nVidia RTX 2080 Ti "Game Performance Preview" 100FPS@4K on Ultra

SixFootDuo

Supreme [H]ardness
Joined
Oct 5, 2004
Messages
5,825
https://wccftech.com/nvidia-geforce-rtx-2080-ti-performance-preview-4k-100-fps/

Seems WCCTech was able to play around with one of the cards and noted that performance was exceptional.

Triple AAA titles hitting 100FPS @4K on ultra settings.

This is exactly the kind of early news I was hoping for.

I pre-ordered 2 x 2080 Ti's yesterday. A MSI Trio and the Gigabyte 3 Fan model. Aurores or something
 
Last edited:
https://wccftech.com/nvidia-geforce-rtx-2080-ti-performance-preview-4k-100-fps/

Seems WCCTech was able to play around with one of the cards and noted that performance was exceptional.

Triple AAA titles hitting 100FPS @4K on ultra settings.

This is exactly the kind of early news I was hoping for.

I pre-ordered 2 x 2080 Ti's yesterday. A MSI Trio and the Gigabyte 3 Fan model. Aurores or something

If true, will be beastly of a card even without RT but of course, gonna have to wait until Kyle and Brent get their hands testing the cards.
 
Numbers will be out way before HardOCP releases them.

I think nVidia is releasing drivers today at Gamerscon? I'm not exactly sure. Reading the same shit everyone else here can go and read.

We will see.

I will say that there is a huge clue a lot of you are just not getting. The $1,200 ( $1,300+ ) with tax that the 2080 Ti costs. nVidia would never ever charge that much for a card that only had a 15% or 20% performance gain over the 1080 ti. If that happens. I can absolutely without a doubt promise you .... a lot of pre-orders would be returned and refunded including both of mine.

I would wait a long ass year + and just keep using a 1080 ti like so many others would do if this happens.
 
no charts to back it up? I'm willing to bet performance increase is less than 20%. Everyone is paying for Ray Tracing Gimmick. That's why the chip size is so big and the cost is so high. Everyone good luck with your $1200 pre-order flop LOL. Get ready to cancel that pre-order real quick.
 
...

I will say that there is a huge clue a lot of you are just not getting. The $1,200 ( $1,300+ ) with tax that the 2080 Ti costs. nVidia would never ever charge that much for a card that only had a 15% or 20% performance gain over the 1080 ti. If that happens. I can absolutely without a doubt promise you .... a lot of pre-orders would be returned and refunded including both of mine.

I would wait a long ass year + and just keep using a 1080 ti like so many others would do if this happens.

I pre-ordered, and in the same mindset. If the #'s disappoint it will just get returned, and I'll go back to using my Titan X. If the card kicks ass, I'll just sell my Titan X for whatever.
 
Can we credit the source https://www.techradar.com/au/reviews/nvidia-geforce-rtx-2080-ti rather than the reposting, makeshitupallthetime asshats at wccftech.

Happily it not being wccftech makes me a lot more confident in the news.

Tech radars performance preview is also painfully one-sided. They didn't mention the demo running shadow of the Tomb Raider at 30 frames per second in actual gameplay at 1080p. While it's nice to fluff up numbers so that every eager gamer gives your website hits and spawns posts like this around the internet, it is egregiously deceptive and is one of the reasons why until you get the numbers from a neutral source with the card in their hands and the ability to benchmark in an environment that is not controlled by Nvidia from top to bottom you're just debating unicorns.
 
Tech radars performance preview is also painfully one-sided. They didn't mention the demo running shadow of the Tomb Raider at 30 frames per second in actual gameplay at 1080p. While it's nice to fluff up numbers so that every eager gamer gives your website hits and spawns posts like this around the internet, it is egregiously deceptive and is one of the reasons why until you get the numbers from a neutral source with the card in their hands and the ability to benchmark in an environment that is not controlled by Nvidia from top to bottom you're just debating unicorns.

Wholeheartedly agree. I just followed the link and throught there should be attribution. Though interestingly that would be a breach on their NDA. On reviews, Kyle basically informs me of my primary home IT spend. Trusted him for almost 20 years and have never been let down.

If he reviewed cars I'd probably buy that too.


This is a lie, I'd have ended up driving some f350 /H2 monstrosity and never forgive him
 
Looks interesting. Pre-ordered to FE 2080 Tis on Monday, says 10/08/2018 arrival date for now. While the pricing isn't where anyone wants it to be on these cards the performance at least at 4k might be good enough to make the higher pricing at least not seem quite like gouging.
 
Numbers will be out way before HardOCP releases them.

I think nVidia is releasing drivers today at Gamerscon? I'm not exactly sure. Reading the same shit everyone else here can go and read.

We will see.

I will say that there is a huge clue a lot of you are just not getting. The $1,200 ( $1,300+ ) with tax that the 2080 Ti costs. nVidia would never ever charge that much for a card that only had a 15% or 20% performance gain over the 1080 ti. If that happens. I can absolutely without a doubt promise you .... a lot of pre-orders would be returned and refunded including both of mine.

I would wait a long ass year + and just keep using a 1080 ti like so many others would do if this happens.

I wouldn't be sure the increased price is evident of a big performance increase. As others have said this card may be good at mining. If nvidia knows that and also knows AMD has no competition for these cards then they can charge virtually whatever they want, even if the in-game performance is slight compared to the 10 series.
 
Ita a cash grab, like ive heard, never before have the ti's dropped before the base line and never before has performance comparisons been ommited ... The focus was on RT.. Irrelevant today, but it makes you feel your pascals are dinosaurs now and You MUST upgrade!!

Of course, i want to be wrong. Just one skeptics opinion when raw data isnt provided and im never one to quickly hand cash over simply because you say what you have is better when the only way you said its better is in RT (equiv to 10 1080tis for rt lol)
 
In addition to the 4K gaming performance, it is also stated that some games (which they obviously can’t name explicitly) were able to run well over 100 FPS at 4K and using Ultra settings

The games that run at over 100fps at 4K have been confirmed. Screenshots:

?u=https%3A%2F%2Fstatic.3drealms.com%2Fmedia%2Fscreenshots%2Fc7185590cf634773b4d56490e26aec1e.jpg


?u=http%3A%2F%2Fdoomwadstation.net%2F2016%2Fdoomstein%2FScreenshot_Doom_20160307_172004.png
 
Maybe if I win the lottery and have tons of extra money to burn on shit.
And even then, I'd probably be kinda sandy about it.
Otherwise, hard pass.
 
no charts to back it up? I'm willing to bet performance increase is less than 20%. Everyone is paying for Ray Tracing Gimmick. That's why the chip size is so big and the cost is so high. Everyone good luck with your $1200 pre-order flop LOL. Get ready to cancel that pre-order real quick.

Ray tracing isn't a gimmick. Not sure why people think this is some gameworks bullshit.

I am not saying these cards are worth their cost or that Ray tracing is worth cutting my fps is half.

But dont talk about ray tracing like its fucking hair works.
 
I will say that there is a huge clue a lot of you are just not getting. The $1,200 ( $1,300+ ) with tax that the 2080 Ti costs. nVidia would never ever charge that much for a card that only had a 15% or 20% performance gain over the 1080 ti. If that happens. I can absolutely without a doubt promise you .... a lot of pre-orders would be returned and refunded including both of mine.

I would wait a long ass year + and just keep using a 1080 ti like so many others would do if this happens.

Riiiiight. Nvidia's top end has always been about providing value to gamers.

Remind me, please, how much faster than the 1080 Ti is the TitanXp?
 
The 2080Ti is going to be about 18% faster than a 1080Ti at base clocks. It looks like Turing may have a frequency ceiling about 5% higher than Pascal, which could make the 2080Ti ~23% faster than 1080Ti OC vs OC. If they're getting 100FPS at 4K on a 2080Ti it's likely an 80FPS experience on a 1080Ti. Of course assuming RTX features off.
 
The 2080Ti is going to be about 18% faster than a 1080Ti at base clocks. It looks like Turing may have a frequency ceiling about 5% higher than Pascal, which could make the 2080Ti ~23% faster than 1080Ti OC vs OC. If they're getting 100FPS at 4K on a 2080Ti it's likely an 80FPS experience on a 1080Ti. Of course assuming RTX features off.

What's your source on 5% higher clocks max OC?
 
I'm seeing ~2100 MHz overclocks on the web. I think most Pascal GPUs do around ~2000...
 
https://wccftech.com/nvidia-geforce-rtx-2080-ti-performance-preview-4k-100-fps/

Seems WCCTech was able to play around with one of the cards and noted that performance was exceptional.

Triple AAA titles hitting 100FPS @4K on ultra settings.

This is exactly the kind of early news I was hoping for.

I pre-ordered 2 x 2080 Ti's yesterday. A MSI Trio and the Gigabyte 3 Fan model. Aurores or something

I hope you didn't base your purchasing decision off Wccftech information? They are just regurgitating the info from the Tech Radar website. And I have massive doubts about the performance figures they mention, they seem contrary to every other website out there. I sometimes wonder was Tech Radar even at the same event!! They are claiming that Tomb Raider ran at 50-57 fps at 4K with full ray tracing. When every other site said it was running at 1080p and was dipping to 30fps a lot. Even the developer acknowledged the performance wasn't that good.

So if Tech Radar got that so wrong, can we believe the rest of their claims? over 100FPS at 4k on triple AAA titles at Ultra settings? Brilliant!! if true. But, I have my doubts.
 
I love how the article tries to brush off the 1080p Shadow of the Tomb Raider performance because the game is 'in such an early state'. The game releases on 9/14 how much optimization can they possibly get if the game is probably already gold by now.

50-57 FPS at 1080p does highlight that the ray tracing features are very demanding for these cards.
 
I love how the article tries to brush off the 1080p Shadow of the Tomb Raider performance because the game is 'in such an early state'. The game releases on 9/14 how much optimization can they possibly get if the game is probably already gold by now.

50-57 FPS at 1080p does highlight that the ray tracing features are very demanding for these cards.

Ray tracing is extremely demanding, you should see rendering farms dedicated to it for movies. Granted it's not the same level here but it goes to reason it's going to hurt fps if you turn it on, I expect it to be near useless at 4K for most. But who knows maybe they will pull a rabbit out of their hat but seems unlikely based on what Shadow of the Tomb Raider is showing us.
 
I hope you didn't base your purchasing decision off Wccftech information? They are just regurgitating the info from the Tech Radar website. And I have massive doubts about the performance figures they mention, they seem contrary to every other website out there. I sometimes wonder was Tech Radar even at the same event!! They are claiming that Tomb Raider ran at 50-57 fps at 4K with full ray tracing. When every other site said it was running at 1080p and was dipping to 30fps a lot. Even the developer acknowledged the performance wasn't that good.

So if Tech Radar got that so wrong, can we believe the rest of their claims? over 100FPS at 4k on triple AAA titles at Ultra settings? Brilliant!! if true. But, I have my doubts.

Yeah that’s really vague. Assuming 5% higher clock, faster VRAM, and 21% more CUDA cores. Lets assume faster VRAM offsets the inefficiency of the cuda core increase.

That’s at least 27% faster in all applications completely ignoring the 1/3 of the die with tensor and ray tracing cores than a 1080ti OC vs OC. Use some of the new tech and it destroys everything else (?). Is that enough in some titles? Heh
 
Ray tracing is extremely demanding, you should see rendering farms dedicated to it for movies. Granted it's not the same level here but it goes to reason it's going to hurt fps if you turn it on, I expect it to be near useless at 4K for most. But who knows maybe they will pull a rabbit out of their hat but seems unlikely based on what Shadow of the Tomb Raider is showing us.

I'm aware of what ray tracing is but the way they marketed these cards as being a Ray Tracing/Rasterizaion combo that is ushering in the next gen graphics I would have thought it would be more optimized than that. Especially since that is the big feature of these $600+ cards.

If the cards can barely muster 50-60 FPS in 1080p they should have waited until at least the next generation to hype this technology. That is unless they just want it to gain steam for later which would make total sense.
 
Last edited:
I’ll post this here too. It’s not uncommon for games to improve now, even AAA like Battlefield, a week or two after launch. Maybe wait for the card to actually launch? Shit it’s not even shipping yet.

AF59CA22-B070-47DC-8086-22555765B503.jpeg
 
Their charts have been spot on in the past... it’s usually AMD that does funny shit (Fury X, RX480, ect.)

It is higher than I would have expected. BS meter tingling a little. It’s where I would expect the 2080ti not the 2080.
 
Their charts have been spot on in the past... it’s usually AMD that does funny shit (Fury X, RX480, ect.)

It is higher than I would have expected. BS meter tingling a little. It’s where I would expect the 2080ti not the 2080.
Agreed. It's higher than the CUDA core count and clock speeds would suggest. Interestingly most of the benches are in HDR, which was shown to have a measurable performance cost on Pascal. I wonder if Turing has some HDR improvements that are contributing to the gains on this chart.

Really looking forward to independent reviews.
 
Did you guys actually read the article that wccftech linked to? it was the biggest click bait ever. It was labeled as hands on review. Then we think this and that. I mean hitting 100fps and sustaining 100fps are totally two different things.
 
Agreed. It's higher than the CUDA core count and clock speeds would suggest. Interestingly most of the benches are in HDR, which was shown to have a measurable performance cost on Pascal. I wonder if Turing has some HDR improvements that are contributing to the gains on this chart.

Really looking forward to independent reviews.

I wouldn't be surprised if this is a straight up money grab experience for Nvidia to rack in cash before 7nm. HDR only results, think about it. I think they are pushing Gsync HDR monitors along with these cards, clearly suggests with the info they are giving out. Grab a 1200 dollar card and spend another 2k on monitor. Seems like high end graphics is turning in to luxury now. I don't see myself rushing to buy 1200 dolalr GPU and spend 2500 on gsync hdr monitor to enjoy. I will wait a few generations until the cost comes down to earth and let the early adopters fund the technology for me. Hahahha.
 
This is promising for rasterized performance:

View attachment 98094

DLSS I think is in replacement of AA. So I can believe that at such high resolution. So That chart can really be deceiving. The way nvidia is representing the charts avoiding detailed results makes me iffy. They sound like AMD right now with those results. DLSS at 4k can increase performance as it is not using AA on the entire image I think, its heavily AI based so it will select what to smooth out.
 
Back
Top