Our initial review of AMD’s brand-new Radeon VII graphics card relied on a punishing combo of a 4K resolution and high-dynamic-range output to bring our field of graphics cards to its knees, and folks, let me tell you: It is a glorious thing to experience smooth 4K gameplay with HDR on an OLED TV over a few days’ worth of gaming with the latest titles.
Our 4K results offer a robust idea of the relative performance offered by today’s graphics cards for the dollar, but I understand that some gamers really want cold, hard numbers for how many FPS they can expect from a given card and title at their preferred resolution. I’ll also concede that I do most of my gaming on 2560×1440, 144-Hz screens with variable-refresh-rate guts to begin with. While those panels might not provide the in-your-face color and brightness of an OLED TV, few things can, and I’ll just as happily take my gaming in high-refresh-rate, tear-free flavor as I will at a higher resolution and lower frame rates.
Certainly the most-requested missing piece from our Radeon VII review was 2560×1440 results, and it just so happened that we had a slightly older data set full of 2560×1440 captures from our recent review of Asus’ Strix RTX 2070 (which is well worth reading in its own right if you’re a 2560×1440 gamer). I don’t usually like commingling data sets produced with older and newer drivers, but given the distinct leap in performance from RTX 2070-class graphics cards to the RTX 2080 and Radeon VII, the single-digit performance differences afforded by most graphics driver updates probably won’t mess up our relative standings too much. The competition between the Radeon VII and the RTX 2080 is a lot closer on paper, though, so I retested the RTX 2080 with Nvidia’s latest drivers to keep things fair. Let’s keep this short and sweet.
Our testing methods
If you’re new to The Tech Report, we don’t benchmark games like most other sites on the web. Instead of throwing out a simple FPS average—a number that tells us only the broadest strokes of what it’s like to play a game on a particular graphics card—we go much deeper. We capture the amount of time it takes the graphics card to render each and every frame of animation before slicing and dicing those numbers with our own custom-built tools. We call this method Inside the Second, and we think it’s the industry standard for quantifying graphics performance. Accept no substitutes.
What’s more, we don’t typically rely on canned in-game benchmarks—routines that may not be representative of performance in actual gameplay—to gather our test data. Instead of clicking a button and getting a potentially misleading result from those pre-baked benches, we go through the laborious work of seeking out test scenarios that are typical of what one might actually encounter in a game. Thanks to our use of manual data-collection tools, we can go pretty much anywhere and test pretty much anything we want in a given title.
Most of the frame-time data you’ll see on the following pages were captured with OCAT, a software utility that uses data from the Event Timers for Windows API to tell us when critical events happen in the graphics pipeline. We perform each test run at least three times and take the median of those runs where applicable to arrive at a final result. Where OCAT didn’t suit our needs, we relied on the PresentMon utility.
As ever, we did our best to deliver clean benchmark numbers. Our test system was configured like so:
Processor | Intel Core i9-9900K |
Motherboard | MSI Z370 Gaming Pro Carbon |
Chipset | Intel Z370 |
Memory size | 16 GB (2x 8 GB) |
Memory type | G.Skill Flare X DDR4-3200 |
Memory timings | 14-14-14-34 2T |
Storage | Samsung 960 Pro 512 GB NVMe SSD (OS) Corsair Force LE 960 GB SATA SSD (games) |
Power supply | Seasonic Prime Platinum 1000 W |
Operating system | Windows 10 Pro version 1809 |
Graphics card | Boost clock (specified) |
Graphics driver version |
EVGA GeForce GTX 1070 SC2 Gaming | 1784 MHz | GeForce Game Ready 417.35 |
Nvidia GeForce GTX 1070 Ti Founders Edition | 1683 MHz | |
Nvidia GeForce GTX 1080 Founders Edition | 1733 MHz | |
Nvidia GeForce GTX 1080 Ti Founders Edition | 1582 MHz | |
Gigabyte GeForce RTX 2070 Gaming OC 8G | 1725 MHz | |
Asus ROG Strix GeForce RTX 2070 O8G Gaming | 1815 MHz | |
Nvidia GeForce RTX 2080 Founders Edition | 1800 MHz | GeForce Game Ready 418.81 |
AMD Radeon RX Vega 56 | 1471 MHz | Radeon Software Adrenalin 2019 Edition 19.1.1 |
AMD Radeon RX Vega 64 | 1546 MHz | |
AMD Radeon VII | 1750 MHz | Radeon Software Adrenalin 2019 Edition 19.2.1 for AMD Radeon VII |
Unless otherwise specified, image quality settings for the graphics cards were left at the control panel defaults. Vertical refresh sync (vsync) was disabled for all tests. We tested each graphics card at a resolution of 2560×1440 and 144 Hz, unless otherwise noted.
The tests and methods we employ are generally publicly available and reproducible. If you have questions about our methods, hit our forums to talk with us about them.
Monster Hunter: World
Monster Hunter: World preyed upon the Radeon VII at 4K, but dropping the resolution back to 2560×1440 allows the Radeon to go from borderline unplayable frame rates to a perfectly enjoyable experience. The Radeon VII’s frame rates come in just short of 60 FPS on average, and frame times remain consistently low for AMD’s 7 nm baby across the board. The RTX 2080 comes a lot closer to delivering a near-perfect 60-FPS experience at 2560×1440, though, as evidenced not only by its average frame rate but also its 99th-percentile frame time.
These “time spent beyond X” graphs are meant to show “badness,” those instances where animation may be less than fluid—or at least less than perfect. The formulas behind these graphs add up the amount of time our graphics card spends beyond certain frame-time thresholds, each with an important implication for gaming smoothness. To fully appreciate this data, recall that our graphics card tests all consist of one-minute test runs and that 1000 ms equals one second.
The 50-ms threshold is the most notable one, since it corresponds to a 20-FPS average. We figure if you’re not rendering any faster than 20 FPS, even for a moment, then you’re likely to perceive a slowdown. 33 ms correlates to 30 FPS, or a 30-Hz refresh rate. Go lower than that with vsync on, and you’re into the bad voodoo of quantization slowdowns. Also, 16.7 ms correlates to 60 FPS, that golden mark that we’d like to achieve (or surpass) for each and every frame.
In less demanding or better-optimized titles, it’s useful to look at our strictest graphs: 8.3 ms corresponds to 120 FPS, the lower end of what we’d consider a high-refresh-rate monitor. We’ve recently begun including an even more demanding 6.94-ms mark that corresponds to the 144-Hz maximum rate typical of today’s high-refresh-rate gaming displays.
Our time-spent-beyond graphs let us put a point on just how close the RTX 2080 comes to delivering a perfectly smooth ride in this title. The RTX 2080 spends under two-tenths of a second on frames that take longer than 16.7 ms to render, while the Radeon VII spends nearly six seconds on such frames. Dropping back to 2560×1440 certainly renders Monster Hunter: World quite playable on the Radeon VII, and indeed, it’s the best AMD card around for playing this title at that resolution. Still, the Radeon VII isn’t beating out the GTX 1080 Ti for delivered smoothness in this title, much less the RTX 2080.
Hitman 2
Hitman 2 may not have the DirectX 12 rendering path of its predecessor, but it can still put the hurt on any modern graphics card. We cranked image quality settings to the max to fully flesh out the game’s world of assassination.
Despite the lack of a DirectX 12 rendering path in this title, the Radeon VII tracks the RTX 2080 well once we take HDR and 4K out of the picture. The AMD card’s average frame rate only slightly tails that of the RTX 2080, and the two cards’ 99th-percentile frame times are nearly identical.
Neither the Radeon VII nor the RTX 2080 put any time up on the time-spent-beyond-16.7 ms board, so we have to flip over to our 11.1 ms and 8.3 ms thresholds to put any light between them. Even then, the Radeon VII spends just a hair longer on frames that take longer than 11.1 ms to finish than the RTX 2080 does, and just about a second longer than its Turing rival to wrap up frames that need more than 8.3 ms to come out of the oven. Either of these cards will make high-refresh-rate 2560×1440 gamers happy.
Far Cry 5
Despite flying Ryzen and Radeon colors in its splash screen, Far Cry 5 doesn’t hand a total victory to the red team. The differences in average frame rates between the GTX 1080 Ti, Radeon VII, and RTX 2080 are going to be near-invisible in this title, though, and the Radeon VII impressively turns in the best 99th-percentile frame time of the bunch.
At 16.7 ms and 11.1 ms, the Radeon VII and RTX 2080 put only the barest slivers of time on the board thanks to a few intermittent frame-time spikes. To really tease out differences between these cards, we have to check out the 8.3 ms threshold. There, the Radeon VII turns in a slight lead over the GTX 1080 Ti and spends only two-tenths of a second longer past 8.3 ms than the RTX 2080 does. We’ll call it a wash.
Forza Horizon 4
Forza Horizon 4 drops drivers into a lovingly rendered rendition of the English countryside. Consequently, it can be rather demanding on graphics hardware if you put the quality-settings pedal to the metal. That’s just what we did to see how our graphics cards perform in this title.
We know from our 4K testing that using MSAA on the Radeon VII can put a greater strain on its resources than the pixel-shader-powered FXAA does. Any hope that dialing back the resolution might have held in this title for the Radeon VII gets dashed by our average-FPS and 99th-percentile frame time results, though. It seems like we’re still running into some kind of ROP bottleneck.
The Radeon VII’s struggles with MSAA in Forza Horizon 4 become painfully evident in our time-spent-beyond-11.1 ms graph. The Radeon VII spends nearly six and a half seconds longer than the RTX 2080 does on frames that take longer than 11.1 ms to finish, and that results in noticeably rougher gameplay.
Since we know MSAA is a punishing technique for removing jaggies on the Radeon VII, I also did a quick test with FXAA enabled to see whether it turns any tables.
Choosing FXAA as our anti-aliasing method doesn’t change the standings between the RTX 2080 and the Radeon VII, but it does make life easier for both cards. Neither pixel-pusher puts up more than a second on tough frames that need longer than 11.1 ms to finish. Still, the Radeon VII can’t unseat the RTX 2080 even with its massive shader array doing the AA work, and at the 8.3 ms mark, the GeForce card spends about four fewer seconds of the 84-second built-in benchmark in this title chewing on tough frames that need longer than 8.3 ms to finish. That’s an improvement one can feel.
Assassin’s Creed Odyssey
The Radeon VII can pull alongside the GTX 1080 Ti in Assassin’s Creed Odyssey, another title that flies Ryzen and Radeon colors on its title screen. Our plot of 99th-percentile frame times puts the Vega 20 card closer to midpack for delivered gaming smoothness, though. The RTX 2080 remains our overall winner.
The Radeon VII’s spikiness in our frame-time graphs doesn’t cause it to put any time on the board at our most concerning thresholds for delivered smoothness, but it ultimately loses out to the RTX 2080 by spending just short of two seconds longer putting the finishing touches on frames that need longer than 16.7 ms to render. Even the two-year-old GTX 1080 Ti provides a smoother gaming experience by this measure.
Gears of War 4
I had high hopes for the Radeon VII in Gears of War 4 at 2560×1440 thanks to its DirectX 12 API and console roots. Even so, for some reason this Unreal Engine game really, really seems to like the Turing architecture at its most punishing settings. The RTX 2080 opens a wide lead on its Radeon competitor here, and it cuts 2 ms off the Radeon VII’s 99th-percentile frame time. That’s impressive delivered smoothness from the green team in this title.
The Radeon VII puts only half a second on the board for frames that take longer than 16.7 ms to finish, but a flip over to the 11.1 ms threshold puts a big exclamation point on Turing’s superiority in this title. The RTX 2080 cuts 6.4 seconds off the time that the Radeon VII puts on the board here, and again, that’s an improvement in gaming smoothness you can feel.
Battlefield V
On top of being the marquee title for Nvidia’s RTX effects so far, Battlefield V boasts a cutting-edge DirectX 12 renderer as part of EA’s Frostbite engine.
At 2560×1440, Battlefield V remains the sole win for the Radeon VII, and only then by a hair. Still, the Radeon outpaces the GeForce in both average FPS and delivered smoothness. Let’s see just how meaningful that result is with our time-spent-beyond graphs.
We have to click all the way over to the 8.3 ms threshold to see any time on the board from the Radeon VII or RTX 2080, and here, the Radeon VII shaves two seconds off of the RTX 2080’s time spent finishing tough frames. That’s a much-needed win for the Vega 20 card.
Conclusions
As always, we’ve created plots of price versus performance by taking the geometric mean of each card’s average FPS and 99th-percentile frame times in all of the games we tested and plotting that value against each card’s suggested price (if stock is no longer available) or against its retail price on Newegg (if stock is available). To make our higher-is-better approach work, we’ve converted the geometric mean of 99th-percentile frame times into frames per second.
I hate to say I told you so, but it’s true: Our relative standings in 4K testing do generally hold when we lower resolution and remove HDR from the picture. The GeForce RTX 2080 Founders Edition still outpaces the Radeon VII at 2560×1440 and offers a smoother, quieter, less power-hungry gaming experience while doing it. Our results with Nvidia’s Founders Edition card might not make the RTX 2080 look like the best value, but custom-cooled 2080s are widely available for just a few bucks over Nvidia’s $699.99 suggested price for partner cards. The Gigabyte RTX 2080 Gaming OC 8G we have in the TR labs is both faster and quieter than the already fast and quiet Founders Edition. Given that the Gigabyte card is just $10 more, that’s bad news for the Radeon VII.
In turn, nothing I wrote in my original conclusion regarding the Radeon VII changes with a different resolution in play. AMD’s 7-nm baby may be a superior sub-$1000 card for those who need a broad range of computing power or large pools of VRAM that you don’t get on consumer Nvidia cards, but gamers will be happier with Turing options at the moment. That’s especially true at 2560×1440, where the Radeon VII’s monster pool of RAM likely won’t get exercised as hard as it is at 4K by games alone. Unless your demands for graphics cards extend beyond pure pixel-pushing, the RTX 2080 remains today’s best dollar-for-dollar choice for fluid and smooth gameplay at 2560×1440.