AMD Radeon VII review: Finally, a high-end card from AMD

AMD’s first enthusiast-grade card in a long while

At CES 2019, AMD unveiled the Radeon VII graphics card, a surprise move that many of us were not expecting. I had thought that the next time I saw a high-end card from AMD, it would be when the company announced its Navi architecture (which is still under wraps), but AMD is clearly not willing to stay quiet for so long.

The new Radeon VII is the world’s first consumer GPU manufactured on a 7nm process technology. It is based on AMD’s existing Vega architecture, and the company is positioning it as a card that can handle 4K gaming at the highest settings. That means that it should be going up against the GeForce RTX 2080, at the very least, an exciting prospect since it’s been a while since AMD released a card that was capable of competing with NVIDIA’s best.

AMD's reference model sports a triple-fan cooler display outputs comprising three DisplayPort and one HDMI connector, which is pretty par for the course these days. Its metal construction gives the card a really solid feel, but I have to say that it still doesn't feel as good as NVIDIA's Founders Edition models. 

2nd-generation Vega architecture

The new 7nm process technology (NVIDIA’s Turing cards are still based on 14nm) is a core part of how AMD managed to improve performance on the Vega architecture. For starters, the company was able to shrink the Vega GPU die from 495mm² to 331mm² , thus creating the space for two additional stacks of HBM2 memory and bringing the total amount of memory to 16GB. In this manner, the company effectively doubled the memory bandwidth of the Radeon RX Vega 64, and the Radeon VII now boasts a total memory bandwidth of 1TB/s.

On top of that, AMD cited optimizations to increase frequencies and reduce latencies, while increasing the bandwidth for the render output units to offer improved gaming performance.

The memory requirements for popular games have increased significantly over the years, so AMD thinks the Radeon VII’s generous 16GB of HBM2 memory and 4,096-bit memory bus will help accommodate the high-resolution textures found in many modern games. According to AMD, while the larger frame buffer may not make that much of a difference if you’re just looking at average frame rates, it can supposedly deliver a more consistent frame rate, which may give a smoother experience.

In addition, the card can also take advantage of AMD's High Bandwidth Cache Controller (HBCC), which reserves a portion of system memory for use by the GPU, effectively expanding the available VRAM. The HBCC then manages the migration of data between the card's VRAM and the system memory.

To sum up, here’s a look at how it stacks up against the previous Vega cards:-

Better thermal monitoring

Temperature monitoring capabilities have also been improved on the Radeon VII. Traditionally, a single sensor is placed in the vicinity of the legacy thermal diode, and readings from this sensor give what we see reported as the GPU temperature. This temperature is then used for fan control and to implement thermal throttling policies.

However, AMD says the Radeon VII now features a network of thermal sensors across the GPU die. The maximum temperature across the entire GPU die then gives something called the junction temperature, and it’s based on the temperature measuring circuitry across the die.

The Radeon VII GPU features 64 temperature sensors distributed across the chip, which is twice the number found on the Vega 64. The card also uses junction temperature for implementing thermal throttling and fan control, which confers several benefits.

First off, this supposedly allows the GPU to more reliably maximize its performance potential, instead of prematurely throttling performance according to reported temperatures from the hottest parts of the chip. In addition, AMD says junction temperatures provide a more effective control point for throttling and improve the reliability of the chips in question. Gamers will be able to view both junction and GPU temperature in Radeon Wattman, which gives them more insight into and control over their GPU.

Test Setup

The detailed specifications of our new graphics card testbed system is as follows:-

• Intel Core i7-8086K (4.0GHz, 12MB L3 cache)
• ASUS ROG Strix Maximus X Hero (Intel Z370)
• 4 x 8GB G.Skill Ripjaws V DDR4-3000 (Auto timings: CAS 15-15-15-35)
• Samsung 860 EVO 500GB SSD
• Windows 10 Pro 64-bit
• ASUS PB287Q, 4K monitor

The full line-up of graphics cards and their driver versions are listed below:

• AMD Radeon VII (Radeon Software Adrenalin 2019 Edition)
• NVIDIA GeForce RTX 2080 Ti (ForceWare 411.51)
• NVIDIA GeForce RTX 2080 (ForceWare 411.51)
• NVIDIA GeForce RTX 2070 (ForceWare 416.33)

[hwzcompare]

[products=658065,649099,649102,650161]

[width=200]

[caption=Test cards compared]

[showprices=1]

[/hwzcompare]

Benchmarks

Next up, here's a list of all the benchmarks used:

• 3DMark
• VRMark
• Ashes of the Singularity: Escalation
• Deus Ex: Mankind Divided
• Far Cry 5
• Middle-earth: Shadow of War
• Shadow of the Tomb Raider
• Tom Clancy's The Division

We used the Fire Strike Extreme benchmark and stress test in 3DMark for our temperature  and power consumption tests respectively.

3DMark

The synthetic 3DMark benchmark tests graphics and computational performance at different resolutions, starting at 1080p and going all the way up to 4K. A series of two graphics test, one physics test, and then a combined test stress your hardware in turn to assess its performance.

The Radeon VII came in somewhere between the GeForce RTX 2070 and 2080 in the Fire Strike benchmark, where it was just under 9 per cent quicker than the former card. However, in the 4K Fire Strike Ultra test, the Radeon VII inched ahead of the GeForce RTX 2080 by around 2 per cent. This is something that you'll see in real-world gaming benchmarks as well, and it seems like the Radeon VII is ever so slightly more competitive at ultra-demanding 4K resolutions, thanks to its massive super fast 16GB frame buffer.

VRMark

Futuremark’s VRMark benchmark is designed to assess a PC’s ability to handle high-performance headsets like the HTC Vive and Oculus Rift. If a PC passes the Orange Room test, it is ready for the latter two systems. The second graph also shows how the cards fared against the benchmark's target of an average FPS of 109. 

The Radeon VII trailed all the NVIDIA cards, coming in around 9 per cent behind the GeForce RTX 2070. 

Ashes of the Singularity: Escalation

Ashes of the Singularity has long been the poster child for the performance benefits a low-level API like DirectX 12 can bring. It is based on the Nitrous engine and can be extremely punishing thanks to the huge number of onscreen units and the sheer level of detail accorded to each unit. However, the CPU does become the limiting factor at lower resolutions and settings. 

The results were sort of a mixed bag for the Radeon VII. It trailed the GeForce RTX 2070 in most of the DirectX 11 benchmarks, but it proved quite capable at taking advantage of the shift to the low-level DirectX 12 API. In the DirectX 12 benchmarks, it pulled ahead significantly and managed to close the gap with the NVIDIA cards.

Deus Ex: Mankind Divided

Mankind Divided features just about every trick to make your game look pretty, including things like volumetric and dynamic lighting, screenspace reflections, and cloth physics. Even though it was released in 2016, the game is capable of bringing even the most powerful systems to their knees. 

The Radeon VII once again fell between the GeForce RTX 2070 and 2080 here, but the shift to DirectX 12 once again allowed it to show some strong results, where it outstripped even the GeForce RTX 2080 at times. However, I should point out that the NVIDIA cards actually suffer from decreased performance in DirectX 12 mode, so it's a matter of picking the API that suits your card best. And when you do that, it's often the case that the Radeon VII's performance is very similar to that of the GeForce RTX 2080. 

Far Cry 5

The latest installment in the Far Cry series is actually an AMD launch title, so it'll be interesting to see how the Radeon VII fares here.

The Radeon VII did better at higher resolutions here, perhaps owing to its generous 16GB of HBM2 memory and high memory bandwidth. Either way, it was able to match, or sometimes outstrip the GeForce RTX 2080, at 1440p and 4K resolutions. 

Middle-earth: Shadow of War

In Shadow of War, the Radeon VII once again turned out numbers between the GeForce RTX 2070 and 2080. However, while it did close the gap with the latter at higher resolutions, it wasn't quite able to outperform it this time. Still, it does come really close to the GeForce RTX 2080. At a 4K resolution and High settings, it was also roughly 13 per cent quicker than the GeForce RTX 2070. 

Shadow of the Tomb Raider

The same pattern repeats itself in Shadow of the Tomb Raider, with the Radeon VII doing a lot better against the Turing cards at a higher resolution. It's clear that the Radeon VII is its weakest at 1080p, where it even falls behind the GeForce RTX 2070 at times. Results like these are often attributed to a card's frame buffer size and in the case of the Radeon VII and its doubled video memory, this holds true where the performance drop off at each resolution step is less impacted than its competitor.

Temperature and power consumption

A look at the card's thermal, power, and acoustic performance highlights some of its most glaring weaknesses. While the peak temperature I measured was a modest 73°C during a 40-loop run of 3DMark's Fire Strike Extreme stress test, the card actually failed to successfully complete the run on quite a few occasions. The triple-fan cooler was also quite noisy, far noisier in fact that NVIDIA's Founders Edition models. 

On top of that, the 73°C temperature recorded here refers to the GPU temperature, there's also a separate junction temperature that you can view through the Radeon WattMan utility. As I mentioned earlier, the junction temperature is based on a network of sensors located across the GPU die, and it peaked at a disturbingly high 110°C. It's difficult to say if the high junction temperature had anything to do with the card failing 3DMark's stress test, but it sure looks like this is one card that runs pretty hot and noisy. 

In the power department, the Radeon VII proved quite power-hungry, guzzling only slightly less than the significantly more powerful GeForce RTX 2080 Ti. That isn't good for it in performance-per-watt terms, and it's pretty disappointing really, considering that the card is based on the more advanced 7nm process. Not to mention, it packs a lot more memory and that too might have contributed to the increased power draw.

Is being good enough actually enough?

With the Radeon VII, AMD finally has a card that can compete with NVIDIA's high-end gaming cards. That's pretty refreshing, especially considering that many of us weren't expecting a top-tier gaming GPU from AMD until Navi's release later this year. 

So how fast is the Radeon VII? Faster than the GeForce RTX 2070 in fact, but it's also retailing for slightly more at US$699. On NVIDIA's end, the GeForce RTX 2070 and 2080 cost US$599 and US$799 respectively, so the Radeon VII literally falls right smack in the middle. The price is illustrative of its performance as well, and the card put out numbers that fell between those two Turing cards for the most part. 

Honestly, this card isn't bad. It's just not very exciting or reassuring. Yes, it's based on a more advanced 7nm process, but that's not reflected in the power consumption numbers. Performance-wise, it's pretty impressive that it can go toe-to-toe with the GeForce RTX 2080 at times, but it also falls behind the GeForce RTX 2070 in other scenarios, including at the less demanding 1080p resolution. In other words, it feels inconsistent, where it's great sometimes, pretty good at others, and just meh the rest of the time. Furthermore, it produces quite a lot of heat and noise and only consumes slightly less power than the GeForce RTX 2080 Ti.

On top of all that, there's the feeling that this is sort of a stop-gap measure from AMD. The company released the Radeon RX 590 late last year, which was basically an overclocked Radeon RX 580. Now, it's announced the Radeon VII, a swanky new card that's the world's first 7nm GPU, but everyone's really just waiting for Navi. NVIDIA introduced some fancy ray-tracing technology with Turing, but the new Vega card isn't really bringing any new talking points to the table, except that AMD has now given us an alternative to Turing.

That said, I can understand why AMD is doing this. It's probably in the company's interest to keep the conversation going and remind consumers that it's still a player in the graphics space, especially at the high-end. The Radeon VII is a capable card, especially at 4K resolutions, and despite its flaws and inconsistencies, it's nice that consumers now have more options to choose from.

However, given what we've encountered while stress testing the card, we're not quite sure if it's a worthy alternative - especially for its asking price. Speaking of value, don't forget that quite a bit of the card's cost goes towards its 16GB of video memory which may not be something everyone can appreciate - depending on the resolution you intend to game.

Our articles may contain affiliate links. If you buy through these links, we may earn a small commission.