The AMD Radeon RX 590 Review, feat. XFX & PowerColor: Polaris Returns (Again)
by Nate Oh on November 15, 2018 9:00 AM ESTClosing Thoughts
As we bring this to a close, the Radeon RX 590 features few, if any, surprises, indeed outperforming the RX 580 via higher clockspeeds but at the cost of additional power consumption. This is enabled by being fabbed on 12nm and being afforded higher TBPs, but in terms of overall effect, the port to 12nm is less of a die shrink and more of a new stepping; in terms of transistor count and die size Polaris 30 is identical to Polaris 10. Feature-wise, whereas the RX 500 series did introduce a new mid-power memory state, the RX 590 doesn't add anything.
Altogether, AMD is helping push the RX 590 along with a three-game launch bundle and emphasizing the value of FreeSync monitors. One might add that the RX 590 also makes the most out of the game bundle, which now presents itself as a direct value-add for the card. But the card by itself isn't providing more bang for your buck than with the RX 580.
By the numbers, then, where does the RX 590 land? Reference-to-reference, the RX 590 is about 12% faster than the RX 580 at 1080p/1440p, and 14-15% faster than the RX 480. Across the aisle, this turns out to be 9% faster than the reference GTX 1060 6GB, though the GeForce card still takes the lead in games like GTA V and Total War: Warhammer II. That matchup itself is more-or-less an indicator of how far Polaris has come - the RX 480 at launch was 11% behind the GTX 1060 6GB, and the RX 580 at launch was 7% behind. Improvements have come over the years with drivers and such, where the RX 480 and RX 580 are much closer, but the third time around, Polaris can finally claim the victory at launch.
That being said, the 9% margin is well within reach of factory-overclocked GTX 1060 6GB cards, especially with a 9Gbps option that the RX 590 doesn't have. Even with the XFX Radeon RX 590 Fatboy the delta only increases to about 11%. In other words, as custom factory-overclocked cards the RX 590 and GTX 1060 6GB are likely to trade blows. And heavily factory-overclocked RX 580s are in the similar situation.
But similar to the RX 580 and 570, the RX 590 achieves this at the cost of even more power consumption. In practical situations like in Battlefield 1, the RX 590 results in the system pulling around 110W more from the wall than with the GTX 1060 6GB. The RX 590 is not a power efficient card at the intended clocks and TBP, and so won't be suitable for SFF builds or reducing air conditioning usage.
Once again though, we've observed that VRAM is never enough. It was only a few years ago that 8GB of VRAM was considered excessive, only useful for 4K. But especially with the popularity of HD texture packs, even 6GB of framebuffer could prove limiting at 1080p with graphically demanding games. In that respect, the RX 590's 8GB keeps it covered but also brings additional horsepower over the 8GB R9 390. For memoryhogging games like Shadow of War, Wolfenstein II, or now Far Cry 5 (with HD textures), the 8GBs go a long way. Even at 1080p, GTX 980/970 performance in Wolfenstein II tanks because of the lack of framebuffer. For those looking to upgrade from 2GB or 4GB cards, both the RX 580 and RX 590 should be of interest.
So in terms of a performance gap, the custom RX 590s seem poised to fill in, though it might be in the range of 5 to 10% when comparing to existing heavily factory-overclocked RX 580s. That gap, of course, would likely have been filled by a 'Vega 11' product, which never came to fruition. Beyond a smaller discrete Vega GPU, 12nm Vega seems less likely given that 14nm+ Vega is no longer on the roadmap. And with 7nm Vega announced as Radeon Instinct MI60 and MI50 accelerators, there's no current indication of those GPUs coming into the consumer space. We also know very little about Navi other than being fabbed with TSMC on 7nm and due in 2019.
So of the possible cards in AMD’s deck, we are seeing a single 12nm Polaris SKU slotting in above the existing RX 580. If there are no further refreshes coming, then gamers will naturally ask when Polaris will be replaced as AMD's mainstream offering, as otherwise the RX 500 series will be holding the fort until then. If this is a prelude to further refreshes, then it becomes a question of how much more performance can be squeezed from Polaris.
The bigger picture is that from the consumer point-of-view, RX Vega in August 2017 was the most recent video card launch before today. There are more pressing concerns in the present: NVIDIA's recently-launched Turing-based GeForce RTX 20 series, as well as Intel's renewed ambition for discrete graphics and subsequent graphics talent leaving AMD for Intel. While the RX 590 avoids the optics of having no hardware response whatsoever in the wake of the RTX 20 cards, it is still a critical juncture with respect to DirextX Raytracing (DXR) development. And regardless, Intel has announced 2020 as the date for their modern discrete GPUs.
So as we head into 2019, it will be a crucial year for AMD and RTG. But returning to the Radeon RX 590, it applies pressure to the GTX 1060 6GB and older GTX 900 series cards, while avoiding direct pressure on existing RX 580 inventory by virtue of pricing. And right now, without anything to compete with the GTX 1080 Ti/RTX 2070 range or above, AMD is likely more than happy to take any advantage where they can. For now, though, much depends on the pricing of top-end RX 580s.
136 Comments
View All Comments
silverblue - Thursday, November 15, 2018 - link
I was talking about the process; I know the 2000 series has done well.I didn't even know TSMC had a 12nm process, but either it's not very well suited to this application or AMD have just clocked it far too highly, so it's not a successful product in that sense.
frenchy_2001 - Thursday, November 15, 2018 - link
Nvidia's Volta and Turing are fabbed on TSMC 12nm.So, it seems to work well for GPUs, but AMD's architecture is just not competitive in their perfs/watt.
silverblue - Thursday, November 15, 2018 - link
I completely forgot about them.It feels like Polaris is bottlenecked in some way, and increasing clock speeds is just a brute force way of alleviating the issue at the cost of significant power consumption. Perhaps the design is just broken to begin with.
Manch - Friday, November 16, 2018 - link
Memory bandwidth is the bottleneck for Polaris.deksman2 - Friday, November 16, 2018 - link
Power consumption - wise, its the process node from GLOFO that's limiting Polaris mainly from achieving high frequencies and low power consumption.GLOFO nodes are designed for low clocks and mobile parts... 12nmLP is designed for the same, and AMD used it for RX 590.
That's why power consumption explodes on high frequencies.
deksman2 - Friday, November 16, 2018 - link
Polaris is bottlenecked by the GLOFO 12nmLP process that's being used here.It's designed for low clocks and mobile parts.
There was NOTHING posted about AMD using TSCM 12nm node for Polaris refresh. TSMC 12nm was slated for Nvidia.
AMD gets to use TSMC 7nm high performance process node for Zen 2, Vega Instinct and Navi.
eek2121 - Saturday, November 17, 2018 - link
"Limited" is the word you want to use to describe Polaris. "Broken" would imply it doesn't work at all. Polaris was never meant to be a high end architecture. They have just been doing refreshes because they are likely reorganizing their GPU division and coming up with a new architecture to replace GCN. Doing something like that takes time, and AMD has to continue generating revenue. Also, the 590 is not a bad card at all, while I have a 1080ti in my machine, I would definitely consider a Polaris card in any new machine I build (for friends, family, etc.)deksman2 - Friday, November 16, 2018 - link
Incorrect on AMD's architecture not being competitive because AMD is still using GLOFO 12nm LP process for RX 590 which is designed for low clocks and mobile parts.TSMC 16nm and 12nm processes are designed for high performance and efficiency... those nodes are superior to GLOFO (that's why AMD's GPU's end up sucking up A LOT of power at high frequencies - its because the process node from GLOFO cannot take it, and partly because the compute performance on Polaris is a lot bigger than on GTX 1060).
porcupineLTD - Thursday, November 15, 2018 - link
I highly doubt that, any source?deksman2 - Friday, November 16, 2018 - link
Actually 'Cooe', you are incorrect.AMD is using 12nmLP process from GLOFO for RX 590.
You can read about that here:
https://www.pcgamesn.com/amd-rx-590-overclocking-p...
Furthermore, the power consumption on RX 590 should be a dead give-away, because this is exactly the same thing that happened for Ryzen+ too (not to mention the fact we had 0 indications that AMD would refresh Polaris on TSMC 12nm process. NV got access to 12nm TSMC process, not AMD... AMD got access to TSMC's 7nm high perf. process and they have reserved Zen 2, Vega Instinct and Navi for that).
They increased the frequencies on 12nmLP, but as a result they also saw an increase in power consumption.
Polaris is clocked WAY beyond the voltage comfort zone on GLOFO processes (which are designed for low clocks and mobile parts).
If they wanted a refresh, they should have just dropped the frequencies down to 580 levels and call it a more power efficient rebrand.
If AMD moved to TSMC 12nm for RX 590, power consumption on this GPU would actually be lower than on GTX 1060 with those frequencies.