NVIDIA GeForce RTX 2080, 2070, and 2080 Ti Specifications Revealed

[Manoa]

@SomethingnewI can say that they've lost the battle yet again & the end users; which is us, are paying the price.

it the true, regardless of if the RT/Tensors are silicon that was worth to spend, such price for graphics is too much - better buy a gun for this price and start shooting people it will give the best ray tracing EVER and unlimited bit-depth graphics lel

 

[bug]

@Somethingnew also, how sure are you that the RTX2070 will be "the same performance of GTX1080" when there isn't any solid evidence being circulated or even proven? How sure are you that it'll be that bad? Does your word carries weight or you're just afraid of a new architecture that made the entire GPU market deemed obsolete because of how it does in ray tracing? Please keep your thoughts to yourself until reviewers got their hands on the reference cards, bench them with current release drivers & write up their findings.
Anyways...
With AMD having no luck or having time constraints in releasing their upcoming Navi or shrunken down Vega cores to keep up with Nvidia's new silicon, I can say that they've lost the battle yet again & the end users; which is us, are paying the price.

You do realize you're talking to someone who has just signed up yesterday and only littered threads about Turing with his posts? Almost like he has an agenda

 

[efikkan]

Most of the talk around the web seems to be heavily negative, especially regarding the performance of the chips; the one thing we don't actually know. There will be a lot of people eating their words when the benchmarks arrives…

BTW; did anyone catch the date of the review embargo?

 

[bug]

Most of the talk around the web seems to be heavily negative, especially regarding the performance of the chips; the one thing we don't actually know. There will be a lot of people eating their words when the benchmarks arrives…

BTW; did anyone catch the date of the review embargo?

September 20? Till then, we have this: https://www.techradar.com/reviews/nvidia-geforce-rtx-2080-ti

 

[Fluffmeister]

Most of the talk around the web seems to be heavily negative, especially regarding the performance of the chips; the one thing we don't actually know. There will be a lot of people eating their words when the benchmarks arrives…

BTW; did anyone catch the date of the review embargo?

That's how this dance works sadly, there is literally not a single scenario where Nvidia could win. Sure Pascal is annoyingly fast and efficient but it's "boring" and "2 years old". Then rumours suggest Ampere/Turing might be nothing more than tweaked Pascal on what amounts to be the same node... how boring!

Now it turns out Turing is huge chip aimed and bringing real time ray tracing closer to reality, throw in AI and the like too... but now it turns out there is no support and no games!!!!111 Also it's not really fast enough as hasn't solved real time ray tracing instantly! O M G FAIL! Maybe they shouldn't have bothered and just made another Pascal +[insert percentage increase here]. But that would obviously be boring and milking the market!

You don't innovate, they hate you. Innovate, they hate you.

 

[efikkan]

Then rumours suggest Ampere/Turing might be nothing more than tweaked Pascal on what amounts to be the same node... how boring!

Well, those rumors are wrong, and if people payed attention they would know that. The SM structure is completely redesigned vs. Pascal. We can't compare them based on CUDA core count etc. because the performance characteristics will be very different. Predicting game performance based on the theoretical specs from Nvidia is the dumbest thing anyone can do right now.

 

[Tsukiyomi91]

time & again, everyone cared about the paper numbers when they haven't even seen anyone benching the cards yet. When Sept 20th come & go, when the benchmarks starts rolling, I wanna see the look on those people's faces & tell them off to eat back at their words.

 

[carex]

You're forgetting that Vega 64 is basically Fury X with a few minor architectural tweaks:
Fury X: 8.9 million transistors, 596 mm², 28 nm TSMC
1080 Ti: 11.8 million transistors, 471 mm², 16 nm TSMC
Vega 64: 12.5 million transistors, 510 mm², 14 nm GloFo
2080 Ti: 18.6 million transistors, 754 mm², 12 nm TSMC

I think NVIDIA wasting their time on RT/Tensor gave AMD a window to steal the gaming crown.

i personally think VEGA was incredibly important for amd as it introduces infinity fabric inside fiji is altogether a different animal

It's faster in RTX which means nothing for 100% of games out there except for the four announced ones which are not even in a beta stage. Nothing is known about the old-school graphics at this point. Reviewers are just receiving the cards. And the RTX 2070 at $500 does sound terrible considering the pricing of the previous generations of NVIDIA cards. NVIDIA has no one to compete with and it shows. At this point I would cease all the debates and wait for reviews to come in.

@W1zzard When will reviews be released? Or even this date is under NDA? ;-)

without review i can easily calculate it will be exactly ~18.5% faster than the 1080ti
about the overclockin ~1900 will be the limit

 

[Prima.Vera]

Most of the talk around the web seems to be heavily negative, especially regarding the performance of the chips; the one thing we don't actually know. There will be a lot of people eating their words when the benchmarks arrives…

Nah, NOT A CHANCE! Did you see the Titan V performance numbers? On average 20-25% faster than the 1080Ti, and that's all, nothing spectacular. And on top of that the 2080Ti even has lower specs than Titan V, so what are we dreaming about here.

Just wake up people, nGreedia is pulling probably the biggest scam in the IT industry. Almost as bad as Apple slowing down their phones with every new IOS release. Almost...

 

[carex]

Just wake up people, nGreedia is pulling probably the biggest scam in the IT industry. Almost as bad as Apple slowing down their phones with every new IOS release. Almost...

we have to analyze each driver update performance from here on there should be a dedicated thread related to this.