltages drop to compensate for the heat. dont know why this is happening because mofsets are heatsinked, plus i have a scythe gt ap-31 5400rpm/150cfm fan on the hyper 212+, which blows a great amount of transient air over the mofset heatsink.
ive oc'ed my 1090t on my old asus m4a785m, which didnt have a heatsink on the mofsets, and is a 4+1 phase board. manually set vcore NEVER climbed above set values, resulting in a stable 3.8ghz @ 1.41vcore OC, with max temps of 49c 12hours prime95 default blend.
im 2 hours into default blend on the m4a88td-v evo set on 200fsb x 20= 4ghz with vcore manually set @ 1.44 and LLC enabled. max core temp 64c so far and max vcore 1.54v. although none of the cores have failed so far, im not confident that it will finish all the iterations (which will take approx 11 hours) with these values and cpu throttling.
im starting to think that my 1090t has a better chance of reaching 4ghz stable on my old mobo with my current cpu fan and heatsinkEdited by piranhamoose - 2/16/11 at 7:40pm