We need some clarity on this issue, it seems.
TLDR: Some AMD RX 560 graphics cards are selling with 2 CUs disabled, resulting in 896 streaming processors to the initially advertised 1024 (64 SPs per CU). Here’s the deal: That card already exists, and it’s called an RX 460; in fact, the first two lines of our initial RX 560 review explicitly states that the driving differentiator between the 460 and 560, aside from the boosted clocks, was a pre-enabled set of 2CUs. The AMD RX 460s could already be unlocked to have 16 CUs, and the RX 560 was a card that offered that stock, rather than forcing a VBIOS flash and driver signature.
The RX 560 with 2CUs disabled, then, is not a new graphics card. It is an RX 460. We keep getting requests to test the “new” RX 560 versus the “old” RX 560 with 1024 SPs. We already did: The RX 560 review contains numbers versus the RX 460, which is (literally) an RX 560 14CU card. It is a rebrand, and that’s likely an attempt to dump stock for EOY.
Microsoft has, rather surprisingly, made it easy to get into and maintain the Xbox One X. The refreshed console uses just two screws to secure the chassis – two opposing, plastic jackets for the inner frame – and then uses serial numbering to identify the order of parts removal. For a console, we think the Xbox One X’s modularity of design is brilliant and, even if it’s just for Microsoft’s internal RMA purposes, it makes things easier for the enthusiast audience to maintain. We pulled apart the new Xbox One X in our disassembly process, walking through the VRM, APU, cooling solution, and overall construction of the unit.
Before diving in, a note on the specs: The Xbox One X uses an AMD Jaguar APU, to which is affixed an AMD Polaris GPU with 40 CUs. This CU count is greater than the RX 580’s 36 CUs (and so yields 2560 SPs vs. 2304 SPs), but runs at a lower clock speed. Enter our errata from the video: The clock speed of the integrated Polaris GPU in the Xbox One X is purportedly 1172MHz (some early claims indicated 1720MHz, but that proved to be the memory speed); at 1172MHz, the integrated Polaris GPU is about 100MHz slower than the original reference Boost of the RX 480, or about 168MHz slower than some of the RX 580 partner models. Consider this a correction of those numbers – we ended up citing the 1700MHz figure in the video, but that is actually incorrect; the correct figure is 1172MHz core, 1700MHz memory (6800MHz effective). The memory operates a 326GB/s bandwidth on its 384-bit bus. As for the rest, 40 CUs means 160 TMUs, giving a texture fill-rate of 188GT/s.
AMD’s RX 560 continues a trend of refreshing the Polaris line, but with a more notable change than the previous RX 580RX 570 refreshes: The RX 560 fully unlocks itself to 16 CUs, up from the previous 14 CUs of the RX 460. This change (in addition to voltage-frequency changes) instantly accounts for performance increases over the RX 460, theoretically making for a more exciting update than was had with the 580 & 570. That’s not to say that the predecessors of this 500 line were unworthy, but they certainly weren’t eye-catching for anyone who’d followed the 400-series launch.
Our review of the Sapphire RX 560 Pulse OC 4GB ($115) card is the first look at this new low-end line from AMD, updating the entry-level, sub-$120 market (in theory) with fresh competition. The incumbent would be the GTX 1050, which we previously thought a better buy than the RX 460. Today, we’re seeing how that’s changed in seven months.
To catch everyone up on the RX 500 refresh thus far, it’s mostly been a glorified BIOS update to the RX 580 and RX 570 cards, driving higher frequency, permitting higher voltage under OV, and trading more power for some performance. Nothing special, but enough to keep AMD in the game until its eventual Vega launch. We found the RX 580 to be a strong competitor to the GTX 1060, particularly at the price point, though noted that owners of RX 480 series cards shouldn’t bother considering an upgrade – because it’s not one. This 500 series is not meant for owners of the 400 series. Tune out until Vega, Volta, or high-end Pascal makes sense.
Sapphire’s RX 560 Pulse OC has one of the weakest cooling solutions we’ve seen of late, but – as we learn in our VRM+VRAM temperature testing – it’s sufficient for this type of card. A low-end GPU doesn’t draw much power, and so Sapphire skates by with its MagnaChip Semiconductor MDU1514 + MDU1517 3-phase power design.
As this content is relatively straight-forward, given the low price, let’s dive straight into testing.
AMD’s taken a page out of nVidia’s book, apparently, and nVidia probably took that page from Apple – or any number of other companies that elect to re-use product names. The new Radeon Pro Duo uses the same name as last year’s launch, but has updated the internals.
The RX 580, as we learned in the review process, isn’t all that different from its origins in the RX 480. The primary difference is in voltage and frequency afforded to the GPU proper, with other changes manifesting in maturation of the process over the past year of manufacturing. This means most optimizations are relegated to power (when idle – not under load) and frequency headroom. Gains on the new cards are not from anything fancy – just driving more power through under load.
Still, we were curious as to whether AMD’s drivers would permit cross-RX series multi-GPU. We decided to throw an MSI RX 580 Gaming X and MSI RX 480 Gaming X into a configuration to get things close, then see what’d happen.
The short of it is that this works. There is no explicit inhibitor built in to forbid users from running CrossFire with RX 400 and RX 500 series cards, as long as you’re doing 470/570 or 480/580. The GPU is the same, and frequency will just be matched to the slowest card, for the most part.
We think this will be a common use case, too. It makes sense: If you’re a current owner of an RX 480 and have been considering CrossFire (though we didn’t necessarily recommend it in previous content), the RX 580 will make the most sense for a secondary GPU. Well, primary, really – but you get the idea. The RX 400 series cards will see EOL and cease production in short order, if not already, which means that prices will stagnate and then skyrocket. That’s just what retailers do. Buying a 580, then, makes far more sense if dying for a CrossFire configuration, and you could even move the 580 to the top slot for best performance in single-GPU scenarios.
AMD’s Polaris refresh primarily features a BIOS overhaul, which assists in power management during idle or low-load workloads, but also ships with natively higher clocks and additional overvoltage headroom. Technically, an RX 400-series card could be flashed to its 500-series counterpart, though we haven’t begun investigation into that just yet. The reasoning, though, is because the change between the two series is so small; this is not meant to be an upgrade for existing 400-series users, but an option for buyers in the market for a completely new system.
We’ve already reviewed the RX 580 line by opening up with our MSI RX 580 Gaming X review, a $245 card that competes closely with the EVGA GTX 1060 SSC ($250) alternative from nVidia. Performance was on-point to provide back-and-forth trades depending on games, with power draw boosted over the 400 series when under load, or lowered when idle. This review of the Gigabyte RX 570 4GB Aorus card benchmarks performance versus the RX 470, 480, 580, and GTX 1050 Ti and 1060 cards. We're looking at power consumption, thermals, and FPS.
There’s no new architecture to speak of here. Our RX 480 initial review from last year covers all relevant aspects of architecture for the RX 500 series; if you’re behind on Polaris (or it’s been a while) and need a refresher on what’s happening at a silicon level, check our initial RX 480 review.
AMD’s got a new strategy: Don’t give anyone time to blink between product launches. The company’s been firing off round after round of products for the past month, starting with Ryzen 7, then Ryzen 5, and now Polaris Refresh. The product cannon will eventually be reloaded with Vega, but that’s not for today.
The RX 500 series officially arrives to market today, primarily carried in on the backs of the RX 580 and RX 570 Polaris 10 GPUs. From an architectural perspective, there’s nothing new – if you know Polaris and the RX 400 series, you know the RX 500 series. This is not an exciting, bombastic launch that requires delving into some unexplored arch; in fact, our original RX 480 review heavily detailed Polaris architecture, and that’s all relevant information to today’s RX 580 launch. If you’re not up to speed on Polaris, our review from last year is a good place to start (though the numbers are now out of date, the information is still accurate).
Both the RX 580 and RX 570 will be available as of this article’s publication. The RX 580 we’re reviewing should be listed here once retailer embargo lifts, with our RX 570 model posting here. Our RX 570 review goes live tomorrow. We’re spacing them out to allow for better per-card depth, having just come off of a series of 1080 Ti reviews (Xtreme, Gaming X).
Part 1 of our interview with AMD's RTG SVP & Chief Architect went live earlier this week, where Raja Koduri talked about shader intrinsic functions that eliminate abstraction layers between hardware and software. In this second and final part of our discussion, we continue on the subject of hardware advancements and limitations of Moore's law, the burden on software to optimize performance to meet hardware capabilities, and GPUOpen.
The conversation started with GPUOpen and new, low-level APIs – DirectX 12 and Vulkan, mainly – which were a key point of discussion during our recent Battlefield 1 benchmark. Koduri emphasized that these low-overhead APIs kick-started an internal effort to open the black box that is the GPU, and begin the process of removing “black magic” (read: abstraction layers) from the game-to-GPU pipeline. The effort was spearheaded by Mantle, now subsumed by Vulkan, and has continued through GPUOpen.
We've been working with AMD since our RX 480 & RX 470 reviews to troubleshoot some driver-related screen hangs. In our July notice, we strongly encouraged that our readers on RX 400 series cards avoid the 16.7.3 drivers, following instability and file corruption on one of our test benches. Since then, we've been in correspondence with AMD on the issue, and finally have some good news for folks who've encountered the green / black / blue / yellow screen hangs with AMD drivers.
The theoretical end of AMD's Polaris desktop GPU line has just begun shipment, and that's in the form of the RX 460. Back at the pre-Computex press event, AMD informed us that the Polaris line would primarily consist of two GPUs on the Polaris architecture – Polaris 10 & 11 – and that three cards would ship on this platform. Two of the three have already shipped and been reviewed, including the ~$240 RX 480 8GB cards (review here) and ~$180-$200 RX 470 cards (review here). The next architecture will be Vega, in a position to potentially be the first consumer GPU to use HBM2.
Today, we're looking at Polaris 11 in the RX 460. The review sample received is Sapphire's RX 460 Nitro 4GB card, pre-overclocked to 1250MHz. The RX 460, like the 470, is a “partner card,” which means that no reference model will be sold by AMD for rebrand by its partners. AMD has set the MSRP to $110 for the RX 460, but partners will vary widely depending on VRAM capacity (2GB or 4GB), cooler design, pre-overclocks, and component selection. At time of writing, we did not have a list of AIB partner prices and cards available.
As always, we'll be reviewing the Sapphire RX 460 4GB with extensive thermal testing, FPS testing in Overwatch, DOTA2, GTA V, and more, and overclock testing. Be sure to check page 1 for our new PCB analysis and cooler discussion, alongside the in-depth architecture information.
We moderate comments on a ~24~48 hour cycle. There will be some delay after submitting a comment.