A Quick Note on Architecture & Features

With pages upon pages of architectural documents still to get through in only a few hours, for today’s launch news I’m not going to have the time to go in depth on new features or the architecture. So I want to very briefly hit the high points on what the major features are, and also provide some answers to what are likely to be some common questions.

Starting with the architecture itself, one of the biggest changes for RDNA is the width of a wavefront, the fundamental group of work. GCN in all of its iterations was 64 threads wide, meaning 64 threads were bundled together into a single wavefront for execution. RDNA drops this to a native 32 threads wide. At the same time, AMD has expanded the width of their SIMDs from 16 slots to 32 (aka SIMD32), meaning the size of a wavefront now matches the SIMD size. This is one of AMD’s key architectural efficiency changes, as it helps them keep their SIMD slots occupied more often. It also means that a wavefront can be passed through the SIMDs in a single cycle, instead of over 4 cycles on GCN parts.

In terms of compute, there are not any notable feature changes here as far as gaming is concerned. How things work under the hood has changed dramatically at points, but from the perspective of a programmer, there aren’t really any new math operations here that are going to turn things on their head. RDNA of course supports Rapid Packed Math (Fast FP16), so programmers who make use of FP16 will get to enjoy those performance benefits.

With a single exception, there also aren’t any new graphics features. Navi does not include any hardware ray tracing support, nor does it support variable rate pixel shading. AMD is aware of the demands for these, and hardware support for ray tracing is in their roadmap for RDNA 2 (the architecture formally known as “Next Gen”). But none of that is present here.

The one exception to all of this is the primitive shader. Vega’s most infamous feature is back, and better still it’s enabled this time. The primitive shader is compiler controlled, and thanks to some hardware changes to make it more useful, it now makes sense for AMD to turn it on for gaming. Vega’s primitive shader, though fully hardware functional, was difficult to get a real-world performance boost from, and as a result AMD never exposed it on Vega.

Unique in consumer parts for the new 5700 series cards is support for PCI Express 4.0. Designed to go hand-in-hand with AMD’s Ryzen 3000 series CPUs, which are introducing support for the feature as well, PCIe 4.0 doubles the amount of bus bandwidth available to the card, rising from ~16GB/sec to ~32GB/sec. The real world performance implications of this are limited at this time, especially for a card in the 5700 series’ performance segment. But there are situations where it will be useful, particularly on the content creation side of matters.

Finally, AMD has partially updated their display controller. I say “partially” because while it’s technically an update, they aren’t bringing much new to the table. Notably, HDMI 2.1 support isn’t present – nor is more limited support for HDMI 2.1 Variable Rate Refresh. Instead, AMD’s display controller is a lot like Vega’s: DisplayPort 1.4 and HDMI 2.0b, including support for AMD’s proprietary Freesync-over-HDMI standard. So AMD does have variable rate capabilities for TVs, but it isn’t the HDMI standard’s own implementation.

The one notable change here is support for DisplayPort 1.4 Display Stream Compression. DSC, as implied by the name, compresses the image going out to the monitor to reduce the amount of bandwidth needed. This is important going forward for 4K@144Hz displays, as DP1.4 itself doesn’t provide enough bandwidth for them (leading to other workarounds such as NVIDIA’s 4:2:2 chroma subsampling on G-Sync HDR monitors). This is a feature we’ve talked off and on about for a while, and it’s taken some time for the tech to really get standardized and brought to a point where it’s viable in a consumer product.

AMD Announces Radeon RX 5700 XT & RX 5700 Addendum: AMD Slide Decks
Comments Locked

326 Comments

View All Comments

  • elwro - Tuesday, June 11, 2019 - link

    Sapphire will fix it
  • AshlayW - Monday, June 10, 2019 - link

    I am really disappointed with the price, like, really disappointed. Honestly they don't really bring a lot new to the table and as far as I can tell with the reference design: 8+6 pin means it isn't really that efficient if it needs that power. And the blower, ugh, just no.

    I knew this wasn't for me, I am happy to stay with my Radeon VII +Kraken G12+Asetek 570LC modification until next year but I was kinda hoping these RX 5700XT and 5700 to be priced like 349 and 299 respectively. Yes... Please don't hate on me, I know AMD isn't a charity and 7nm process is likely expensive but this chip succeeds Polaris and i was hoping it would be priced like it. Honestly I think custom Vega 56 for £250 is still the champ - these cards are readily available for this price here.

    I await Big Navi RDNA+HBM2, or maybe I get an Nvidia 7nm card, but I will not pay more than 699 US / £650 for the card and I will not buy with less than 11 GB of Video Memory.
  • AshlayW - Monday, June 10, 2019 - link

    Please don't hate me for "bashing AMD" : I am just a bit negative today and seeing the down sides of all things.
  • Opencg - Tuesday, June 11, 2019 - link

    I agree. The rumors had these things priced WAAAY lower. If they come out slightly agead of the 2070 then there's maybe reason for excitemen. But I honestly think that the favorable comparison amd showed off today was all best case benchmarks. Either way maybe they have some room to come down. So a price war in the mid teir may be what they are trying to encite vs nvidia. (They would probably benefit more to eliminate nvidias margins before they drop)

    Oh well wait for the benchmarks I guess. I really freaking hope they lose the f$^^#&= blower syle cards for navi 20. seriously
  • Meteor2 - Tuesday, June 11, 2019 - link

    People seem to be under the illusion that 7nm is cheap to manufacture. It’s not! It costs more than the next-larger nodes.
  • PixyMisa - Tuesday, June 11, 2019 - link

    Currently it costs as much per transistor as 12nm, so nearly twice as much for a given die area. That will come down, but right now it is not cheap.
  • Opencg - Tuesday, June 11, 2019 - link

    I agree. I bought into the hype for 2 reasons. Navi was supposed to use a new architecture that could scale well using inerposers. If that is true well they didn't use it for these cards.

    The second reason is that the current generation of cards are still greatly inflated in price due to rtx and tensor cores taking up silicon. I was hoping that AMD could just match pascals value and that would give them an edge. They only got halfway there though. Perhaps they are making insane margins on these and have room to go down.

    Either way it's a half let down for me. Im still waiting for the day when a dollar could buy as much performance as the 10 series when mining wasn't inflating.
  • Irata - Tuesday, June 11, 2019 - link

    Same here as far as price is concerned - was hoping for at least $50 less, but if nVidia responds with either faster spec and or lower price RTX 2060 / 2070 then we may still see lower prices across the board with everyone winning, regardless of which brand they prefer.

    In the end, we will have to wait for reviews and what else RX5700 contains in terms of features.
  • eva02langley - Thursday, June 13, 2019 - link

    This is what fanboys don't understand. It was planned by AMD when they heard about Super.
  • catavalon21 - Friday, July 5, 2019 - link

    As it turns out, the 5700 XT just became $50 less - 2 days before launch!

Log in

Don't have an account? Sign up now