NVIDIA GeForce RTX 3080 Pictured?

[TheLostSwede]

https://twitter.com/i/web/status/1269356041885626373

 

[Scrizz]

Hopefully this lets people understand.

Blue: Intake
Red: Exhaust

 

[Xzibit]

Wouldn't that cause havoc for air flow inside the case

 

[$ReaPeR$]

Weird card design.. can't wait to see the tests, especially about the cooling design.

 

[theGryphon]

Wouldn't that cause havoc for air flow inside the case

That's what I said above too... it would.

 

[Vya Domus]

Hopefully this lets people understand.

Blue: Intake
Red: Exhaust

Here's another flaw, those kinds of fans will blow the air directly perpendicular onto the surface below, so that means little to no air reaches the middle of the heatsink.

 

[Xzibit]

The rear backplate fan will be fighting with the CPU fan for fresh air if you don't have an open top case. Plus the I/O fan will feed hot air from the backplate fan due to case intake fans. Mounting the AIO in front of the case intake effect.

Its an interesting take

 

[Applewarlord]

It's very possible that this front/rear fans setup is more efficient than the classic one-sided fans setup, hence they'd like to implement it for the first time.
If it's worse, why bother at all ?

Very likely, considering the new open-air fin-stack design provides more surface area for heat exchange and easier future maintenance. I love it alrdy.
Altho they could've gotten the idea of extending the heat-sink/fan-shroud over the PCB from someone else, say, Sapphire 5700xt Nitro+, Nvidia sure took it up a notch.
Another interesting thing is that the two white stripes we can see faintly from the rear of the card under the semi-seethrough plastic back.

My first guess is they are heat-interface materials usually stick on VRMs but what are they doing on the back of the board? There's no metal backplate to assist on spreading heat. Maybe it's just a habbit. Don't know they are doing it

 

[Caring1]

People will have to rethink case fans on the bottom of the case blowing air up towards the GPU.

 

[Applewarlord]

Also in case no one noticed there is almost no place for the die to be on board. You can't have the fan directly over the die as that would limit the area of the coldplate and heatpipes too much.

IMO the fan on the front side of the card is NOT over the GPU but over VRM instead. It's a non-conventional board design. Notice in the rear-view picture there are two faint white stripes under the plastic see-through back plate? They could be thermal-interface materials which in recent years make contact with a metalic back plate. Truely I have no idea what they are doing here but it could be a human error. Someone just slapped it on by habbit.

 

[Ruru]

I'd say that it's a mockup or a prototype if it's not a fake. Well, there's been interesting designs so I can't see why something like this couldn't exist.

 

[LemmingOverlord]

Almost as corny as Intel's Xe renders.

So Nvidia concludes that instead of a blower reference design, it's going for a sucker reference design????? Ironic.

 

[Vayra86]

The aesthetic is a real departure for Nvidia. I'm not sure it's hmm, manly enough for a top end card as strange as that may sound. Maybe it'll grow on us?

Shrouded in mystery, that. But manly enough? We have waifu cards now, it will work out just fine. Manly is a Korean esports team looking badass in 2020

 

[reflex75]

This design will create hot air loop.

 

[efikkan]

Frankly, I wouldn't care if the card were My Little Pony themed. It's going into a box, I care about performance, noise, airflow, temperature and price of course.

For those who do care about looks, rest assured there will be plenty of AiB variants.

I only wish at least one of the could go for more of a "tower cooler" design. I wouldn't mind the cooler being 6 slots as long as it pulled the air out in a silent fashion, and a such design would make much more sense in a typical computer case.

 

[SL2]

Seems like the front exhaust will feed the rear fan with hot air. If the fans were on the same side then I guess the heat from the front exhaust would go directly to the CPU area..
View attachment 158066

I changed my mind. As someone pointed out, this cooler has probably a vapor chamber. If so, how can it possibly exhaust air in the underside of the card? (see pic in quote) Unlike heatpipe heatsinks, vapor chamber heatsinks doesn't let air straight through. Like someone else said, we can't see anything between those fins on the underside anyway, which is highly unlikely if that indeed was an exhaust. Maybe it's just fins on the underside of the vapor chamber, which makes this part (front, underside in case) passively cooled. (Yeah it can be a combination of VC and heatpipes, but still, we can't see anything in there anyway.)
Here's the vapor chamber of the 2080 TI FE. Pretty much no air, or light, comes through.

 

[8BitZ80]

The design seems far too elaborate and refined to be what I would consider "fake". Knowing Nvidia, they've been trying to produce more out-of-the-box designs and move away from the standard multi-fan design. I mean hey, a radical change is a great way to justify those radical MSRPs.

 

[Vya Domus]

I changed my mind. As someone pointed out, this cooler has probably a vapor chamber. If so, how can it possibly exhaust air in the underside of the card? (see pic in quote) Unlike heatpipe heatsinks, vapor chamber heatsinks doesn't let air straight through. Like someone else said, we can't see anything between those fins on the underside anyway, which is highly unlikely if that indeed was an exhaust. Maybe it's just fins on the underside of the vapor chamber, which makes this part (front, underside in case) passively cooled. (Yeah it can be a combination of VC and heatpipes, but still, we can't see anything in there anyway.)
Here's the vapor chamber of the 2080 TI FE. Pretty much no air, or light, comes through.
View attachment 158149

No open air cooler is designed such that air comes through.

 

[SL2]

No open air cooler is designed such that air comes through.

My post isn't based on what you've seen, or haven't seen, I'm talking about what everyone else is talking about in this thread.
This thread is 99 % speculations, and there are countless posts that suggests so.

 

[Vya Domus]

My post isn't based on what you've seen, or haven't seen, I'm talking about what everyone else is talking about in this thread.
This thread is 99 % speculations, and there are countless posts that suggests so.

It doesn't matter, open air coolers are simply designed to dissipate the heat through the sides of the heatsink not directly through it, that's not a speculation it's fact.

 

[SL2]

It doesn't matter, open air coolers are simply designed to dissipate the heat through the sides of the heatsink not directly through it, that's not a speculation it's fact.

That doesn't stop anyone from doing different. I have no idea what you're trying to prove here. It's not like such design defies physical laws.

 

[Decryptor009]

It doesn't matter, open air coolers are simply designed to dissipate the heat through the sides of the heatsink not directly through it, that's not a speculation it's fact.

Anyone with 2 brain cells could tell you that air cannot move through solid metal, but upon contact it cools it depending on temperature of the air.

The flow of air once it hits though depends on which direction the gaps dictate.

Saying they are designed that way is not correct, it is physically impossible for air to dissipate any other way with such a design.

 

[SL2]

Or maybe we're talking about two different things.

 

[yeeeeman]

It's very weird why AMD did what they did with Navi 1st generation.
Why no Big Navi and why not competing with Vega 64/ Vega 64 derivatives in the below Big Navi segments.

They had a 256 mm2 HD 4870 but back then they had a card with 70-80% higher performance.

Now they have nothing.

And if the rumours are true, the smaller Navi 2X chips won't be out before 2021.

I can tell you why by talking to some engineers involve in its making.
rdna 1 is a new uArch with lots of changes and because the teams are small and the task is huge (nvidia is much larger than AMD), budget/timing constraints mean that navi 10 had a lot of hardware bugs that meant it is not efficient and required a lot of software tweaking (read drivers) to even make it work. You can see all these traits in the 5700xt. It eats a lot of power for a small-ish chip on 7nm, that even turing on 12nm is better. As regarding the drivers and the bugs...people are still complaining about crashes, so yeah. They were forced to sell it at a low price and create lots of subproducts from it in order to sell as many dies as possible, unfortunately for cheap. The strategy worked partially but many users got frustrated with bad drivers (in essence the drivers are OK, but you can do so much to hide hardware bugs) so they switched to nvidia. Anyway, this meant they couldn't launch a big die gpu based on rdna1 within reasonable power consumption/budget and thermal, so the focus was shifted on rdna 2 which is an improvement in many regards to rdna 1 but especially it has bug fixes for all the power/functionality related stuff.
So short answer to your question is that AMD cannot really make a great chip with new uArch from first try. If they would have succeeded to make rdna 1 what rdna 2 will be, then nvidia would have had a hard time competing with 2000 series. The 2080 ti would sell with 400-500$.

 

[Nero1024]

Disappointing design if it will actually end up being the real thing. Maybe one of the prototypes?