AMD Radeon RX Vega 56 Review
By Jimmy Thang on
While AMD’s $499 RX Vega 64 was built to go toe-to-toe with Nvidia’s GeForce GTX 1080, its more affordable Vega 56 sibling is designed to be the GTX 1070 killer. We’ve recommended the GTX 1070 for more gamers, since it’s considerably cheaper, but is powerful enough to max out games at 1440p or to push high-refresh rate 1080p displays. AMD’s RX Vega 56 is priced the same at $400, and makes a strong case at dethroning Nvidia’s venerable card.
Table of Contents
The RX Vega 56 uses the same external design as the RX Vega 64. Both cards measure approximately 10.5 inches long and use a black chassis coupled with a 30mm blower-style fan. Its connections, which include three DisplayPorts and one HDMI connection, are also the same. Its biggest difference is that it’s a 210-watt TDP card, which draws less power than it’s 295-watt sibling. It still requires two 8-pin connectors, however, and you’ll want at least a 700-watt PSU to power the GPU.
AMD Radeon RX Vega 56
AMD Radeon RX Vega 64
Nvidia GeForce GTX 1080
Nvidia GeForce GTX 1070
CUDA Cores/Stream Processors
Memory Bus Width
In terms of specs, the RX Vega 56 is a slightly lesser version of the RX Vega 64 in many ways. Its name is derived from the fact that it has 56 compute units, as opposed to 64 on the RX Vega 64. This also means that it has less stream processors and texture units. It still uses 8GB of AMD’s second generation high-bandwidth memory (HBM2), but its memory clock is reduced from 945MHz down to 800MHz.
The RX Vega 56 is a 10.5 teraflop GPU, which is 2.2 TFLOPS fewer than the RX Vega 64, but a whole 4.0 TFLOPS more than the GTX 1070. Don’t let the teraflop count mislead you, however, the competition is much closer than that number will have you believe. We’ll dive into performance in the next few slides.
Since AMD is pitting the RX Vega 56 against the GTX 1070, we're going to make direct comparisons against it. We're also going to add in benchmark numbers for the RX Vega 64 and GTX 1080 to give more context as well.
To ensure benchmarking consistency, we're using the same test rig for all GPUs, which is a system equipped with an Intel Core i7-5930K Haswell-E CPU clocked at 3.9GHz and 16GB of DDR4 RAM clocked at 2133MHz running in quad-channel mode.
We're going to test the cards using the same suite of synthetic, VR, and game benchmarks. These tests encompass three resolutions (1080p, 1440p, and 4K). We also ran each benchmark at their respective max settings to really put the cards through their paces.
3DMark 11 Extreme is a synthetic 1080p test that provides a score and measures a GPU's DirectX 11 performance. As you can see from the chart above, the RX Vega 56 scored a 7481, which is five percent less than the GeForce GTX 1070.
The rest of our 1080p tests show the RX Vega 56 and GTX 1070 trading blows, with AMD’s card winning three out of the five benchmarks here. Nvidia’s card managed to beat the RX Vega 56 in our Unigine Valley and BioShock Infinite tests, which are tests the RX Vega 64 also fell behind the competing GTX 1080.
At 1080p, it’s really too close to declare a definitive winner between the RX Vega 56 and the GTX 1070. AMD’s GPU never got below 70 average FPS in any of our 1080p tests, which indicate that it’s probably a bit overkill for standard HD gaming, unless you have a really high-refresh rate monitor.
While the GTX 1070 continues to beat the RX Vega 56 in Unigine Valley and BioShock Infinite, those leads diminish at 1440p. In the benchmarks where AMD’s card wins, it increases its lead. Overall, we give the slight advantage to the RX Vega 64 at 1440p.
Like the RX Vega 64, RX Vega 56 is really able to push itself more at 4K. Here, it’s able to beat the GTX 1070 in all of our tests. We suspect the card’s HBM2 video memory has a lot to do with the card being able to kick it into high gear here.
While the GPU can max out many 4K games, with it getting a 22.3 average FPS in Metro Last Light, it proves that it won’t be able to muster the performance to max out the most graphically demanding games at 2160p.
According to Valve’s VR Performance Test, the RX Vega 56 is a card that’s rated “very high” for virtual reality. When you look at the frame data, the benchmark suggests that the RX Vega 56 is four percent better for VR.
Acoustics and Temperatures
The RX Vega 56 is noticeably cooler than the RX Vega 64, which makes sense considering it’s a lower TDP card. When we gamed with it, we saw it generally stayed in the low 70 Cs and hit a high of 73. Its fan noise was also largely tolerable with it spinning at around 2,100RPM.
Unless you heavily prefer Nvidia’s ecosystem and have a G-Sync monitor, it’s hard to recommend the GTX 1070 over the RX Vega 56, considering they cost the same. While they’re pretty much neck and neck at 1080p, AMD’s card generally beats Nvidia’s GPU at 1440p, 4K, and VR.
The RX Vega 56 ends up being a great graphics card for high-refresh rate 1080p gaming, a standout card for 1440p, and one that can help you dip your toes into 4K. It, arguably, represents best-in-class performance for its price.