The next generation of AMD graphics cards are coming soon and it is rumored that they will finally have a real competitor to Nvidia's high-end graphics cards. We already know that AMD will provide the graphics performance for the upcoming Xbox Series X and Playstation 5. However, this architecture is also used in a high-end GPU that can compete with the RTX 3080 and 3090 recently introduced by Nvidia.
The architecture behind these new GPUs is known as RDNA 2, which is used in both these new PC graphics and the new consoles. AMD has proven to have the best performance per watt, as demonstrated by the success of the Radeon RX 5700 XT.
With the so-called "Big Navi", AMD will reportedly focus on achieving both better energy efficiency and performance improvements and adding support for ray tracing to its GPU product line for the first time. Here's everything we know so far.
Release date and price
AMD has started a countdown clock, and we can expect the company's next generation graphics cards to be unveiled on October 28th. The new Zen 3-based processors will be announced on October 8th.
The AMD # RDNA2 architecture and Radeon RX 6000 series graphics cards bring the best of Radeon to gamers worldwide. Find out more October 28th pic.twitter.com/CZJRxTBe6m
– Radeon RX (@Radeon) September 9, 2020
During AMD's Financial Analyst Day earlier this year, CEO Lisa Su promised that the next-generation graphics architecture will be ready by the end of 2020. The timeline fits well with Microsoft and Sony's plans for their respective consoles, as both hardware manufacturers have promised that the Xbox Series X and PlayStation 5 will be available for the 2020 holiday season.
This could give the company a significant boost in its intellectual stake as it seeks to fill the void with rival Nvidia. If AMD is able to ship on consoles, Big Navi could serve as a Halo product on that platform, which in turn will generate future sales for PCs.
Although AMD has been aggressively pricing in the past, PC gamers looking for a heavy discount on Big Navi may be disappointed. At press conferences earlier this year, Su had stated that AMD has slowly increased the prices of its GPUs over the past few years in an attempt to generate better margins, as AMD finds that its investments in graphics add significant value to gamers. We anticipate this will continue to be the case with Big Navi, especially as the demand for games and entertainment continues to grow in the face of new work, study, and games in home settings created by the global health pandemic.
As a reference, pricing for Nvidia's current flagship GPU, the Radeon VII, is very close to Nvidia's pricing. Benchmarks showed that the Radeon VII was very similar to the GeForce RTX 2080, valued at $ 699, and AMD's $ 679 price tag revealed that choosing Team Red did not give gamers a huge discount. Since Nvidia is pricing the RTX 3080 at $ 699, we expect the Radeon RX 6000 on PC to be just below that price.
Nvidia's aggressive price for the RTX 3080 and the card's seemingly strong performance will likely put pressure on AMD to either deliver a suitable performance – if Big Navi is to be positioned as a competitor to its competitor's Amp architecture – or at the price compete.
The RX 6000 brand confirmed
The trademark "6000" was discovered in Fortnite as an Easter egg. Gamer @MissGinaDarling posted on Twitter that she was taken to a secret Radeon room within the game with the special code "6000" that insiders speculated could be used in the final branding of the Radeon RX 6000. After this enthusiasm, Frank Azor, AMD's chief architect for gaming solutions, confirmed the Easter egg and indicated that further clues would come.
Hey @AMDGaming @AMDRyzen @AMD @sherkelman. Found this on Fortnite. What does that mean? pic.twitter.com/C3cpOTx8DU
– Gina (@MissGinaDarling) September 4, 2020
The AMD Radeon RX 6000 family line is expected to include multiple cards that include entry-level, mid-range, and premium GPUs. Like the 5000 series, AMD could offer an RX 6300 as the successor to the RX 5300 as an entry-level model and position its RX 6500 and RX 6300 as mid-range offerings to take over the GeForce RTX 3070 from Nvidia. On the premium end, AMD could replace the RX 5700 with a new RX 6700. Given that AMD had offered a premium Radeon RX 5700 XT and a Radeon XT 50th Anniversary Edition worth $ 499, the company could Offer high-end cards with a new RX 6900 model that will be designed to compete against Nvidia's GeForce RTX 3080 and RTX 3090.
AMD also offers discrete mobile graphics for laptops with models such as the RX 5300M, 5500M, 5600M, and 5700M. These chips are also available in Pro configurations for mobile workstations that are designed for business use. The RX 6000 series could similarly be available in similar configurations for use in laptops and mobile workstations for Nvidia RTX Studio laptops.
It's unclear what strategy AMD will be pursuing for the Mac, as Apple recently announced it would begin the transition from Intel processors to its own Apple Silicon for the Mac. The RX 5000 series is also available on the iMac as the Radeon Pro 5300, 5500XT, 5700, and 5700XT. When Apple's transition is complete, it will use its own integrated GPU solution on its custom ARM-based processors.
RX 6000 comes to APUs
According to David Kumar, CFO of AMD, Big Navi will be integrated into AMD's APUs as an integrated GPU solution in addition to premium GPUs and consoles. This is a move similar to what Intel is taking with its Intel Xe graphics architecture.
The company's Ryzen 4000 APUs with built-in Radeon graphics offer enough power to play many 1080p titles on the game's lowest graphics settings. This would allow simple desktops – and mobile laptops – to play basic games without requiring an expensive or bulky setup with discrete GPUs. When AMD compared its Ryzen 7 4700G desktop APU to the Intel Core i7-9700, it showed that the GPU performance was up to 274% higher than the competition.
The RDNA 2-based APUs will likely give AMD an extra boost, but Intel's new integrated Xe architecture on its 11th generation processors also brings a significant performance boost. Intel's integrated Xe graphics, installed on Tiger Lake-based laptops this fall, promises to be 2.7 times faster than content creation, more than 20% better performance for office tasks, and twice as much Performance versus previous generation 11 integrated GPU. Intel's Xe platform has 96 execution cores for a total of 768 cores with 3.8MB L3 cache.
Intel's 11th generation Tiger Lake will likely compete against AMD's Renoir and Nvidia's discrete and lightweight GeForce MX in gaming. Intel's own numbers show that Xe outperforms the competition, including the Ryzen 7 4800U. Given that Intel is ramping up AMD on integrated GPU performance, it will be interesting to see how a future Ryzen APU with integrated RDNA 2 graphics will compete with the new Xe architecture.
A recent tweet from @coreteks suggests that following Nvidia's announcement of the GeForce RTX 3000 series, AMD could position the Radeon RX 6000 with 16GB of RAM – almost 50% more than what's available on the flagship RTX 3080 – Priced between $ 499 and $ 599, where $ 549 is the likely cost of ownership for this card.
Since the new GDDR6X or G6X memory standard was developed by Micron in collaboration with Nvidia, it is used exclusively for the RTX 3080 and 3090. This means AMD will have to rely on its upcoming RDNA 2 card for the G6 non-X GDDR6 memory format. While you don't have that much bandwidth here, the new RDNA 2 cards could potentially come with even more video RAM – up to 16GB – which could offset some of the speed gains Nvidia has with its 10GB of RAM on the RTX 3080 For reference, the GDDR6X memory can run up to 760 GB per second, while the 8 GB GDDR6 memory of the Radeon RX 5700 XT 50th Anniversary Edition supports a maximum memory bandwidth of 448 GB per second.
It used to be rumored that AMD's new cards might be available in HBM or high bandwidth memory, but recent leaks show these plans have been abandoned. HBM's high cost would likely have contributed to higher retail costs, which AMD may want to avoid as it competes against Nvidia in both price and performance. It's still unclear what AMD plans for HBM, but including the feature could make a high-end RX 6000 card for avid gamers and creatives competitive against Nvidia's $ 1,499 ultra-premium RTX 3090.
Dan Baker / Digital Trends
Unlike the numerous leaks that occurred prior to the announcements of rival Nvidia's GeForce RTX 3000 series, the design information for the upcoming AMD Radeon RX 6000 cards has not surfaced until an official reveal. While Nvidia has chosen a more controversial design language for its RTX 3000 cards in order to better control the heat of its high-performance card, it is unclear what strategy AMD will pursue.
The company had already confirmed that it will move away from the blower heat sink this year in favor of a more attractive design. With more premium aesthetics, Team Red players can expect to spend more on RGB backlight kits so that Big Navi can shine as the centerpiece on gaming rigs with clear tempered glass panels. For reference, the RX 5000 series' metal-clad industrial aesthetic was a bit bland, and the one-fan design made the card look anemic.
With a more powerful offering this year, AMD could potentially add more fans to the card and give it a more aggressive design. However, it is unclear whether the map will also get bigger. By comparison, the RTX 3090, which Jensen Huang, CEO of Nvidia, referred to as the BFGPU, has a triple-slot design. In general, the new RTX series was introduced with a new 12-pin connector that is more compact and meets the new performance requirements of the cards.
Big Navi will use AMD's next generation GPU architecture core, which is an evolution of the RDNA architecture introduced on the Radeon RX 5000 series graphics cards. Prior to the launch of the Radeon RX 5000 series, AMD's older Vega graphics relied on the GCN platform. According to AMD executives, Big Navi's RDNA 2 architecture will provide a 50% per watt performance increase over RDNA, a similar improvement to the transition from GCN to RDNA.
This means that RDNA 2 is expected to use up to 33% less power while maintaining similar performance levels to RDNA, or it can perform 50% better for the same power consumption.
In 2020, AMD will likely only launch its RDNA 2 PC GPU in a reference design, which, according to TweakTown, has a new cooler design and offers better and faster performance. Unfortunately, there doesn't seem to be enough time for third-party cards, and these designs could be postponed until next year.
AMD's new GPU will likely be made with a more advanced 7nm + node. While Nvidia turned to Samsung's 8nm node for the RTX 3000 series, AMD's new RX 6000 series will likely be manufactured by partner TSMC. AMD previously announced that the new RDNA 2 will use a redesigned microarchitecture that offers improved performance per clock.
AMD is also expected to implement Microsoft's new DirectStorage feature on its new Radeon RX 6000 cards, which will allow GPUs and SSDs to communicate directly with each other to avoid loading times for games. Rival Nvidia demonstrated this feature by calling it RTX I / O on its new RTX 3000 series.
In addition to supporting Variable Rate Shading (VRS), which is part of the Microsoft DirectX 12 API under Windows, the card will also include ray tracing in the AMD product range for the first time. AMD will likely use its own optimizations and software to provide ray tracing for games. While the company hasn't released details on how to implement ray tracing, the company's early discussions about Big Navi suggest it will use the same Bounding Volume Hierarchy (BVH) approach as its rival.
However, early rumors suggest that ray tracing is not available to all members of the Big Navi or Navi 2x family, which could suggest that the coveted feature will only roll over to AMD's flagship cards. This could mean that the ray tracing feature may require more complex hardware that may not be found on low-end cards, and enabling the feature across the board could result in performance degradation that Nvidia may find with cards the entry-level RTX 2000 series.
With RDNA 2, AMD has the chance to take over Nvidia in the premium segment. While AMD has an excellent track record in the midrange market, the RDNA 2 architecture allows AMD to compete at any cost. While AMD's current offering for gamers is strong in the 1080p and 2K 1440p segments, Big Navi will bring "uncompromising 4K gaming" according to a slide the company has shown investors.
"Nice start from @Nvidia yesterday on their new graphics card," tweeted Azor. “They will combine well with our latest @AMDRyzen CPUs. Cant wait to show you all of the great products our @Radeon team has been working on! What a great year to be a gamer !!! "
Early rumors suggest that Big Navi could see up to a 30% increase in performance compared to Nvidia's Turing flagship, the GeForce RTX 2080 Ti, a $ 1,200 graphics card. Given the RTX 2080 Ti’s strong performance in 4K gaming, a 30% increase in performance compared to this card should give Big Navi enough headroom to handle ray tracing at higher resolutions. The introduction of the flagship RTX 3080 from Big Navi and Nvidia should lead to even greater demand for 4K games. As a result, we'll likely see more 4K game-ready monitors.
Given the 5,120 stream processors, which assume the same ratio as AMD at GCN, and a potential clock speed of 1,700 MHz, the performance of Navi 2x should be 17.5 TFLOPs. This is twice the number of stream processors available in the ultra-premium edition RX 5700XT 50th Anniversary.
The performance could be even higher if the clock speeds were higher or if AMD could push more stream processors onto the architecture. This compares favorably to the 12 TFLOPs Microsoft has promised for the upcoming console. The PlayStation 5 should have 11 TFLOPs of computing power and be supplied with 52 CUs that are clocked at 1,743 MHz.
Azor's comment could anticipate AMD's performance for the Radeon RX 6000, which according to Igor & # 39; s Lab and Hardware Times could match Nvidia's flagship RTX 3080 in performance.
The publication found that Big Navi with 300 watts is equivalent to the 320 watt RTX 3080 in terms of its performance. At 300 watts, RDNA 2 would correspond to the power consumption of the Radeon VII.
At 275 watts, Big Navi's performance is somewhere between that of the RTX 3070 and the RTX 3080. Unfortunately, AMD doesn't seem to have a product that can compete with Nvidia's ultra-premium RTX 3090, which can play 8K games at 60 FPS delivers.
According to Wccftech, Big Navi is expected to have a total of 5,120 cores. Twitter user @_rogame confirmed the number of cores in his own leak, adding that up to four shader engines, up to 2 shader arrays per engine, up to five WGPs or 10 compute units per shader array and up to four RBs should appear per shader engine. Overall, the tweet confirmed that 80 arithmetic units should be displayed on a card called NAVI21, or twice as many as on the RX 5700 XT 50th Anniversary Edition.