AMD may be headed back to the well with Polaris, this time launching the new GPU on a 12nm version of GlobalFoundries 14nm process node. There’s been rumor of this coming, though there have also been questions about whether AMD would attempt a 12nm refresh given the reality of Nvidia’s Turing launch.
According to Phoronix, a new PCI ID, 0x6FDF, has been added to Linux. It’s not mentioned in any database of Radeon drivers or GPU models anywhere online. It’s listed as being part of the Polaris 10 family — in Linux, Polaris 20 GPUs (RX 500 series) are part of the Polaris 10 family as well, which increases the chance that this is a 12nm die shrink. What might we expect from such a part?
The Ryzen 7 2700X improves on its predecessors' 1T performance by 1.10x, courtesy of its faster clocks. Multi-threading ticks up a hair more, at 1.11x. In multi-threaded code, the Ryzen 7 2700X is a full 1.26x faster than the Core i7-8700K.
7zip's decompression test has always been Ryzen-friendly and it scales well with SMT — another Ryzen strength. Both Intel and AMD use simultaneous multi-threading, but AMD tends to pick up more performance from its implementation than Intel does.
Our Handbrake encode resizes a 4K video file into the 1080p "Normal" legacy profile with a 17,100 Kbps VBR in both H.264 and the newer HEVC / H.265. Ryzen has always led in H.264 but lagged Intel in H.265. Here, the 2700X's higher clocks helps close the gap with Intel, though the Core i7-8700K remains top dog by a significant margin.
Maxwell Render 4's "Benchwell" scene has typically favored Intel CPUs over AMD and scales strongly with additional cores. AMD's performance has improved significantly since Ryzen launched; the 2700X beats out the closest Intel competitor by 1.15x.
Dolphin Emu's benchmarks likes Intel CPUs and high clocks, but the 2700X does gain ground here. It can't match the 8700K, but it's effectively tied with the Core i7-6900K.
Our three Blender scenes have been strong points for Ryzen 7 and Ryzen 5 since these chips debuted. The 8700K managed to pull almost even with the 1800X, but the 2700X has opened the field again. The 8700K takes 1.12 - 1.18x longer to render the three scenes.
Company of Heroes 2 is the RTS sequel to the acclaimed Company of Heroes and it remains downright stingy with its performance. The R9 390 remains basically tied with the RX 480, but the RX 580 pulls ahead on all fronts, only surpassed by the much more expensive GTX 1070.
Metro Last Light Redux is an updated version of the original Metro Last Light. Our tests put a heavy load on all GPUs by using supersampled antialiasing. Here, the GTX 1060 still leads the RX 580, with the R9 390 delivering the fastest AMD performance overall.
AoTS was the first DX12 game on the market and remains an impressive demo of what that API can do. The RX 580 has just enough advantage over the RX 480 to take the 1080p benchmark and tie with the GTX 1060 at 1440p. The R9 390 is a bit faster in 1080p than RX 480, but ties it at 1440p.
In Hitman, the RX 580 performed extremely well, winning by a whisker against the GTX 1060 and leaving its own predecessors well behind. The R9 390 is a whisker ahead of the RX 480, but the difference falls into the margin of error.
Shadow of Mordor is one of the best-regarded RPG/action hybrids of the past few years. Nvidia cards typically perform better than AMD hardware in this title, so it's not surprising to see Team Green win overall. The RX 580 is faster than both the RX 480 and R9 390, and effectively tied with the GTX 1060 at 1440p.
Rise of the Tomb Raider is the second game in the rebooted Tomb Raider franchise. Here, the RX 580 again pulls well away from the RX 480 and the R9 390 and noses past the GTX 1060 to overall victory. The R9 390's additional ROPs (64, compared to 32 for the RX 480 and RX 580) don't give it any help here.
We've moved from Total War: Rome 2 and to the fantasy-themed Warhammer title with DX12 support. The performance gaps between the RX 480 and RX 580 are quite small here, and the R9 390 actually ekes out a win at 1440p compared to either AMD card. The GTX 1060 enjoys a very small advantage at 1440p, but the R9 390 very nearly ties it.
DiRT Rally is the most recent rally game by Codemasters, and is somewhat newer than the Autosport we've used for our recent CPU tests. Here, Nvidia scores some of the biggest wins we've seen in this review, vastly outpacing any of the AMD cards. The RX 580 does turn in the best performance overall, however, with frame rates somewhat ahead of the R9 390.
We tested Doom at 1080p in both Vulkan and OpenGL. The results aren't all that surprising — AMD is well behind Nvidia in OpenGL, but ahead of the 1060 in Vulkan. The R9 390 can't quite beat the GTX 1060 in that API, but the RX 480 ties it and the RX 580 pulls ahead.
Holes in Turing’s Product Line
The idea that AMD would pull a refresh like this seemed unlikely back when Nvidia was thought to be launching a top-to-bottom stack refresh. But today, that doesn’t look to be the case. The performance hit from enabling features like RTX is so heavy, it’s not clear if Nvidia can even launch the feature below the RTX 2070, which means Turing may effectively be a refreshed GPU family with a lot of Pascal cards hanging on. Nothing is stopping NV from eventually launching new “2060” cards that are basically rebadged 1060s, and both AMD and Nvidia have gone down this road before.
But Nvidia may not want to take that step this time around. The question for the company is whether it makes more sense to position the new RTX family as entirely devoted to ray tracing and DLSS with the old Pascal numbering system reserved for non-RTX cards or if bringing these other GPUs forward into the Turing product line and labeling them with the new naming convention would confuse buyers. I’m inclined to think it would; buyers are unlikely to grasp the distinction between an “RTX 2070” and a hypothetical “GTX 2060” if Nvidia tries to make the “G” alone the distinguishing factor for whether new features are available. Regardless of whether or not Nvidia keeps Pascal in market or attempts to re-brand older cards as part of the Turing family, AMD may have an opportunity to better position itself in the mid-market.
Of course, all of this is supposition, based on some PCI ID codes for a product family, not actual hardware. But if AMD can snag a win across the midrange for a die-shrunk version of its Polaris family, the company may well go for it. Improving its competitive midrange position is a good way for AMD to win back market share, even if it isn’t competing with the RTX family.