Ever since AMD refreshed most of its Ryzen product range last year using GlobalFoundries’ 12LP process—an enhanced version of the foundry’s 14 nm FinFET node with higher-performance front-end-of-line transistors but only optional density improvements—and expanded that refresh to Radeons with the RX 590, I sort of expected the company to do the same with its Vega 10 graphics chip at some point. Instead, AMD surprised us at CES by introducing a consumer version of its Vega 20 data-center GPU on board a new graphics card: the Radeon VII. Taking Vega 20 out of the data center means AMD is first to market with a gaming GPU fabricated on TSMC’s cutting-edge 7 nm FinFET process.
Vega 20 initially made an appearance in the Radeon Instinct MI50 and MI60 accelerators introduced late last year. Instead of making the biggest chip possible as quickly as possible like Nvidia has with its compute accelerator chips of late, Vega 20 only modestly expands the processing capabilities of the Vega 10 GPU to support some reduced-precision data types useful in deep learning applications. TSMC’s 7-nm process is a real node shrink with all of the areal scaling such an advance implies, and it does most of the work in delivering a generation-on-generation performance improvement.
Vega 20 takes most of the existing Vega 10 GPU and shrinks it to 331 mm², down from 495 mm² for Vega 10. By shrinking the die in this way, AMD made room on Vega 20’s underlying silicon interposer to add on two more stacks of HBM2 RAM. Vega 20 now talks to its on-package memory over a 4096-bit wide bus, and with HBM2 speeds of 2 Gb/s per pin, that works out to a whopping 1 TB/s of theoretical memory bandwidth. Compare that to the 484 GB/s that Vega 10 enjoyed on board the RX Vega 64, and we should expect a significant performance increase from this move alone.
TSMC has claimed that its first generation of 7 nm transistors can deliver up to 20% higher performance or 40% power savings at the same performance versus its 16 nm node. AMD pushed the power-savings-versus-performance-increase slider all the way to the right for the Radeon VII. The company has already said that this card would deliver its improved performance at the same power as the RX Vega 64.
As a result, the Radeon VII’s board power rings in at 300 W, up 5 W over even the notoriously power-hungry RX Vega 64. In return, AMD clocks the Radeon VII’s Vega 20 chip at a 1750 MHz typical boost speed, up about 13% over the Radeon RX Vega 64 in its air-cooled configuration. As we’ll see from some back-of-the-napkin calculations to come, that clock-speed increase more than makes up for the fact that AMD has had to disable four of Vega 20’s 64 compute units, presumably for yield reasons. Along with those compute units, the Radeon VII appears to lose 16 of its texture units. We’re confirming that figure with AMD and will update our numbers as soon as we can.
In its descent from the data center, the Vega 20 GPU loses some features that will be of little use to enthusiasts. For example, Radeon Instinct accelerators using Vega 20 support both PCI Express 4.0 and coherent card-to-card communications using the Infinity Fabric interconnect, but the Radeon VII sticks to the far more common PCIe 3.0 standard and drops the data-center class inter-card connection. Vega 20 also offers support for FP64 (aka double-precision) data types at up to 1/2 the FP32 rate, and unusually, some of that FP64 processing capability survives on the Radeon VII in a form that may be useful to folks who need it.
This morning, AMD told Anandtech that it’s ultimately decided to enable 1/4-rate FP64 on the Radeon VII after earlier statements suggesting that 1/8 would be the rate of choice. Some more back-of-the-napkin math suggests that in its final form, the Radeon VII could crunch through FP64 workloads at 3.35 TFLOPS. That’s quite a bit better than the 1.6-TFLOP peak rate that the Kepler-powered GTX Titan offered back in the day. GTX Titan cards have endured as relatively cheap founts of FP64 performance for those who need it, so AMD might sell some Radeon VIIs to scientific-computing folk who want useful FP64 performance alongside an otherwise much more modern graphics card.
Here’s a table comparing the Radeon VII’s basic specs to several of today’s most common pixel-pushers:
|RX Vega 56||1471||64||224/112||3584||2048||410 GB/s||8 GB|
|GTX 1070||1683||64||108/108||1920||256||259 GB/s||8 GB|
|RTX 2060 FE||1680||48||120/120||1920||192||336 GB/s||6 GB|
|RTX 2070 FE||1710||64||120/120||2304||256||448 GB/s||8 GB|
|GTX 1080||1733||64||160/160||2560||256||320 GB/s||8 GB|
|RX Vega 64||1546||64||256/128||4096||2048||484 GB/s||8 GB|
|Radeon VII||1750||64||240/120?||3840||4096||1 TB/s||16 GB|
|RTX 2080 FE||1800||64||184/184||2944||256||448 GB/s||8 GB|
|GTX 1080 Ti||1582||88||224/224?||3584||352||484 GB/s||11 GB|
|RTX 2080 Ti FE||1635||88||272/272||4352||352||616 GB/s||11 GB|
|Titan Xp||1582||96||240/240||3840||384||547 GB/s||12 GB|
|Titan V||1455||96||320/320||5120||3072||653 GB/s||12 GB|
From this chart, a couple things should stand out. One is that the Radeon VII offers by far the highest theoretical memory bandwidth of any card we’ve ever had our hands on. At $699, the Radeon VII is also the cheapest way to get 16 GB of RAM on a graphics card yet. We typically don’t see that much VRAM until one gets into pro visualization products, but with the advent of the Radeon VII, AMD is pushing 16 GB as a new enthusiast standard. Since memory capacity is one of the Radeon VII’s clearest spec wins over the RTX 2080 and its 8 GB of RAM, it’s no surprise that AMD is making hay of this feature. The company suggests a number of current and upcoming games can benefit from having more than 8 GB of video memory available for gaming at 4K and maximum settings, and it also highlighted the value of that memory for prosumer workloads like 4K and 8K video processing with Adobe Premiere.
On a gut level, this push for higher memory capacities on cards destined for 4K gaming makes sense. We know that texture sizes are growing and seem likely to grow further as gamers demand higher and higher quality visuals, and I know a couple video producers who might appreciate having more video memory available without paying through the nose for it. AMD says Far Cry 5 can occupy as much as 12.9 GB of VRAM at 4K and max settings, and the company provided us a scary-looking frame time graph that suggests major inconsistency could arise if an RTX 2080 runs over its available memory pool. Far Cry 5 is one of the games we use in our test suite, so that’s an easy point to return to for further analysis.
|RX Vega 56||94||330/165||5.9||10.5|
|RTX 2060 FE||81||202/202||5.0||6.5|
|RTX 2070 FE||109||246/246||5.1||7.9|
|RX Vega 64||99||396/198||6.2||12.7|
|GTX 1080 Ti||139||354/354||9.5||11.3|
|RTX 2080 Ti||144||473/473||9.8||14.2|
As a hotter-clocked Vega chip, the Radeon VII offers modest but meaningful theoretical performance improvements versus its RX Vega 64 predecessor in every one of our usual theoretical peak figures. Really, there are no surprises here. The basic resource allocation of Vega 20 didn’t change as part of the move to the 7 nm node, and AMD’s widening of the chip’s memory bus largely overshadows the clock-speed-related gains in the table above. Graphics tends to be a memory-bandwidth-bound workload, and making sure the beast is fed is unsurprisingly a priority for GPU architects. Perhaps the addition of that wider memory bus speaks to what sat atop the list of fixes AMD envisioned for Vega after its first time out on shipping products.
The Radeon VII card itself should be a well-known quantity by this point. Yours truly wasn’t on the ground at CES, but the card was, and the press enjoyed extensive access to it during the show. The Radeon VII takes the basic bead-blasted, diamond-cut aluminum shroud design introduced on limited-edition Radeon RX Vega 64 cards and turns it into a home for three fans blowing onto an open-style heatsink more typical of custom cooler designs from board partners.
We’re hesitant to tear down and repaste coolers that need to go back on large GPU packages with HBM RAM like Vega 20, but the fine folks at Gamers Nexus have no such qualms, and they’ve already torn down their card ahead of today’s embargo lift. From GN’s work, we know that the Radeon VII cooler uses a massive, complicated copper vapor chamber to wick heat away from the Vega 20 package and into an aluminum fin stack. Perhaps thanks to that vapor chamber, AMD is able to keep the Radeon VII reference design within the boundaries of two slots, although the cooler shroud does jut slightly beyond the edge of the expansion-card bracket. The 300 W board power we alluded to earlier manifests itself in this card’s pair of eight-pin PCIe plugs.
It’s interesting to see AMD follow in Nvidia’s Founders Edition footsteps for Turing by introducing an open-air cooler on the Radeon VII despite its much higher board power than any GeForce RTX card. The RX Vega 64’s blower-style heatsink may not have been the best thing going for absolute noise levels, but it did dump the copious heat that card produced out the rear of a PC. Systems with a Radeon VII inside, on the other hand, will need to be well-ventilated to remove the hot air exhausted by its open-style cooler, although that demand is no different than the challenges posed by Nvidia’s latest reference coolers.
AMD stickers the Radeon VII at $699, or the same suggested price as GeForce RTX 2080 partner cards. The company has drawn a bead on Nvidia’s best-performing Turing card under $1000 over and over in the run-up to this morning. Let’s see if AMD hit its target now.
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|
|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|
|OS||Windows 10 Pro version 1809|
Thanks to Intel, Corsair, G.Skill, and MSI for helping to outfit our test rigs with some of the finest hardware available. AMD, Nvidia, and Gigabyte supplied the graphics cards we used for testing today, as well.
|Graphics card||Boost clock
|Graphics driver version|
|Nvidia GeForce GTX 1080 Ti Founders Edition||1582 MHz||GeForce Game Ready 418.81|
|Gigabyte GeForce RTX 2070 Gaming OC 8G||1725 MHz|
|Nvidia GeForce RTX 2080 Founders Edition||1800 MHz|
|Nvidia GeForce RTX 2080 Ti Founders Editoin||1635 MHz|
|AMD Radeon RX Vega 64||1546 MHz||Radeon Software Adrenalin 2019 Edition Press|
|AMD Radeon VII||1750 MHz|
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 3840×2160, unless otherwise noted. We enabled HDR in games where it was available. Our HDR display is an LG OLED55B7A television.
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.
Hitman 2‘s world of assassination comes with stunning lighting effects and lots of scene complexity, and it’s drop-dead gorgeous in 4K with HDR. All that makes for a tough time getting to 60 FPS for most of our graphics cards. The Radeon VII gets off to a strong enough start by matching the GTX 1080 Ti nearly frame for frame, but the RTX 2080 has an undeniable advantage in both average frame rates and 99th-percentile frame times.
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 each 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 generally 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.
At the 16.7 ms threshold, the Radeon VII proves the equal of the GeForce GTX 1080 Ti. Both cards spend about six seconds of our one-minute test run rendering frames that take longer than 16.7 ms to finish. The RTX 2080 impresses by spending less than one-third the time completing such frames, though.
Forza Horizon 4 with MSAA
Imagine MSAA at 8X here.
Forza Horizon 4 hit the road as a showcase for Radeon graphics performance, but after breakout results for the red team early on, that lead has dwindled to performance parity with Nvidia cards. Problem is, our pedal-to-the-metal settings with 4K and HDR in this title resulted in a substantial—and atypical—GeForce lead. Even after moving the Radeon VII to several different test systems and trying any number of platform-related fixes, we couldn’t get our delivered results to mesh with the expectation of performance parity for Radeons in this title.
After bringing up this issue with AMD, the company advised us that using 8X MSAA in Forza Horizon 4 will reduce performance on its products more than it will on GeForces. That suggests the 64-ROP complement that’s graced high-end Radeons since the Hawaii days is perhaps no longer enough to handle MSAA at 4K in this title.
My understanding is that MSAA delivers a one-two punch because it both relies on fixed-function hardware and can be quite memory-bandwidth intensive, even with modern color compression techniques. It appears that the Radeon VII’s terabyte per second of theoretical bandwidth can’t overcome bottlenecks elsewhere in the GPU.
Presuming we are looking at a ROP bottleneck with these settings, that resource constraint hits our Radeons hard. Both the RTX 2080 and the GTX 1080 Ti spend less than half the time past 16.7 ms on tough frames that the Radeon VII needs here, and the Vega 64 completely runs out of gas.
As part of our conversation with AMD, the company suggested that FXAA was the optimal technique for achieving the best balance between image quality and performance on its cards in this title. We didn’t want to discard our MSAA test results, as it should come as no surprise that using less-expensive computational techniques can result in increased performance. Still, we figured we at least ought to give that alternate technique a shot to see whether it resulted in noticeably lower image quality in exchange for performance.
Forza Horizon 4 with FXAA
Flipping off MSAA puts a better face on performance for all of our graphics cards, and it may well be the optimal way to play this game with minimal jaggies. Forza Horizon 4’s FXAA implementation is quite good, and even my jaded graphics reviewer’s eye has to admit that there’s precious little difference in image quality between running 8X MSAA and just using FXAA in this title.
Since FXAA is implemented as a pixel shader rather than a technique that employs fixed-function hardware, all of our cards get a nice speedup, and the Radeons can put their prodigious shader resources to use rather than getting bottlenecked by ROP throughput. The net result is that the Radeon VII pulls even with the GTX 1080 Ti on average FPS and 99th-percentile frame times, but the RTX 2080 is just that little bit better yet.
None of our cards need longer than 50 ms or 33.3 ms to render a frame, and both the GTX 1080 Ti and the Radeon VII spend just under a second on tough frames that need longer than 16.7 ms to finish. The RTX 2080 proves even better still, spending only a vanishing fraction of a second on such frames. Still, any of our $699-and-up cards would make a fine choice for 4K HDR gaming with Forza Horizon 4 using FXAA.
Assassin’s Creed Odyssey
To test Assassin’s Creed Odyssey, we cued up the Very High preset at 4K with HDR enabled.
Our 4K HDR test of Assassin’s Creed Odyssey reveals stormy seas for the Radeon camp. We’ve been aware that enabling HDR on Radeon cards can cause worse frame-time spikes than usual in the latest Assassin’s Creed titles ever since we tested 4K HDR performance in Assassin’s Creed Origins for our RTX 2080 review, but we didn’t really dig into it at the time.
Whatever the cause for those spikes, it appears to have persisted with the release of Assassin’s Creed Odyssey. The Radeon VII’s average frame rate lands between that of the GTX 1080 Ti and the RTX 2080, but its 99th-percentile frame time trails even the RTX 2070. That’s not good news for gaming smoothness.
Our time-spent-beyond-X graphs show that the worst of the Radeon VII’s spikes cause it to put up some time at the 33.3-ms mark, and those are hitches you can feel as you traverse Odyssey‘s rendition of ancient Greece. At the 16.7-ms mark, though, the Radeon VII pulls ahead of the GTX 1080 Ti, even as both cards spend about one-fourth of our one-minute test run finishing frames that take longer than 16.7 ms to render. If AMD can fix those frame-time spikes, it’ll make the Grecian roads a lot less bumpy.
Far Cry 5
As we did for Assassin’s Creed Odyssey, we tested Far Cry 5 at 4K with HDR enabled.
Far Cry 5 is another title that has a rough relationship with Radeons when HDR is switched on, and the Radeon VII’s performance in rural Montana echoes its time in ancient Greece. AMD’s 7 nm darling nearly catches the RTX 2080 on average frame rates, but its 99th-percentile frame time comes in way behind that of its Turing rival.
It’s a shame that those hitches spoil the Radeon VII’s delivered smoothness by putting time on the board at the 50 ms and 33.3 ms marks, because the card’s performance for 4K gaming is otherwise quite capable. The Radeon VII shaves nearly a second and a half off the GTX 1080 Ti’s time spent finishing fussy frames at 16.7 ms, even as the RTX 2080 shaves yet another second off that result yet. Again, if AMD can fix those intermittent spikes, the Radeon VII will be quite the competitor for Nvidia in this title.
Monster Hunter: World
Monster Hunter World once again proves that it’s a monster of a game to run well as we dial in a 4K resolution and demanding settings, including HDR. The Radeon VII can’t even outpace the RTX 2070, let alone the GTX 1080 Ti or RTX 2080. At least AMD’s latest and greatest keeps 99th-percentile frame times under control.
Given the low frame rates from the Radeon camp, it’s most relevant to start our time-spent-beyond analysis at the 33.3 ms threshold. Despite its relatively low average frame rate in this title, the Radeon VII doesn’t spend more than a few milliseconds on frames that take longer than 33.3 ms to finish. If 30 FPS gaming at 4K in this title is your goal, the Radeon VII will at least not pummel you with frame-time spikes that lead to an unplayable experience. Still, our 16.7 ms graph shows that the RTX 2080 proves a far more effective tool for getting the job done in Monster Hunter: World at 4K.
Chalk up a much-needed win for the Radeon VII as we close out our gaming tests. AMD’s 7 nm baby turns in a sterling 60 FPS average and a 99th-percentile frame time that suggests only minor roughness in Battlefield V at 4K and with HDR enabled. Meanwhile, our Turing GeForces all appear to struggle with a stuttering issue that pushes up 99th-percentile frame times.
A peek at our 50 ms and 33.3 ms graphs shows that the spikes GeForces are experiencing aren’t the kinds of punch-you-in-the-face fuzziness we abhor the most, but they do leave their marks at the 16.7 ms threshold. Here, the Radeon VII spends a little under two seconds of our one-minute test run on frames that take longer than 16.7 ms to finish, while the RTX 2080 puts up more than double the time. Even without the spikes of its Turing successors, the GTX 1080 Ti spends more than three times as long wrapping up such frames. If you don’t have any interest in Battlefield V‘s DirectX Raytracing effects, the Radeon VII proves unusually capable in this title.
A quick look at noise and power consumption
The most nagging question I’ve had since CES was just how efficient the Radeon VII would be thanks to its underlying TSMC 7 nm FinFET process. Back then, AMD CEO Lisa Su said the Radeon VII would deliver 25% higher performance for the same power as its RX Vega 64 predecessor. To find out whether that’s the case, we turned to the 2016 reboot of Doom and kept an eye on our Watts Up power meter to monitor system power draw from the wall.
There’s a shocker. Despite being rated for a similar board power to that of the Radeon RX Vega 64, the Radeon VII actually lets our test system consume quite a bit less power while delivering much higher performance. We wouldn’t be surprised if the actual improvement in performance per watt for the Radeon VII greatly exceeds that 25% figure.
Even though the Radeon VII marks a much-needed improvement in performance per watt for the red team, the net effect of that shrink has been to bring the Radeon VII’s approximate performance per watt on par with that of the GTX 1080 Ti, whose GP102 GPU was fabricated on TSMC’s 16 nm FinFET process—in other words, a whole node behind 7 nm FinFET.
Going by these results, I’d venture that AMD’s graphics processors badly need a Maxwell moment (or a moment of Zen, if you prefer) that puts the seven-year-old basic design of the Graphics Core Next microarchitecture to rest. GCN has evolved plenty over time, to be sure, but Vega 20 shows that the fruit of that evolution is far off the pace AMD needs to compete with Nvidia on performance per watt.
As we first saw with AMD’s Polaris products, the red team has let process advances take the lead in delivering both performance and performance-per-watt improvements of late, while the stagnation at 28 nm apparently spurred Nvidia to find ways to improve its products’ performance per watt independent of process gains—an investment that has paid off handsomely over time. The Turing architecture appears to have a virtual node and change worth of advantage on AMD in the performance-per-watt race, on top of whatever benefits Nvidia’s chips may yet enjoy from migration to a next-generation process. Scary stuff.
It’s not all roses for the Radeon VII once we pull out the decibel meter, either. Despite the fact that this card draws less power than the RX Vega 64, and even with its elaborate copper vapor chamber and full-length fin stack, the Radeon VII reference design has to spin its trio of fans at high RPMs to move heat away from the card. In turn, the Radeon VII is actually louder at full tilt than the already noisy RX Vega 64 reference design.
Meanwhile, the RTX 2080 Founders Edition comes in more than 10 dBA quieter than the Radeon VII Since dBA is a logarithmic scale, a rough rule of thumb is that such a difference is enough to produce a halving of perceived noise levels. Sure, one can don headphones or crank the volume knob to drown out the racket the Radeon VII makes, but you will notice this card’s din in quiet parts of games where low noise floors might matter, and its whine may annoy other people in a room with you.
Given this card’s price tag, it’s hard not to be disappointed with how loudly it runs. The Radeon R9 Fury X managed to incorporate a whisper-quiet liquid cooler for $50 less than the Radeon VII commands, and the liquid-cooled edition of the RX Vega 64 rang in at the same $699.99 price tag as this card. Perhaps AMD can fine-tune the Radeon VII’s fan curve in a future driver update to better balance noise levels and delivered performance, but for now, we think its noise character is unbecoming of a premium product.
Let’s sum up the hundreds of graphs on the preceding pages with our famous value scatter plots. To create these charts, we take the geometric mean of both average frame rates and 99th-percentile frame times across all of our tests, then convert that 99th-percentile number into FPS to make our higher-is-better logic work. We’ve included graphs that incorporate data from Forza Horizon 4 with both MSAA and FXAA so that you can verify how our use of a geometric mean prevents a stumble in one game from affecting our overall results too much.
For pricing data, we sought out e-tail listings for cards still being sold. For cards that are no longer available, we used the manufacturer’s last known suggested price.
Stop us if you’ve heard this one before. On the whole, the Radeon VII’s performance potential (as measured by average frame rates) lets it slightly outpace the venerable GeForce GTX 1080 Ti, making it the fastest single-GPU Radeon card we’ve ever tested by that metric. Our final reckoning puts the Radeon VII just 8% behind the RTX 2080 on that measure, too. That’s not bad, considering that the RTX 2080 Founders Edition we tested is a more expensive and hotter-clocked card than a reference-spec RTX 2080 may be.
Going by our 99th-percentile FPS metric for delivered smoothness, however, the GTX 1080 Ti and RTX 2080 both outstrip the Radeon VII. Right now, even Nvidia’s $500-ish RTX 2070 will generally offer you a 4K HDR gaming experience as smooth as what AMD’s $699 fighter can deliver, and that’s with our most favorable test case for Forza Horizon 4 rolled in. If you prefer MSAA to FXAA in that title, the ride on Radeons gets even rougher, and our conversations with the company suggest that’s because of a ROP bottleneck that’s not going away.
Even with its best foot forward in Forza Horizon 4, it’s disappointing to see a product as expensive and as critical as the Radeon VII go down such a bumpy road at launch. We’ve been banging this drum for seven years now. AMD has proven that it can iron out those wrinkles with driver updates and time, and we don’t think a 15% improvement in 99th-percentile frame rates is an insurmountably high bar to clear if the goal is to catch the RTX 2080. Such a figure might be a deterrent to dropping $700 on one of these cards until AMD’s drivers shape up, though.
On performance per watt, the Radeon VII marks an important advance for AMD. It consumes less power than the Radeon RX Vega 64 to deliver much higher performance. Even so, it needs quite a bit more juice to do its thing than both the RTX 2080 and GTX 1080 Ti. AMD’s impressive-looking triple-fan cooler comes tuned for maximum performance, too, so it’s even slightly louder in operation than the notoriously noisy blower on the RX Vega 64. Both Nvidia’s reference blower for Pascal and its biggest dual-fan Founders Edition cooler for Turing will keep a room’s noise floor noticeably lower than that of the Radeon VII’s heatsink.
Perhaps because of its data-center DNA, the Radeon VII can’t even move the stubbornly stationary performance-per-dollar bar. This card delivers the same bang for the buck that the GTX 1080 Ti did at launch, and aside from 16 GB of RAM that’s of unclear value to gamers today, it lacks any forward-looking features like the nascent real-time ray tracing and DLSS functions of Nvidia’s Turing architecture for curious folks to kick the tires on. With four stacks of costly HBM2 RAM and a cutting-edge lithography process at its heart, we doubt AMD could trim much fat from the Radeon VII’s price tag without whittling bone. If $699 is the price the Radeon VII has to command, though, it makes the card a tough sell for what it delivers.
For even money between the RTX 2080 and the Radeon VII, we’d put our bet on the green team for the moment. Perhaps thanks to the arrival of the Radeon VII, swift and whisper-quiet RTX 2080 partner cards are now selling for only small premiums over Nvidia’s $699.99 suggested price. You’ll enjoy faster and smoother gaming for your money than the Radeon VII can offer right now. We’ll need to defer final judgment on the value of the Radeon VII’s 16 GB of memory versus the RTX 2080’s 8 GB complement, but that deficit didn’t appear to cause issues for the Turing card even in Far Cry 5, a title that AMD highlighted as one of the worst memory hogs around for 4K gaming at max settings.
We imagine the Radeon VII might be the right card for some people. Perhaps your day-to-day work eats VRAM like there’s no tomorrow, and you only care about gaming on the side. Maybe you don’t care in the least about what you’ve seen of hybrid rendering with real-time ray tracing, and you passed up the GTX 1080 Ti at its zenith. Maybe you just can’t bear the thought of putting one red cent in Jensen Huang’s jacket fund. If any or all of those things describe you, the Radeon VII is as good as it gets for an alternative choice in high-end graphics right now. We just wish it was a smoother, quieter, and cheaper one.