How Does the GTX 1080 Ti Stack Up in 2020?

[Julio Franco]

Today we're taking a look back at the mighty GeForce GTX 1080 Ti, the previous-gen Nvidia flagship that has become somewhat of an iconic GPU, and for good reason. We were impressed with what Nvidia managed to achieve at the time. The 1080 Ti was designed to enable a level of performance never seen before and it accomplished just that.

Read the full article here.

https://www.techspot.com/review/2017-geforce-1080-ti-vs-rtx-2070-super-vs-radeon-5700-xt/

 

[Vulcanproject]

The GTX10 Pascal cards in general were just very good. 1080Ti tops the lot and it's easy to see why nobody was in a hurry to move to the RTX replacements when performance is still so strong.

This is what I emphasize about building a PC, so much is about buying the right hardware at the right time. It can be expensive but you can also get your money's worth as this shows. I doubt anyone who had a 1080 or 1080Ti soon after launch and still have it today regret their purchases.

Buy a reasonably fast card right at the start of a new GPU generation, when there is significant increase in performance and you'll get years of top quality gaming out of it. It'll still be worth a considerable amount when it comes to trade in against the next GPU. The cycle continues.

 

[krizby]

How come the 2070 Super is 9% faster than 5700XT here (without DLSS) but 7% when compare directly the 2 in 2070S vs 5700XT article ? Some rounding errors ?

 

[kmo911]

Still aming for good gaming. other cards with 8 gb run out o ram in some games. try out wolfenstein youngblood for 4 gb -8gb and then run 4k on gtx 1080 ti FE. still a lovely card that can emulate rtx rts cores low qualyty.

 

[poohbear]

Bought a used Gigabyte Aorus GTX 1080ti for my new 34" alienware widescreen gaming monitor at the beginning of covid19 lockdown for $425. sold my GTX1070 for $270, so can't complain. Its a great card that's serving me well for all the gaming im catching up with during this lockdown. The $1200 RTX 2080ti is out of the question for me, so ill be holding on to this 1080ti well into 2021.

 

[Lounds]

It would be interesting to test the 5700XT against the 1080ti in new games in 3-4 years time. The evidence always shows radeon holds up better past the 5 year mark on new games.

 

[enemys]

For me it still holds on awesomely. Works flawlessly in VR, 4K and high refresh rate 1440p gaming, especially since I rarely play AAA titles at launch, sticking mostly to games released a couple of years ago. While I'm not very fond of Nvidia in general, I have to admit 1080Ti is one of the best hardware pieces I've ever bought.

 

[krizby]

It would be interesting to test the 5700XT against the 1080ti in new games in 3-4 years time. The evidence always shows radeon holds up better past the 5 year mark on new games.

How does the Fury X hold up against 980Ti in 2020 , assuming there are still working Fury X out there....
Anyways a 700usd 2017 card took a dump on a 400usd 2019 one is a massive shame already. I still remember how the 250usd HD 4890 killed the 650usd 8800 GTX which released 2 years prior. How the mighty ATi have fallen...

 

[amstech]

The GTX 1080 is still a beast as well.
I believe the RTX 2070 has pulled ahead of it some but overall they are quite close.

 

[ZackL04]

Nice fun to read article Steve.

Makes me feel even better about the reference 5700 I just bought and flashed with XT BIOS for $250

 

[QuantumPhysics]

1080Ti 1440p Benchmarks

1440p Average = 101.0 fps
Ultrawide 1440p Average = 86.3 fps
Average 4K performance = 60.3 fps





The 1080Ti was a beefy piece of hardware which was only superseded by the 2080Ti.

But keep in mind, the Titan X, Titan Xp and RTX Titan still do exist and in many cases Titan/ Titan Xp can be purchased cheaper than 1080Ti which are dried up on the use market since people willing to spend money on those are more likely to go for a 2080Ti.

Nvidia is a power player that is determined to dominate the market.

My issue is, the bulk of the market really doesn't need this level of power when you analyze their systems. STEAM shows us the majority of registered users use a GTX 1060.

The majority of online gamers don't have a 4K monitor... many don't have 1440p either.

1080p gaming @ 60fps is still the vast majority of the audience in 2020.

The other tell is the benchmarks you provide showing the 2060 Super exceeding 60 fps at both 1080p and 1440p in every test, as well as the fact you didn't bother doing any 4K testing for these benchmarks.

So many new games are designed to run on new gen Core i3 with just a 1060 or better that a car like the 1080Ti, 2080Ti or TITAN is unnecessary unless you are willing to spend more for bragging rights or future proofing (investment).

I have an Alienware 1440p 34" gaming monitor running on a 2080Ti. According to these results, I could have gotten a 1080Ti or a 2060 Super.

But then I wouldn't be able to access Ray Tracing on the level I can.

Therefore, the new standard of measurement should include whether or not Ray Tracing is used, in addition to frames per second in addition to being able to game in 1080p, 1440p or 4K.

The 2080Ti is simply the only card right now that can check all those boxes.

 

[godrilla]

"Moving on to the breakdown with the Radeon RX 5700 XT, overall the GTX 1080 Ti was 11%"
The better comparison is the 1080ti vs 5700xt . When asking why are people paying more for a used 1080ti 3 years later? The problem goes back to the root cause of lack of competition. While the 5700xt is a great gpu it still does not dethrone a 3 year old flagship. The 2070 super on the other hand and anything rtx today is a bad value because of lack of competition in raytracing and other features. This is probably the worst time to purchase a gpu. Once amd comes out with a rdna 2.0 gpu then these 1080ti will fall hard in price in the second hand market. If you need a gpu the 5700xt is probably the best value and will also hold its value in second hand market if one decides to upgrade later on this year vs the competition.
Also there is absolutely no need to pay for rtx premium unless you want to use those features.

 

[nnguy2]

I have an Alienware 1440p 34" gaming monitor running on a 2080Ti. According to these results, I could have gotten a 1080Ti or a 2060 Super.

LOL was wondering if you were gonna mention your monitor and gpu...was not disappointed

 

[veLa]

Just goes to show you what a beast the GTX 1080 Ti was when it came out and how little the bar has moved since it set the bar.

I think Steve Burke from Gamers Nexus said it best when NVIDIA released the RTX 2080 Super: We get it NVIDIA, you can make a 1080 Ti.

 

[hover389]

Why are you testing this class of card at 1080p instead of 4K? Sure some people love gaming on 1080p screens with a very high framerate but to not have 4k testing results doesnt make any sense.

 

[QuantumPhysics]

Why are you testing this class of card at 1080p instead of 4K? Sure some people love gaming on 1080p screens with a very high framerate but to not have 4k testing results doesnt make any sense.

For the time it was released, 1080p gaming was the standard.

Regardless what you may here, 4K gaming isn't mainstream yet. The majority of monitors being sold now are 1440p or 1080p.

The 1080Ti averages around 60 fps or less at 4K. I don't hold that number against it because it' came out so many years ago.

The 2000 series should be measured against 1440 primarily and 4K secondly.

 

[Evernessince]

How come the 2070 Super is 9% faster than 5700XT here (without DLSS) but 7% when compare directly the 2 in 2070S vs 5700XT article ? Some rounding errors ?

Aside from just variations in silicon between any two chips, margin of error is 3-5%

My issue is, the bulk of the market really doesn't need this level of power when you analyze their systems. STEAM shows us the majority of registered users use a GTX 1060.

The steam survey isn't and has never been an accurate representation of the market. Case in point: https://www.pcgamer.com/chinese-is-...ar-language-according-to-its-hardware-survey/

How does the Fury X hold up against 980Ti in 2020 , assuming there are still working Fury X out there....
Anyways a 700usd 2017 card took a dump on a 400usd 2019 one is a massive shame already. I still remember how the 250usd HD 4890 killed the 650usd 8800 GTX which released 2 years prior. How the mighty ATi have fallen...

About as well as it did at launch, which isn't great. Even being $50 cheaper than the 980 Ti, you were only getting full value if you used it's compute performance.

 

[amstech]

The steam survey isn't and has never been an accurate representation of the market.

When 75%-80% of the discrete GPU market share is owned by Nvidia, the 86% of Steam users having Nvidia GPU's is very accurate. These are not polled results, these are results taking directly from real world users and real world systems.
IMO, and for many other people I don't think there is a more accurate market representation, going on about 10 years.

 

[Evernessince]

When 75%-80% of the discrete GPU market share is owned by Nvidia, the 86% of Steam users having Nvidia GPU's is very accurate. These are not polled results, these are results taking directly from real world users and real world systems.
IMO, and for many other people I don't think there is a more accurate market representation, going on about 10 years.

Better start learning Chinese then, because according to that logic it's now the dominant language playing games and thus the language devs should program to.

Steam is so real world, they've had issues multiple times in the past with their numbers and clearly still are when it comes to counting multiple users per PCs.

Sounds like an excuse for the fact that steam doesn't provide any transparency about how it gathers it's numbers and it's slow to fix problems with them.

I think you are one of the few that do believe steam is an accurate representation of the market.

 

[Rayneofpayne]

How come the 2070 Super is 9% faster than 5700XT here (without DLSS) but 7% when compare directly the 2 in 2070S vs 5700XT article ? Some rounding errors ?

Driver tweaks? The software changes with time and can alter performance also different versions of DX and Windows can have an effect on performance, Why do people still make comments like this, usually they list hardware and software with a method of testing for this very reason.

 

[Mike89]

All I can say is my plane jane EVGA 1070 is still kicking butt at every game I throw at it. I still play at 1080p and am not even thinking about a video card right now. This thing still plays all my games smooth as butter and I don't even have to put syrup on it.

 

[ZorracK]

Great article about a great card!
good to see that kind of benchs!
I'm glad now that I didnt swap my 1080ti for a 5700XT OC Se

 

[amstech]

I think you are one of the few that do believe steam is an accurate representation of the market.

Yeah I don't think so.
And not only is Steam very accurate, its even better for getting gaming results because while you can get other apps on Steam, lets be honest/real here, its a gaming application and 95% or more are just, mostly gaming. If anything this makes their hardware results even more distinguished.

 

[hel0sn8]

All I can say is my plane jane EVGA 1070 is still kicking butt at every game I throw at it. I still play at 1080p and am not even thinking about a video card right now. This thing still plays all my games smooth as butter and I don't even have to put syrup on it.

Exactly! 1080 here, came from a 980. 100% feel the same as you. Will last for a good while yet. 1080p @ 144Mhz is what I like. And it'll be good enough for time to come.. This article shows that a 1080ti still kicks arse and if you got one or a nearby G card at good price then your still laughing.

Sell / buy at the right time and hold for as long as poss till value starts to dip and get a cross over upgrade for minimal money. I've always upgraded that way. Just need to keep an eye on the value of things. Right now with COVID is the worst time..

Some people always always, side step the point of an article to lay down their agenda or POV which honestly most of us just dont care about. I mean seriously; 1 showing their AMD love, claiming will last longer than Nvidia. 2 showing off they have latest high end 2080ti and wanting to make us unworthy, when in fact they just show how much a mug they are, thinking they are better etc. 3 going off topic and starting an argument about which has market domination - Red or Green and then totally getting it all wrong, especially in the language department (seriously, is it too hard to beleive that chinese language is becoming very very common! LMAO Pretty sure programmers go out of their way to include the language for obvious sales reason.)

Get the point of an article or roundabouts and comment accordingly. Anything else belongs in a new Thread.

 

[krizby]

Driver tweaks? The software changes with time and can alter performance also different versions of DX and Windows can have an effect on performance, Why do people still make comments like this, usually they list hardware and software with a method of testing for this very reason.

The 2070 Super vs 5700XT article was last week, Techspot (or Hardware Unboxed) use the same card for every testing anyways. What hardware they use are only mentioned on their videos.