Recently, we covered a story on the new EVGA GeForce GTX 970 ACX having a possible design flaw or manufacturing defect where the GPU die only makes contact with 2 of the 3 heatpipes included in their cooler design. You can check out the full story here. Since then, we’ve contacted EVGA and asked for a response to give any substance to the claims and if the cooler design was intentional.
EVGA have given their response below:
“The way the EVGA GTX 970 ACX heat sink was designed is based on the GTX 970 wattage plus an additional 40% cooling headroom on top of it. There are 3 heat pipes on the heatsink – 2 x 8mm major heat pipes to distribute the majority of the heat from the GPU to the heatsink, and a 3rd 6mm heatpipe is used as a supplement to the design to reduce another 2-3 degrees Celsius. Also we would like to mention that the cooler passed NVIDIA Greenlight specifications.
Due to the GPU small die size, we intended for the GPU to contact two major heat pipes with direct touch to make the best heat dissipation without any other material in between.
We all know the Maxwell GPU is an extremely power efficient GPU, our SC cooler was overbuilt for it and allowed us to provide cards with boost clocks at over 1300MHz. EVGA also has an “FTW” version for those users who want even higher clocks.
Regarding fan noise, we understand that some have expressed concerns over the fan noise on the EVGA GTX 970 cards, this is not a fan noise issue but it is more of an aggressive fan curve set by the default BIOS. The fan curve can be easily adjusted in EVGA PrecisionX or any other overclocking software. Regardless, we have heard the concerns and will provide a BIOS update to reduce the fan noise during idle.“
Our take on the above quote is a bit of a strange one, as EVGA claim that the heatsink is based on the GTX 970 wattage + 40% extra cooling performance to give more headroom giving us a total cooling power of 200W TDP, but as the reference GTX 970 uses essentially the same cooler as the Titan Black which has a 250W TDP, we can only assume that EVGA’s GTX 970 ACX could it fact be a worse performer than a reference model.
It is also worth noting upon the part of the quote which talks about the fan curve being so “aggressive” and the cooler being oversized for the GPU and TDP – If this is the case, why does the fan curve have to be so aggressive that the card has such a noisy idle of around 30dbA?
Even after the official response above, someone has brought it to our attention that the 3rd heatpipe isn’t even sealed off therefore having no functioning operation whatsoever as shown below.
So there you have it, but is this enough to satisfy you, or do you think EVGA are using this as an excuse for them clearly trying to cut costs by using the same cooler from other model cards? Only you can decide that, but as always, we invite you to let us know your thoughts in the comments below.
Well I’m pretty miffed I just took delivery of two ! I don’t want to remove the cooler to check, just wondering should I be returning them for a refund now.
Nahhh, don’t worry about it. These are still kick ass cards and the minimal difference that extra heat pipe would make, you won’t even notice. Especially since you have two of them.
goodthing i use asus 😛
They did the same thing with the Direct CUII on the R9 290/x as well :/
Always pays to do some research beforehand no matter the company.
maybe but i only use nvidia cards so no idea what goes on with amd cards lol
Well you’d love to know that the DCU II on the GTX 780/Ti also does the same thing, again it pays to do your research first.
are you trying to be one of them know it all people…
Informative, not “know it all”
Same thing, you call it Tomatoes, I call it Tamatoes. no difference you still come off as a smart ass.
I don’t particularly care how i come across the internet to random people, i posted some useful information to someone that obviously buys because of the name alone and as evidenced by EVGA (Nvidia No #1 AIB) that’s not the wisest choice.
The way the EVGA GTX 970 ACX heat sink was designed is based on the GTX 970. That is all BS coming from higher up.
So EVGA is saying that they had this design since the GTX760. Even then the ACX on the GTX 760 was not that good either.
It is not a new design, they just recycle the old and painted the heatpipes. Look at Gigabyte, Asus, MSI, Zotac they all have new design for the 970/980.
This is the exact same heat sink used on the GTX760 SC, and that only made contact with 2/3 heat pipes too, the third one is actually smaller/less important than the main two that do make contact. It was a non issue for the 760, it’s a non issue for the 970, just blown out of proportion.
It is an issue. The position of the GPU is off a bit. Which makes part the die touch the metal plate on the side. The GTX 760 did not have that problem, as the ACX was designed for it, not 970.
The position of the GPU is the same as it was on the GTX760, hence why the TIM markings are identical.
Look at the picture. The die doesn’t touch the plate. That’s excess thermal compound. Look at the outline. The die is no bigger than the 2 pipes it touches.
Who gives a crap if they used a little extra thermal paste. it does not effect the GPU in any way. as for the smaller die not covering the third tube. Heat does not contain its self to just that small area. heat spreads out over the entire plate. so it will still dissipate the heat. Some people are blowing this way out of proportion.
good thing i use AMD FTW
If you are so happy for AMD then why are you commenting on nvidia related article?
It having no functioning whatsoever is quite harsh to say. The copper in the heatpipe, even without the gas inside, is still very much conductive, more so than the aluminium that connects the pipes together at the base. The heat would still travel through the copper at a higher rate than the aluminium base would transfer it, with or without the gas inside.
So it still works, and as long as the GPU core is not placed on the pipe itself, and the heat has to travel through the aluminium, it works just as well as it is right now.
At least that’s my take on it.
I agree with you. I’d also like to say that I feel they did try to cut costs. It wont matter much if you aren’t oc’ing but if you are, an aftermarket cooler of the same price will probably do better.
If they took the third pipe away, would it make the cooler more efficient? No. But it would increase the cost, which would be passed on to the customer. So they used an existing item, with a third pipe that isn’t used, which lowers the cost of manufacturing because they already have a warehouse full of the things. Why does everybody have such an issue with that?
The cost would not be increase and pass to consumers. Why is Asus, Gigabyte, MSI and Zotac all came out with new design heatsinks and yet cost the closely the same.
Im more concerned with the excessive use of thermal paste by manufacturers.
Are they using paste, or compound? ..
So? It doesn’t matter if there is some spilling out at the edges so long as its thinly spread over the actual contact area, which looks to be the case.
Look it up. You might learn something.
Oh and btw. This was written 5 months ago. Way to necro!
Doesn’t really matter when it was written, people are still buying this card, so they will still be reading the comments on this article, and it’s still important that they get the right information.
And I don’t need to look it up, I understand it very well. Having a bunch of extra insulating material smeared around the edges does pretty much nothing, since *by design* the heat isn’t really expected to transfer through that area anyway, it’s just dead space. 99% of the heat is going through the contact area with the metal, where it looks to be appropriately thinly spread.
Highly doubt people will by this card now. Maybe used, but highly doubt that aswell. It has been replaced with the ACX 2.0 version.
One should care about the thermal pase/compound or whatever you want to call it. Optimally, you don’t want any at all. But we don’t live in a perfect world where the heatspreader and cooler have 100% contact.
That may be true, but the ACX 2.0 design was not changed in this area so the discussion is still relevant: http://www.bit-tech.net/hardware/graphics/2014/09/19/nvidia-geforce-gtx-970-review/3
And I agree that people should care about thermal paste *in general*, and I never said otherwise. I’m trying to explain to you why the way they are doing it is okay, because for the most part you *only* need to care about the way it is spread *over the contact area*. I thought I was pretty clear about this. The thermal paste spread *around* but not *on* that specific area makes about as much difference as if the way you might spread it on the outside of your case.
However it does make a difference. No matter how blinded you are by your own opinion.
It’s not just an opinion if I’ve actually explained to you why it doesn’t matter, which I have. What you have offered in response, on the other hand, *is* just an opinion, since you’ve failed to provide any argument.
Do you understand anything about heat transfer, or did you just read ‘less paste is better’ somewhere without understanding the subtleties of it or the actual reason behind it?
I’m going to put the next part in capitals, but for the record I’m not angry/shouting, I’m just trying to put some emphasis on it since it doesn’t seem to be getting through to you: THE AMOUNT OF PASTE ON THE CONTACT AREA IS THE ONLY PART WHICH IS IMPORTANT, BECAUSE THAT’S WHERE THE HEAT TRANSFER HAPPENS.
Haha, sure mate, I’m sure you’ve set a rig up and done dozens of tests, controlling for all other factors, including the amount actually left over the contact area, just to address this exact issue, I’m totally going to take your word on that one. You wouldn’t just make shit up to win points in an argument lol.
Look, even if you’re telling the truth about having done this, since I actually understand how this works (objectively, learn what opinion means), I’d put far higher odds on you just having smeared it badly on the tests where you left some around the outside and left air bubbles, or too much paste, in the contact area where it actually makes a difference. Smearing by hand is a really unreliable method, after all. Its no surprise you didn’t get great results. The reason you got poor temps, though, was categorically nothing at all to do with the excess paste round the outside. Use your head, try to actually think about what makes sense here. Doubling down on your misunderstanding is just making you look silly.
I don’t think you understand what objective and subjective mean. I have described the physical processes involved to you which result in the surrounding paste having no effect. This is an objective claim. It might be an *incorrect* claim, but that doesn’t make it subjective, and if it *is* incorrect you have still repeatedly failed to explain why you think this is the case.
And no, I’ve not done tests. I never claimed to. You were the only one that lied about that. I actually have arguments which have value and stand on their own, which, again, you have repeatedly failed to counter. Obviously no tests means no *proof*, but I never claimed to have proof, just a much stronger argument than you. Laugh it up, clown 🙂
Did you not read my last comment, or did you just not understand it? *You* prove it. Or at least offer *something* in the way of an actual argument, as I’ve done.
Or just post another whiny, petty, non-comment so you can get the last word in.
Well. You have not provided anything. You bring accusations. You act like a know-it-all. Take it as you will. I don’t fucking care.
The 3rd heatpipe is sealed, it doesn’t extend all the way through the holes in the fins because there’s likely some components in that area which is why there’s a cutout in the fins there in the first place.
Is the EVGA GTX 980 ACX 2.0 plagued with the same problem?
No it’s a different design. 5 heat pipes with full contact
GTX 980 ACX 2.0
GTX 980 ACX 2.0
http://forums.evga.com/download.axd?file=0;2222804&where=&f=980.jpg
EVGA GTX 970
http://cdn.overclock.net/d/d2/d2678403_cooler2.jpeg
which model number is this as many different evga gtx970 2954 2955 3756 3754 1298 etc ???
I never had problem one with my 970 SC ACX 2.0 cooler! I can hit a boost clock of 1500mhz and it still stays under 70c! I think it’s a very nice cooler! They should have designed it to cover all V-Ram chips and cool the VRM a little better. But other than that it’s fine! You don’t see MSI cards hitting the boost speeds that EVGA’s cards do! At stock mine hits 1379mhz.
Ok after about 8 months of use, And after playing newer games that use more GPU power and generate more heat, I can now confirm that the ACX 2.0 cooler is not the best in the world. One of the main problems is the crappy fan profile, This cooler should have never been paired with a passive fan mode. And when the fans finally do spin the profile is not nearly aggressive enough. Also the V-Ram chips add to the heat allot since they are left bare with no heatsinks of their own. If this cooler had V-Ram heatsinks on the top and bottom it would not be so bad. But it would still not be perfect. I bought this card because I did not want to worry about upgrading for a few years. But heat never helps a card to last long, Ask my old 5850 about that. EVGA could have done better than this. I know it has a lower rated TDP than the 760, But anyone with a brain knows it’s going to use more wattage than a older 760 under heavy loads. Using a old 760 cooler was a dumb idea.
i have the EVGA GeForce GTX 970 FTW + ACX 2.0 and it ran at 60C at idle with no fan turn the fan to 20% and it drops down to 35C…just liquid cooled it with an nzxt g10 and x40 with push pull config and over clocked to 1519mhz and get max temp of 50C runs at 28C at idle….THE ONE PROBLEM I HAVE IS “COIL WHINE” its loud under a load considering my graphics card and cpu are liquid cooled. turn the volume up…lol….evga says its not a problem just annoying.