Why you can trust Tom's Hardware
Our expert reviewers spend hours testing and comparing products and services so you can choose the best for you. Find out more about how we test.
Looking at the 1440p ultra results, we see the EVGA RTX 2080 Super XC Hybrid averaged 105.8 fps across all nine games. The air-cooled Asus ROG Strix 2080 Super OC averaged 105.9 fps while the RTX 2080 Super Founders Edition averaged 104.9 fps—a virtual tie in overall performance. Every game averages more than 60 fps, and several are over 100 fps, allowing for high refresh gaming at these settings.
AMD’s current flagship Radeon RX 5700 XT holds its own, managing 90 fps on average with the factory overclocked ASRock EX 5700 XT Taichi, and 86.8 fps with the reference 5700 XT. Both are well behind any 2080 Super, but they're nearly tied with AMD's slightly older Radeon VII. Obviously, AMD's RX 5700 series competes on price and not just raw performance, and until Big Navi arrives this is AMD's best solution.
We should also note that all GPUs are tested using DirectX 12 in Borderlands 3, The Division 2, Metro Exodus, and Shadow of the Tomb Raider—and the Vulkan API for Red Dead Redemption 2 and Strange Brigade. We mention this because in many games that have DirectX 11 as an option, they actually run slightly better on Nvidia cards in that mode. Borderlands 3 in particular is typically around 10% faster in DX11 on Nvidia GPUs at 1440p and 4K, but even Division 2 and Metro Exodus see a modest 3-5% uplift. And of course, for games like Metro Exodus and Shadow of the Tomb Raider, DX12 is required if you want to use ray tracing on an RTX card.
We felt it best to standardize on one API for each game, as it helps simplify testing. Otherwise, we'd need to check performance in every API offered for each game to determine the best choice for each GPU at each resolution and setting. That increases the number of test runs and thus testing time by 66%, plus we'd need a lengthy explanation saying which API was used on each GPU at each setting. Besides, long term we expect more games to begin using DX12 and/or Vulkan exclusively, especially once the next generation of consoles arrive with ray tracing hardware support.
I would've thought a liquid cooled gpu like that would've performed a fair deal better than an air cooled model like the ROG Strix...
Did they take a FE 2080 Super PCB and slap the AIO on it? The FE only has a 280w power limit, and Asus Strix has a 313w limit on the OC model.
It would be nice to see comparisons of the different hybrid cooled cards out there (EVGA's Hybrid, Gigabyte's Waterforce, Inno3D iChill, etc.). I use these for my iTX build, and apparently there are many differences in noise and cooling performance. And while there are some reviews online, it's really hard to come across comparisons.
Some of us like to use the high end cards for Folding. Most card reviewers don't give a single sentence in that area as to how various cards do in that arena. So, we have to search that info out in other areas. "It would be nice" to see a reference in video card reviews - even if it's only a sentence or two.
The EVGA RTX 2080 Super XC Hybrid performed well, averaging over 100 fps at 1440p ultra and 60 fps at 4K ultra. The hybrid cooling kept the card much cooler than other cards, but the pump has a soft hum. Priced at $789.99, it's one of the most expensive 2080 Super cards available, but only similar Hybrid setups can cool as well.
EVGA RTX 2080 Super XC Hybrid Review: Cool Running, but Worth It? : Read more
Why didn't you guys review the FTW3 version of this card which has a higher boost of 1845Mhz? I paid $769 for it too on Newegg back in September. It's actually going for around the same price right now.
Everything in your review is spot on. It rarely eclipses 53 degrees even when it's working it's hardest. It can overclock it's butt off but I put it back to stock as it's really not necessary since I'm gaming on the Dell S3220DGF monitor. The pump wine does sort of bother me but I've gotten used to it. I wish it could be controlled.
The one real issue with this is card is that the radiator fan can't be software controlled with Precision X1 or Afterburner. You can only independently control the GPU fan. All software reports them running at the same speeds which certainly isn't true.
Some of us like to use the high end cards for Folding. Most card reviewers don't give a single sentence in that area as to how various cards do in that arena. So, we have to search that info out in other areas. "It would be nice" to see a reference in video card reviews - even if it's only a sentence or two.
Are you aware of any objective way to test that? F@H doesn't have a benchmark tool (that I'm aware of) and of course they can't just sit there hoping they get a WU, wait for 2 hours, repeat to average results, and just cross their fingers that one WU will be the exact same as another from the same day let alone of months.
They can perhaps derive it from the other tests done but that specific one would have to be speculation.
