This is an uncomplicated comparison between the previous generation GeForce flagship, the GTX 1080 Ti, which competes in 39 games against the RTX 2070 Super and the RX 5700 XT. We will learn how the GTX 1080 Ti holds its own compared to these new "midrange" GPUs. If you should buy one, this will be answered before this article ends.
For testing the red team we have the PowerColor RX 5700 XT Red Devil and for the green team the MSI RTX 2070 Super X Trio and the MSI GTX 1080 Ti Gaming X Trio. Our standard GPU test bench was used, which includes a 5 GHz Core i9-9900K with 16 GB DDR4-3400 memory. We tested all games with 1440p and 4K resolutions for a total of 39 titles.
As usual, we'll go through some of these game benchmarks where we see remarkable data to highlight and towards the end of the feature have some big breakdown charts that compare all performance data.
First we have the newly released Gears 5. The GTX 1080 Ti dominated the 2070 Super and 5700 XT, both newer GPUs achieved an average of 76 fps at 1440p, while the Pascal GPU achieved an average of 88 fps. At 4K, the 1080 Ti was also 14% faster than the 2070 Super and 25% faster than the 5700 XT.
The controls are another innovation in our series of benchmarks, but this seems to be very unoptimized. All three GPUs couldn't reach more than 60 fps at 1440p, so that's pretty awful. The game looks good, but it's not amazing and certainly not good enough to justify this level of performance. At least for the moment, Control seems to be a real FPS pig.
Next up is War Thunder and here the 1080 Ti outperformed the 2070 Super at 1440p by 13%, although both delivered a similarly low 1% performance. The 1080 Ti also knocked out the 5700 XT by 34%, and we're seeing a similar thing at 4K, despite a similarly low 1% performance.
World of Tanks plays very well on the GTX 1080 Ti, here the Pascal GPU was 15% faster than the 2070 Super and 24% faster than the 5700 XT. This is a solid result for the older 1080 Ti, although frame rates were high on all 1440p GPUs at maximum setting. The 4K result was more impressive: 94 fps on average of the 1080 Ti enabled an extremely smooth gaming experience.
The GeForce GPUs are slightly stronger in Division 2, for example the 1080 Ti was 13% faster than the 5700 XT at 1440p and 15% faster at 4K. The 1080 Ti also displaced the 2070 Super with 1440p, but adapted it with the 4K resolution.
Interestingly, both the 5700 XT and 2070 Super beat the 1080 Ti comfortably in the Counter-Strike Global Offensive, but given the frame rates, it's hard to say how important it is. I also doubt that many will play at 4K, but if you do, you can still expect over 200 fps with each of these GPUs.
The Forza Horizon 4 RTX series recently saw a 20-30% increase in performance with the latest drivers. However, there was no performance increase for the Pascal GPUs, and as a result, the 1080 Ti still follows the 5700 XT, at least at 1440p. The results are more competitive at 4K and here the 1080 Ti matched the 5700 XT and was not much slower than the 2070 Super.
We noticed how well the older Pascal GPUs performed well in older titles or at least in titles with outdated game engines. In Fortnite we see that the 1080 Ti outperforms both the 2070 Super and 5700 XT at 1440p and 4K. The 4K resolution was 14% faster than the 2070 Super, so a pretty practical win here.
Like Fortnite, PUBG uses the Unreal Engine 4 and so we see that the 1080 Ti easily hits the 5700 XT, although this time it wasn't much faster than the 2070 Super.
The performance in F1 2019 was practically identical for these three GPUs with average frame rates between 112 and 114 fps. The results at 4K were also very competitive, the GTX 1080 Ti matched the 5700 XT exactly.
Wolfenstein: Youngblood is a relatively new addition to our benchmark series, and it plays very well on the newer Turing GPUs. The 1080 Ti was 10% slower than the 2070 Super and with an average of over 140 fps that's not a big deal. The 1080 Ti was comparable to the 5700 XT at both 1440p and 4K.
All three GPUs performed very well in Apex Legends, especially at 1440p. The GTX 1080 Ti beat the 5700 XT by a small margin, while not being much slower than the newer RTX 2070 Super. At 4K, the GeForce GPUs delivered essentially the same performance, while the 5700 XT dropped about 12%.
To test Rainbow Six Siege, we manually force a 100% render scale with TAA enabled. Here we see all GPUs that can reach an average of over 100 fps at 1440p. The 1080 Ti and 5700 XT delivered a similarly low result of 1%, but the Pascal GPU was slightly faster on average. Compared to the 2070 Super, the 1080 Ti was 8% slower and we see something similar in the 4K resolution.
The last game we're going to watch before jumping into the 39 game glitch is Monster Hunter World.
Here we see very competitive performance with both resolutions tested. The 1080 Ti was a bit faster than the 2070 Super and 5700 XT, but overall the gaming experience was similar.
A quick note on power consumption. We do not take this into account in every game benchmark comparison, as the consumption values do not change compared to the first tests and we refuse to include thermal data, as this can vary considerably from model to model. E.g. Not all RTX 2070 Super graphics cards are operated at the same temperature, so every single model has deviations. The same applies to electricity consumption. Some models are more power hungry than others, but the discrepancy here is less.
When it comes to power consumption, the GTX 1080 Ti GPU consumes around 27% more power than the 2070 Super GPU and around 35% more than the 5700 XT GPU …
When comparing total system consumption, the 1080 Ti consumed 15% more power than the 2070 Super and 12% more power than the 5700 XT. The newer GPUs are more efficient, but have no noticeable impact on your system.
As expected, the two-year-old Pascal GPU is still ahead and can easily keep up with the current performance options from AMD and Nvidia. We only looked at the results for about a dozen titles. So let's see how they are compared across all 39 games tested.
GTX 1080 Ti against RTX 2070 Super
As you can see in the graphic below, the older GTX Pascal was on average 3% faster at 1440p. Overall, this means a very similar performance on average, although in some cases we see big differences when looking at individual titles. For example, the 1080 Ti was 16% faster in Gears 5, 15% faster in World of Tanks, and 13% faster in Quake Champions and War Thunder. It was also up to 10% slower in CSGO and Wolfenstein Youngblood, 9% slower in Forza Horizon 4 and 8% slower in Rainbow Six Siege.
Of the 39 games tested, the margin was 32% or less at 32. Then we saw a margin of 5% or less in 19 of the games tested. As the 3% average difference suggests, performance was largely similar.
GTX 1080 Ti against Radeon 5700 XT
Compared to the RX 5700 XT, the 1080 Ti with 1440p was on average 9% faster, and as this figure shows, the GeForce GPU has the edge in significantly more titles. Outliers include a 34% win in War Thunder and 32% in Vermintide 2, while it was slightly faster in World of Tanks and Metro Exodus.
The GeForce GTX 1080 Ti will remain a very powerful graphics card at the end of 2019, no surprises.
Many of you wanted this test to answer two main questions: First, did Nvidia start giving up Pascal on driver optimization, in other words, did the 1080 Ti fall behind Turing-based GPUs like the 2070 Super? Second, should you buy a 1080 Ti today and what price range should you aim for?
Let's take driver optimization first. We will not look at the Nvidia gimping or nerfing performance of older generations because we have looked at it several times in the past and have always found that the claims are wrong. Sure, sometimes we've seen driver bugs that hurt older generations because the focus is on bringing the current generation up to date. We saw this when Forza Horizon 4 was first released as an example, but Nvidia has always solved these problems.
Forza Horizon 4 is actually a good example. Recently, the RTX 2070 Super saw a 30% increase in performance, while the 1080 Ti saw a zero percent increase in performance. This means that when released, the 2070 Super was 13% slower than the 1080 Ti in this title, today it's 10% faster. But is this an example of how Nvidia does optimizations for Pascal, or is it just an example of how the Turing architecture is better used? We would tend to the latter, maybe now you can better use the simultaneous execution of FP32 and INT32 by Turing.
However, as the only example here, it is impossible to draw a real conclusion. What we can tell you is that when released, the RTX 2070 (not awesome) was 17% slower than the 1080 Ti. We also know that the RTX 2070 is 12% slower on average than the 2070 Super, so the fact that the 1080 Ti is still 3% faster on average, seems about right.
Should you buy the 1080 Ti now? Only if the price is right. If we look at the completed deals on eBay, we see that GTX 1080 Ti are currently sold for around $ 500. The cheapest example costs around $ 450. We honestly expected them to be much cheaper.
You can buy a brand new RTX 2070 Super for $ 510. A 1080 Ti saves ~ $ 60 on a used card that has likely been a little misused and doesn't seem like a wise investment. In addition, you can currently buy brand new 5700 XTs for around $ 425. Given that the 1080 Ti was only 9% faster on average, it's not worth paying a little more for a used model. Based on today's results, we would not pay more than $ 400 for a used GTX 1080 Ti. In fact, they would only tempt us around the $ 350 mark.