waves from the land of 980mv or less in every game and benchmark
Yuk you got the movie reference in the jokeHumans have 21 grams of water? I think you confused it with the movie 21 grams where they say the human soul weighs that much.
XTX or XT? I'm talking about xtx. I realize they may not be the same in post. But even ltmatt was crashing in GoW in under 1080mv, iirc.
Ran warzone yesterday at 3050 core, 1075mv, 2750mem,15+pl. No issues so far. With hwinfo in the background voltage went no higher 1041. 1075 will not be stable for other games but thats a benefit you get from amd. You can overclock per game which is nice
MBA XTX faulty cooler, vertical, open air.
Exactly my findings. 3020 MHz core, 1125mV, 2776 MHz memory is what I converged to after hours of testing and gaming. I have a MERC 310 7900XTX. 32.7k in TS. I can get better scores that "appear" stable but will eventually cause a driver timeout/crash... Mainly during transient load situations. I am happy with the performance.
This is true but is the % difference worth it? In my eyes no.
Agreed, "One Stable mv to rule them all" ! Sooner or later it's gonna crash, and I bet it's not when you wan't it to. I use Scum all epic settings (nV game UE4) to find stable mv @DX12, it will find any unstable card eventually (BTW: this game won't run DX11 at all on XTX, first driver bug in a game I have seen so far, as 6900 XT did it fine, need to force it to DX12 via steam library). Also I have found on my Red Devil Ltd. XTX, that set 2730mhz get 2710'ish at 1090 gives me desent power usage even at +15%, anything above like 2830mhz and wattage runs away completely. @2730mhz set 1090mv I'm around 370-390w max, that gives me a silent card and still close enough to the "wild" benches here. Did I win silicon lottery this time? As usual, no....
It's not a massive difference but my temps are nice because of it . 1100 mv avg Hotspot 68. 1085 and 1075mv avg Hotspot 62c
OMG! Glad to read rest of rig ok.
Has anyone received one of the EK-Quantum Vector² RX 7900 XTX waterblocks yet?
There is more total mass in TIM on the card than there is water in a vapor chamber, probably by a fair bit (those pads and putty add up). So unless you were careful to reuse only what came with the card, weighed it beforehand, or are going to repaste it in exactly the same way, it will be hard to put the difference down to the vapor chamber contents.
Best working fluid for electronics cooling in most applications. Very high heat of vaporization and a boiling point that can be easily tuned to the intended operating temperature range of an ASIC.
