There have been teasers, press conferences, architectural announcements, and pricing games all around – and all before the first card has even shipped. The run-up to the launch of AMD’s new Radeon RX 5700 series of video cards has been a dynamic and memorable time, and a very fitting outcome for a family of cards named after AMD’s legendary Radeon HD 5700 series. However, at some point all the showmanship must come to an end and the cards will fall where they may: launch day is upon us for AMD’s Radeon RX 5700 series and RDNA GPU architecture.

I’m not sure there’s anything traditional about an AMD video card launch at this point, but today’s launch is about as non-traditional as they come, right on down to the Sunday launch date. AMD announced their video cards almost a month ago at an epic (ed: that’s EPYC) E3 event, taking the wraps off of both their new CPUs and GPUs. Rather than hold anything back, AMD came to E3 with everything front-loaded: specifications, prices, architectural details; everything except a pile of cards to sell. So we’ve been waiting for this moment for some time now, to test AMD’s claims about power, performance, and features, and see how they translate into real-world gaming performance. AMD has a lot that they want to do in the video card space, and riding high on their success with Zen the company’s ambition is once again palpable.

Getting down to business then, today is the launch of AMD’s next generation of video cards, the Radeon RX 5700 series. Aimed at what these days is the midrange segment of the video card market, AMD is looking to carve out a new place for the company in the hearts of gamers who are looking for high performance video cards that won’t break the bank. These parts are, in turn, based on AMD’s Navi 10 GPU, which is the first GPU using the company’s new RDNA architecture. And, while Navi 10 is not AMD’s first 7nm GPU – an honor the Radeon VII and its Vega 20 beat it to – it’s the first 7nm GPU that you’re actually going to want to pay attention to.

Altogether, AMD is rolling out two(ish) cards today. At $399 we have AMD’s new class-leading Radeon RX 5700 XT, which is a full-fledged Navi 10 card with all the trimmings. Meanwhile, for the slightly more budget conscious, we have the $349 Radeon RX 5700 (vanilla), a cut-down Navi 10 card that gives up some performance for lower pricing and lower power consumption. Finally, AMD is also launching their own “Anniversary Edition” version of the 5700 XT, which features a factory overclock and will sell for $449. (This card will be a footnote for today’s article, as it’s a limited-edition card that AMD isn’t sampling)

AMD Radeon RX Series Specification Comparison
  AMD Radeon RX 5700 XT AMD Radeon RX 5700 AMD Radeon RX 590 AMD Radeon RX 570
Stream Processors 2560
(40 CUs)
(36 CUs)
(36 CUs)
(32 CUs)
Texture Units 160 144 144 128
ROPs 64 64 32 32
Base Clock 1605MHz 1465MHz 1469MHz 1168MHz
Game Clock 1755MHz 1625MHz N/A N/A
Boost Clock 1905MHz 1725MHz 1545MHz 1244MHz
Throughput (FP32) 9.75 TFLOPs 7.95 TFLOPs 7.1 TFLOPs 5.1 TFLOPs
Memory Clock 14 Gbps GDDR6 14 Gbps GDDR6 8 Gbps GDDR5 7 Gbps GDDR5
Memory Bus Width 256-bit 256-bit 256-bit 256-bit
Transistor Count 10.3B 10.3B 5.7B 5.7B
Typical Board Power 225W 180W 225W 150W
Manufacturing Process TSMC 7nm TSMC 7nm GloFo/Samsung 12nm GloFo 14nm
Architecture RDNA (1) RDNA (1) GCN 4 GCN 4
GPU Navi 10 Navi 10 Polaris 30 Polaris 10
Launch Date 07/07/2019 07/07/2019 11/15/2018 08/04/2016
Launch Price $399 $349 $279 $179

For both gamers and AMD, the launch of the RX 5700 is an important one – and likely to be the most significant video card launch of the year. For consumers, this is the first series of high-volume video cards built on a 7nm process, pushing performance up and prices down at a time where video card pricing has been sluggish improve. For AMD, this launch gets their incredibly important RDNA architecture out the door to its first users, all the while renewing their competitiveness in the midrange market. The RX 5700 series aren’t flagship-level cards, but make no mistake: for AMD they’re still as important as any flagship launch.

AMD’s Radeon DNA (RDNA) architecture, in turn, is an interesting development. We’ll get into much greater detail later on, but at a high level RDNA is the most significant architectural development for AMD since the launch of Graphics Core Next in 2011. AMD and its engineers have made changes to their GPU architecture at some of its most fundamental levels, which comes with significant ramifications for performance, efficiency, and more. This is all to prepare their next-generation architecture for the company’s grand goals: RDNA and its successors will be in PC video cards, in APUs, in game consoles, and thanks to a partnership with Samsung, even in mobile GPUs.

Still, RDNA is only as valuable as the performance it brings, and this will especially be the case for the RX 5700 series. The first iteration of this architecture is all about changing the internal plumbing of AMD’s GPUs. As a result, there are no real marquee hardware features to speak of – AMD isn’t rolling out paradigm-changing features like ray tracing or the next Rapid Packed Math – so for consumers, the RX 5700 cards are essentially interchangeable with 2017’s Vega cards in terms of graphics functionality. Which is not to say that AMD is showing up empty-handed, but what new features it is rolling out – Radeon Anti-Lag and Radeon Image Sharpening – are software-based features that will be available to the entire Radeon product family. This means that the payoff for AMD needs to be in pricing, power consumption, and performance; the RX 5700 needs to deliver on the fundamentals.

Ultimately there are a lot of words I could spill on the subject of AMD, especially on today of all days, the launch of both their next-generation CPU and GPU architectures. But perhaps it’s best to keep things simple: today’s launch of the company’s Radeon RX 5700 series video cards and the RDNA architecture is a much-needed opportunity for the company to reset and recover in the video card space. Vega was ultimately underwhelming, Polaris is very long in the tooth, and AMD is still feeling the hangover from the cryptocurrency boom & busts. The company has held on to their consumer market share through aggressive pricing – RX 500 series cards are disruptively cheap – but AMD needs to be more visible and more profitable if they want to remain a viable competitor to juggernaut NVIDIA. Not to mention shoring up their position as Intel ramps up to become the third player in the video card space in the next couple of years.

Product Positioning, Availability & the Competition

Leading up to today’s launch, the announcement Radeon RX 5700 series has created quite a butterfly effect across the greater video card industry. With AMD having shown their cards early, NVIDIA, who has essentially been dominating the $300+ space since the start of the year, made their own preemptive move with the launch of the GeForce RTX 2070 Super and the RTX 2060 Super. A price cut in everything but name, the new Super cards saw NVIDIA essentially shift the performance of its $699 RTX 2080 and $499 RTX 2070 cards down to $499 and $399 respectively. These cards won’t go on sale for another two days (on a more traditional Tuesday), but it’s a launch that was clearly intended to shore up NVIDIA’s own position while taking some steam out of AMD’s launch.

AMD in turn made their own adjustments, cutting the price of their cards on Friday before they even launched. While the RX 5700 XT was originally set to launch at $449 and the RX 5700 (vanilla) at $379, these became $399 and $349 cards respectively before the first board was ever sold. These kinds of last-minute pricing shenanigans are not unheard of – first impressions count for a lot – however it’s been a long time since we’ve seen AMD and NVIDIA trading shots in quite such a direct manner. The net result is that, at least for the $349 to $499 segment of the video card market, the performance-per-dollar ratio just went up even more.

Overall, outside of today’s unusual launch date, this should be a pretty standard launch cycle for AMD. The company is launching with reference cards first, meaning that AMD and board partners alike will be selling cards based on AMD’s reference PCB and blower. Custom and semi-custom cards will come later, as supplies ramp up, board partners qualify their coolers, and new, GDDR6-capable PCBs are engineered. AMD has not disclosed how many cards are being distributed for the launch, but these days it’s rare to see a new generation of midrange (or better) cards not sell out at launch. So anyone interested in an RX 5700 card may need to act quickly.

Within AMD’s product stack, these new cards will be the backbone of AMD’s product lineup. The company’s Vega family of cards was already on its way out due to competition from NVIDIA, so this will be the final push for those cards. That will leave the Radeon VII above the RX 5700 cards, and then the RX 500 series below it. And while AMD hasn’t announced any other Navi GPUs, sooner or later those cards will get pushed out too by a lower-tier Navi GPU.

Trickier, perhaps, is placing the new cards within a historical context for AMD’s product lineups. Is the Radeon RX 5700 series the successor to the RX Vega series, or the RX 500 series? In terms of pricing and absolute performance, it’s closer to the former. However in terms of die size and relative performance gains, these cards feel a lot like the next-generation successors to Polaris. It’s a bit of an academic question – buyers are going to focus on perf-per-dollar first and foremost – but how these cards are framed will have an impact on how they’re received.

Sizing up the competition, AMD was originally going to launch these cards against NVIDIA’s GeForce RTX 2070 and RTX 2060. Now with the launch of the Super cards and the impending retirement of the RTX 2070, those matchups have changed. NVIDIA’s competition is now the RTX 2060 Super at $399, and the RTX 2060 (vanilla) at $349, both prices directly opposite AMD’s new cards. AMD in turn comes in with the edge on manufacturing process, as they’re using TSMC 7nm versus the 16nm offshoot that NVIDIA uses, however NVIDIA comes in with a notable feature advantage thanks to ray tracing and variable rate shading support. AMD and NVIDIA’s cards are not equal in features, and that will play a big part in their value.

Ultimately, today’s launch isn’t a case of a next-generation card coming in and wiping the floor with its last-generation competition; like the RTX 20 series launch last year, the RX 5700 launch is a more gradual shift in pricing and performance. GPU development is expensive, 7nm is even more expensive, and everyone is playing things a lot more conservatively than they did earlier this decade. For the moment then, the RX 5700 series can largely be considered to be part of the same generation of GPUs as the RTX 20/GTX 16 series, for all of the benefits and downsides that entails. The upside is that we get more to talk about, and get to form a more nuanced opinion, but for anyone looking for a simple recommendation for a new video card, there won’t be anything quite that simple with this launch.

Finally, both AMD and NVIDIA will be looking to tip the scales with game bundles. On the AMD side, the company is launching both its new CPUs and GPUs with their new Xbox Game Pass for PC bundle, which will see the products come with a 3-month voucher for Microsoft’s new game subscription service. Meanwhile NVIDIA is bundling Wolfenstein: Youngblood with its RTX 2060, while the new RTX 2060 Super and RTX 2070 Super cards get that plus Control as well.

July 2019 GPU Pricing Comparison
Radeon VII $699 GeForce RTX 2080 Super
  $499 GeForce RTX 2070 Super
Radeon RX 5700 XT $399 GeForce RTX 2060 Super
Radeon RX 5700 $349 GeForce RTX 2060
Meet the Radeon RX 5700 XT & Radeon RX 5700
Comments Locked


View All Comments

  • eastcoast_pete - Tuesday, July 9, 2019 - link

    Actually, I like the idea used in the graph you linked to: $ per fps, averaged from 18 games, all at 1080p very high settings. It allows a value comparison all the way from lower to high-end cards.
  • Meteor2 - Monday, July 8, 2019 - link

    C'mon jjj you're better than that.
  • sgkean - Monday, July 8, 2019 - link

    How does enabling the various advanced features (Ray Tracing, AMD Fidelity FX, AMD Image Sharpening) affect the game scores? With the performance being so close, and these new features/technologies being the main difference, would be nice to see what effect they have on performance.
  • Wardrop - Monday, July 8, 2019 - link

    I assume the noise of these is such due to the use of a blower? I'm guessing we'll have to wait for custom PCB's and coolers to get something quieter, or otherwise got with water cooling.
  • xrror - Tuesday, July 9, 2019 - link

    Argh... yet again, it seems like AMD is pushing beyond the sweet spot of the process node to try and force as much raw performance out as they can.

    I really don't want to be yet another person bashing on Raja. He probably did get a bit of "short changed" on personnel resources at AMD as Ryzen really DID need to succeed else AMD dies. And he did deliver on giving good GPU compute GPU cores for the higher margin workstation markets.

    But... it just feels like AMD needs to get to terms with their fabrication node and how to get GPU cores to "kickith the butt" beyond beating Intel IGP graphics.

    Which... feels unfair in a way. The only reason AMD "sucks" is that nVidia right now is so stupid dominant in discrete graphics (and major kudo's to nVidia for mastering that on an "older node" even). I mean even Intel had really bad problems porting it's IGP graphics to 10nm Cannon Lake.

    But that all said, RX 5700 really feels like it's fighting against the process node to not suck. Intel may (hopefully, might) actually get it's s**t together and bring forth a competitive descrete card (and if they "fail" guess what, that fail will hammer the lower end market) and nVidia...

    well like, nVidia even -2 process nodes behind at this rate would probably still be faster. Which is stupid. All credit to nVidia, it's just I really hoped for a few more process "rabbits out of the hat" before GPU's slammed into the silicon stagnation wall.

    I just wish we could have gotten maybe a doubling of graphics performance for VR before "market forces" determined that a VR/4K capable video setup is going to cost you over $1000.
  • Meteor2 - Tuesday, July 9, 2019 - link

    "RX 5700 really feels like it's fighting against the process node to not suck." -- what are you talking about?
  • peevee - Thursday, July 11, 2019 - link

    Actually, for GPUs with their practically linear scaling of performance from ALUs, using the densest nodes is the right approach. They probably should have used denser, low-power variant (libraries) of TSMC's "7nm" process and add more ALUs in the same space at the expense of frequency, but that would be different from what Ryzen 3, so add the extra expense to R&D.
  • CiccioB - Tuesday, July 9, 2019 - link

    In few words, AMD just used a lot of transistors and W just to get near Pascal efficiency.
    Thanks to the new 7nm PP they manage to create something that looks like acceptable.
    But as we already saw in the past, they somewhat filled the gap only because Nvidia is still waiting for the new PP to become cheaper.
    Once it will, Nvidia new architecture is going to leave these useless piece of engineering in the dust. Be it just a Turing shrink with no other enhancements.
    10 Billions transistors to improve IPC of about 1.25x and spare just few W thanks to the 7nm PP. And be on par to Pascal at the end. 10 Billions transistors without the support of a single advanced feature that Turing has, such has VRS that is going to improve performances a lot in future games and is going to be the real trump card for Nvidia against this late Pascal, no mesh shading or similar, no FP+INT, no RT and no tensors that can be used for many things included advanced AI.
    10 billions transistors that simply have given evidence that GCN is problematic and really needs a lot of workarounds to perform well. 4.4 millions transistors used to improve GCN efficiency. And that resulted in a mere 1.25x.
    10 billions transistors spent on fixing a crap architecture that would not be enough to make it look good but, again, if the frequency/W curve would not have been ignored completely making this chip consume the same as the rival which is on a older PP. Like for all the previous failing architectures starting from Tahiti.

    In the end this architecture is a try to fix an un-fixable GCN and relies only on the delay that Nvidia has in the 7nm adoption. On the same node it would have been considered the same as Polaris or Vega, big, hot, worthless to the point to be sold with no margins.
    As we can see this is equal in being a waste of transistors and W and has been discounted even before launched. Worthless piece of engineering that will be "steamrolled" by next Nvidia architecture that will pose the basic path for all the next graphics evolution while already extending what is already available today thought Turing.
    AMD has still to put all those missing features, and it already has a really big transistor budget to handle today. 7nm, though by some revision, are here to stay for long time. If AMD is not going to change RDNA completely they won't be able to compete but by skipping the support of the more advanced features in the next years and are going to enjoy this match of the performances for just few months. Of course the missing features will be considered useless until they will eventually catch up. And they have still the console weapon to help them keep the market to a stall as they are quite behind with what the market can provide in the next years. RT is just the point of the iceberg. But also advanced geometry like mesh shading features that could already boos the scene complexity to the moon. But we just learnt that with NAVI AMD just managed to match Maxwell geometry capacity. Worthless piece of silicon, already discounted before launch.
  • Meteor2 - Tuesday, July 9, 2019 - link

    "In few words, AMD just used a lot of transistors and W just to get near Pascal efficiency." -- that makes no sense at all.

    Didn't bother reading the rest of your comment, sorry not sorry.
  • CiccioB - Wednesday, July 10, 2019 - link

    I just wonder what you have seen.
    NAVI gets the same perf/W that Pascal has and the same exact features.
    No RT, no tensor, No VSR, no geometry shading, no Voxel acceleration (that was already in Maxwell), no doble projection (for VR).
    7nm and 10 billions transistor to be just a bit faster than a 1080 that is based on a 5.7 billion transistor chip. And using more power do to so.

    Don't bother reading. It is clear you can't understand what's written.

Log in

Don't have an account? Sign up now