Question Sapphire Vega56 Pulse VRM Temps at stock

[JakubVe]

Hi all,
Could anyone here with the Sapphire pulse share your GPU VRM and SOC VRM temperatures (GPU-Z naming)? In stock.

I just replaced the cool pads and had to use 0.5mm to get proper contact between the GPU and heatsink. And if I leave it stock, the temps are climbing to over 90° in few min in Heaven and it actually crashed at 91°(seems bit soon to be fair).

Underclocked at 1000mV with +30% power limit and 50-60% fans(1600-2000rpm) the temperatures don't go above 80% which I would consider quite good. The higher fan curve makes most of the difference so the 0.5mm pads do have contact just not as efficient at that thickness I would guess. It just worries me that at stock it crashes under constant load.

Thanks

 

[Lutfij]

Which pads did you replace the stock pads with? I've used Fujipoly but they aren't cheap on past mods for GPU's. Might want to also measure up the thickness of the stock pads, if you still have them around. The wrong thickness of pads can and will amount to uneven pressure/mounting for the rest of the card which could lead to more heat buildup. So even if temps on the VRM's are good, the rest of the card might be heating up. Try working with higher fan curves since the GPU is a hot head.

PSU's make and model?

 

[JakubVe]

Hi, I've replaced it with pads from extremecool360, nothing fancy but doubt its worse than the stock pads. The problem is I didnt have the stock pads as I got the card without them. The previous owner wasnt sure what thickness was on originaly. But as I wrote I tested 4 different thicknesses and the only one where I got the proper contact between heatsink and GPU chip was with 0.5mm.

The ex-360 were bit hard so I got another ones which seems bit softer and I now got 1mm on(also 6 W/mK instead of 5 W/mK) and have no problem with contact on GPU, but also the temps didnt really improve, maybe couple of degrees. This kind of leads me to believe that these are fairly standard temperatures. The crash on 91 degrees seemed to be random as it never happened again with either of the pads.

As I mentioned, the higher fan curves do help a lot but it also gets louder and given the VRMs are attached to heatsink, they shouldnt really get too hot, in theory. Shame there is noone with the card as I would really like to know what are the standard temperatures for this card if I should be investigating further or if I actually reached the standard temps. I often sell my old hardware so I wanna make sure that it works as standard version would.

The PSU is BeQuiet Straight Power 580W, its gold rating so dont think this causes any troubles.

 

[JakubVe]

Also to clarify, Do you, or anyone reading this, know when GPU-Z is mentioning GPU VRM and SOC VRM temperatures which parts of the board would that be? Is it those 4 "phases"(not sure if thats the right term) on top of the chip and 3(4) to the right side?