EVGA 1080 ACX Fail?

[Thavion Hawk]

So about a year or so ago a guy came into the store I work at and asked if we could build his new system for him? We said yes and so he brought in the parts that he had ordered. Now it's not uncommon around this area for college students that want a nice computer but have a budget to go out and order parts to build a system themselves without fully grasping how hard it could be to build the system(Relative to skill and understanding) We normaly charge about one and a half to two hours labor to do a build for a customer depending on the complexity of the build and/or the OS or software that they want loaded. In this guys case(No pun intended) we had to charge two hours labor for the build because... well here's the pictures of his parts off my Instagram 

 

 Now as you can see that's one heck of a parts list. This is the kind of list I imagine gets made by going on PC Parts Picker and picking close to if not the most expensive parts on a platform. At the time of the build it was the single most expensive system by parts value I had ever worked on.

 

Please forgive my pining for his system specs. Given the point of this post I have no regrets having my system and not his...

 

Okay now that the intro is over, here's the deal. This build was done just as the EVGA 1080FTW ACX Thermal problems came to light and the Mass Recall stuff went down and wouldn't you know it, that's a pair of effected and there for theoretically Faulty 1080's. The customer had just picked the system up the day before the first videos hit YouTube about the situation and I said to myself, "If that guy watches any of this he's going to freak out..." Sure enough about a week later we get a call from the guy asking if he could bring the cards to us so we could take them apart and install the EVGA Thermal Pad Kits for him because he doesn't want to RMA his new cards and risk getting back different faulty cards... Well we said no because we don't want to risk damaging product that we didn't sell(I've taken GPU's apart many times and would do it for my own card in a heart beat but no way am I risking it with a customer part...) when EVGA would cover all the cost of shipping the cards to them to get the pads installed by them.

 

The customer relented and sent the cards to EVGA and all was well until today. Today the customer came in and told me how, after about 6 months, one of the graphics cards had failed. He RMA'd the card and all was well until again just yesterday another GPU failed. Well he thinks it's the GPU given where the magic blue smoke came from. He opted to bring it in so that we could diagnose the problem. If it's just another faulty GPU, the cards have warranties and EVGA is damn good with dealing with RMA's. If it turns out his PSU some how was killing his GPU's, or the Mobo has a fault or what ever the case may be otherwise, then we will be able to diagnose and deal with it for him. 

 

This guy spent close to $3,000 on this computer and factoring in GPU values it's worth more now then it was when he built it. I fell for the guy and want to help him out I just hope that his problem is just the GPU. We don't have any PSUs larger then an EVGA 750GQ in stock so if it's a PSU then he's stuck RMA'ing his Corsair unit or waiting for a larger one to be ordered in(It's got a good warranty too so RMA is only logical) If it's the motherboard, well we have an ASUS Prime Z270-A in stock we can drop in that will work, but it will make his ROG Head unit worthless unless he again opts to RMA the motherboard(Again Warranties are nice to have) 

 

Well that's the situation all. This guy has himself a beast of a gaming rig and it's had one GPU go out on him in the past year with what looks to be a second(And no I didn't get clarification if this one is the one that came back from EVGA after the first one failed.) What do you all think? Is it going to be the GPU again? Is it something else that's contributing to killing the GPU's? Or is this guy just running some crypto mining in the background of his computer cooking the GPUs? Honestly as he said this failure happens while he was browsing the web and not gaming makes me think he may have been mining via that browser based mining script...

 

 

[Mick Naughty]

Well he had money, so he should swap the parts and find out. 

 

What is he even doing to use both cards? All the parts have warranties so that’s a start there. Gpu would kill more then one card repeatedly. If it was the same one in the same slot over and over it would be clear it’s the board but given such little detail of use and temps, it’s anyones guess. 

 

Mining isnt gonna kill a card as it would use the same profile as it would gaming and the cards are smart enough to crash if there’s a heat issue. 

 

Going off the pic poor cooling is a big factor regardless. 

 

 

[Thavion Hawk]

I recommended the customer let me mount the rad on the top, but he wanted to keep the case cover on the vents. There is tape holding the cover on. Temps on the GPU's caused by lack of air flow may well be contributing to his problems. He showed me his rad and it was clogged with dust. He didn't think to blow it out from the inside to clear the dust...