AMD Radeon RX 6000 Series vs 5000 Series: Should You Upgrade?

Updated 12th November: With the Ryzen 5000 CPUs benchmarked, we cannot wait to get our hands on the RX 6000 GPUs to see what the full potential behind Smart Access Memory truly is.

Check out our Where to Buy pages to get your hands on the latest GPUs from AMD:

• Where to Buy 6900 XT
• Where to Buy 6800 Series
• Where to Buy 6700 Series

So the big reveal has come and gone, the RX 6000 series of GPUs has finally arrived, and many are wondering if they should pull the trigger and upgrade from their existing RX 5000 series setups.

The key thing to note is who these new cards are aimed at. Previously AMD’s target market was the middle-to-low range, where they did a good job of competing against Nvidia, leaving their rival to almost monopolize the high-end enthusiast market in recent years. The 6800, 6800 XT, and 6900 XT however are direct competitors to Nvidia’s RTX 3070, RTX 3080, and RTX 3090 respectively. At this stage, there have been no cheaper cards announced by AMD in the 6000 series, which makes comparing them against the 5000 series somewhat difficult. Nevertheless, there are certain circumstances where upgrading your 5000 series card to the 6000 series would be recommended.

Without further ado let’s walk through the main differences between the two generations and see which option is best for you.

The biggest omission from the 6000 series reveal was the lower price entries in the series, though this isn’t entirely surprising. It is normal for both AMD and Nvidia to not release the cheaper cards in each new series until a while later, as to encourage sales of the more expensive offerings beforehand. If we take a look at the 5000 series, for example, the 5700 XT, 5700 XT Anniversary Edition, and 5700 cards were the first released on July 7th, 2019, followed by the 5500 and 5300 XT on October 7th, 2019, the 5500 XT on December 12th, 2019, and the Radeon 5600 and 5600 XT on January 21st, 2020. 

We can expect a similar staggering of releases for the 6000 series, and if history is a guideline, news about any cheaper entries towards the end of this year or the beginning of 2021. If the current models of the 6000 series GPUs are out of your price range, we would recommend considering holding fire until then.

Whether or not these anticipated (though not confirmed) lower entries in the series will actually be worth buying is of course uncertain at this juncture. The 5700 XT was so successful because it offered a $100 cheaper alternative to Nvidia’s 2070 Super for roughly the same performance, besides the omission of ray tracing technology (though at this point few games supported it so it wasn’t much of a factor). The later entry of the 5500 XT however was more or less the opposite story – Nvidia’s RTX 1600 series were superior cards overall, meaning most considered the AMD card to be relatively pointless. Hopefully, any further entries in the 6000 series will be competitive enough to justify their purchase.

RDNA is the codename for the architecture of AMD’s 5000 series of GPUs which was generally considered to be a bit of a flop on release. RDNA 2 is the new version of the architecture for the 6000 series and looks to be a considerable improvement. As with their Ryzen processors, AMD have kept the same 7nm process but have managed to squeeze out great efficiencies and performance improvements from their latest generation of GPUs. 

First and foremost among these improvements is the “Infinity Cache” which is based on the same technology as the large L3 cache on their Ryzen 5000 CPUs. According to AMD, the Infinity Cache allows the memory bus to operate at a bandwidth 2.17x greater than you would normally expect a 256-bit memory bus to be capable of, reducing DRAM bottlenecks and latency. This is one of the reasons the 6000 series should outperform dramatically on 4K and other high-resolution screens, where bottlenecking can lead to low minimum FPS – no matter the clock speed of a graphics card, if the memory bandwidth isn’t sufficient this can cause stuttering in gaming as the frame rate noticeably fluctuates.

Other improvements in the manufacturing process have also allowed AMD to get 30% more energy efficiency out of their compute units, meaning greater clock speed for the same power expenditure and heat generation. All of these improvements combined are expected to lead to a 54% uplift in performance-per-watt between the 5000 series and the 6000 series.

RDNA is the codename for the architecture of AMD’s 5000 series of GPUs which was generally considered to be a bit of a flop on release. RDNA 2 is the new version of the architecture for the 6000 series and looks to be a considerable improvement. As with their Ryzen processors, AMD have kept the same 7nm process but have managed to squeeze out great efficiencies and performance improvements from their latest generation of GPUs. 

First and foremost among these improvements is the “Infinity Cache” which is based on the same technology as the large L3 cache on their Ryzen 5000 CPUs. According to AMD, the Infinity Cache allows the memory bus to operate at a bandwidth 2.17x greater than you would normally expect a 256-bit memory bus to be capable of, reducing DRAM bottlenecks and latency. This is one of the reasons the 6000 series should outperform dramatically on 4K and other high-resolution screens, where bottlenecking can lead to low minimum FPS – no matter the clock speed of a graphics card, if the memory bandwidth isn’t sufficient this can cause stuttering in gaming as the frame rate noticeably fluctuates.

Other improvements in the manufacturing process have also allowed AMD to get 30% more energy efficiency out of their compute units, meaning greater clock speed for the same power expenditure and heat generation. All of these improvements combined are expected to lead to a 54% uplift in performance-per-watt between the 5000 series and the 6000 series.

Other improvements in the manufacturing process have also allowed AMD to get 30% more energy efficiency out of their compute units, meaning greater clock speed for the same power expenditure and heat generation. All of these improvements combined are expected to lead to a 54% uplift in performance-per-watt between the 5000 series and the 6000 series.

Other improvements in the manufacturing process have also allowed AMD to get 30% more energy efficiency out of their compute units, meaning greater clock speed for the same power expenditure and heat generation. All of these improvements combined are expected to lead to a 54% uplift in performance-per-watt between the 5000 series and the 6000 series.

Software and drivers have been one area that AMD has lagged behind Nvidia in the past. Whilst it seems Nvidia still maintains this lead, based on the reveal of the new 6000 series this gap seems to be narrowing.

Unlike the older generation, the 6000 series is compliant with “DirectX 12 Ultimate” which should offer much more futureproofing for the latest graphical effect options we can expect to see crop up in AAA games in years to come, especially compared to the 5000 series which do not have DX 12 Ultimate support. Specific supported graphical features for the 6000 series include mesh shading, variable-rate shading, and sampler feedback as well as Microsoft’s DirectStorage tech that reduces loading times. “Radeon Boost” and “Radeon Anti-Lag” technologies should also increase response time and reduce latency for the new cards.

A big addition for the later cards is the inclusion of Ray-tracing support. Although early third party testing indicates that Nvidia’s version of the technology could be 33% – 59% more effective than AMD’s, some Ray-tracing is still a lot better than zero Ray-tracing, and this technology is completely absent from the 5000 series. In a similar vein, DLSS (Deep Learning Super Sampling) remains an exclusive proprietary technology of Nvidia’s, which the new AMD cards lack. AMD did mention they are working on their own version of this tech, which essentially uses Artificial Intelligence to assist in upscaling images to high resolutions, but we do not know when this so-called “Super Resolution”, will be available, or how effective it will prove – either way, this is another advantage the new graphics cards will have over the older AMD generation.

Finally, the two features of “Rage Mode” and “Smart Access Memory” (SAA) were the big talking points of the reveal. Further information about these two features can be found here, but in essence, Rage Mode is a one-click overclocking ability the new cards will come with and SAA, the more interesting of the two, is a way for the new Ryzen CPUs to interact more directly with the memory of the graphics card, improving performance. For people with Ryzen 5000 CPUs, who also have the necessary 500 series motherboard, the uplift of this feature could mean between 2% – 13% FPS performance increase for the RX 6800 XT for example, over what the card would be capable of on a different setup. Whilst we do not know how this additional FPS performance would be replicated across any lower entry-point cards which may be revealed for the 6000 series going forward, we can definitely expect them to have this same technology. The extra bang for buck that this could bring if you intend to purchase one of the new Ryzen CPUs could make an upgrade to a 6000 series Ryzen that much more tempting.

Software and drivers have been one area that AMD has lagged behind Nvidia in the past. Whilst it seems Nvidia still maintains this lead, based on the reveal of the new 6000 series this gap seems to be narrowing.

Unlike the older generation, the 6000 series is compliant with “DirectX 12 Ultimate” which should offer much more futureproofing for the latest graphical effect options we can expect to see crop up in AAA games in years to come, especially compared to the 5000 series which do not have DX 12 Ultimate support. Specific supported graphical features for the 6000 series include mesh shading, variable-rate shading, and sampler feedback as well as Microsoft’s DirectStorage tech that reduces loading times. “Radeon Boost” and “Radeon Anti-Lag” technologies should also increase response time and reduce latency for the new cards.

A big addition for the later cards is the inclusion of Ray-tracing support. Although early third party testing indicates that Nvidia’s version of the technology could be 33% – 59% more effective than AMD’s, some Ray-tracing is still a lot better than zero Ray-tracing, and this technology is completely absent from the 5000 series. In a similar vein, DLSS (Deep Learning Super Sampling) remains an exclusive proprietary technology of Nvidia’s, which the new AMD cards lack. AMD did mention they are working on their own version of this tech, which essentially uses Artificial Intelligence to assist in upscaling images to high resolutions, but we do not know when this so-called “Super Resolution”, will be available, or how effective it will prove – either way, this is another advantage the new graphics cards will have over the older AMD generation.

Finally, the two features of “Rage Mode” and “Smart Access Memory” (SAA) were the big talking points of the reveal. Further information about these two features can be found here, but in essence, Rage Mode is a one-click overclocking ability the new cards will come with and SAA, the more interesting of the two, is a way for the new Ryzen CPUs to interact more directly with the memory of the graphics card, improving performance. For people with Ryzen 5000 CPUs, who also have the necessary 500 series motherboard, the uplift of this feature could mean between 2% – 13% FPS performance increase for the RX 6800 XT for example, over what the card would be capable of on a different setup. Whilst we do not know how this additional FPS performance would be replicated across any lower entry-point cards which may be revealed for the 6000 series going forward, we can definitely expect them to have this same technology. The extra bang for buck that this could bring if you intend to purchase one of the new Ryzen CPUs could make an upgrade to a 6000 series Ryzen that much more tempting.

The 6000 series is AMD’s first real foray into the world of 4K gaming, and the jump in performance from the previous generation, according to AMD’s figures, is approximately double. The RX 5700 XT, the biggest hitter of the 5000 series, was the card used by AMD in the benchmarking, though they didn’t specifically mention which 6000 series card they were comparing it against.

Ultimately the 5000 series was only ever really meant to be used for 1440p displays; though 4K use was possible, it was never going to compete against Nvidia’s higher offerings. The aforementioned Infinity Cache hardware improvements and future support of Super-Resolution driver technology should mean that even if the new 6000 series cannot beat Nvidia in the 4K and 8K arenas across the board, they can certainly offer a challenge. If you have 4K gaming in mind, then this is an area in which the choice to upgrade to the 6000 series seems clear, even at this stage.

The 6000 series is AMD’s first real foray into the world of 4K gaming, and the jump in performance from the previous generation, according to AMD’s figures, is approximately double. The RX 5700 XT, the biggest hitter of the 5000 series, was the card used by AMD in the benchmarking, though they didn’t specifically mention which 6000 series card they were comparing it against.

Ultimately the 5000 series was only ever really meant to be used for 1440p displays; though 4K use was possible, it was never going to compete against Nvidia’s higher offerings. The aforementioned Infinity Cache hardware improvements and future support of Super-Resolution driver technology should mean that even if the new 6000 series cannot beat Nvidia in the 4K and 8K arenas across the board, they can certainly offer a challenge. If you have 4K gaming in mind, then this is an area in which the choice to upgrade to the 6000 series seems clear, even at this stage.

Besides comparing the RX 5700 XT against the newer generation in 4K FPS performance (as shown above) there weren’t any additional comparisons between the 5000 and 6000 series on a game-by-game basis in AMD’s presentation. Whilst we cannot look at any proper 1440p or 1080p benchmark comparisons until independent testers get their hands on the new GPUs, we can draw some conclusions from the benchmarking we do have. 

The lowest entry of the new 6000 series, the RX 6800, was shown against the Nvidia RTX 2080 Ti in a series of games, with an FPS increase across the board varying from slight to substantial (albeit with Smart Access Memory enabled). We know that vs the 5700 XT, the most powerful of AMD’s 5000 series, the RTX 2080 Ti shows an FPS uplift of approximately 25-30%. This illustrates the substantial difference in performance between the 6000 series currently available and the 5000 series, which are ultimately aimed at different price points and target markets. Ultimately, direct benchmarking of the RX 5000 cards against the RX 6000 series are not that useful until lower-end RX 6000 cards are announced which will present a more comparable offering.

Besides comparing the RX 5700 XT against the newer generation in 4K FPS performance (as shown above) there weren’t any additional comparisons between the 5000 and 6000 series on a game-by-game basis in AMD’s presentation. Whilst we cannot look at any proper 1440p or 1080p benchmark comparisons until independent testers get their hands on the new GPUs, we can draw some conclusions from the benchmarking we do have. 

The lowest entry of the new 6000 series, the RX 6800, was shown against the Nvidia RTX 2080 Ti in a series of games, with an FPS increase across the board varying from slight to substantial (albeit with Smart Access Memory enabled). We know that vs the 5700 XT, the most powerful of AMD’s 5000 series, the RTX 2080 Ti shows an FPS uplift of approximately 25-30%. This illustrates the substantial difference in performance between the 6000 series currently available and the 5000 series, which are ultimately aimed at different price points and target markets. Ultimately, direct benchmarking of the RX 5000 cards against the RX 6000 series are not that useful until lower-end RX 6000 cards are announced which will present a more comparable offering.

Bringing all this together then, we can certainly say that the 6000 series offers substantial improvements over the 5000 series cards, at both an architectural level and in terms of the accompanying software and driver support. If you are looking for an AMD card that performs well at 4K resolutions then the RX 6800, RX 6800 XT, and 6900 XT are pretty much the only game in town. Similarly, if you are planning on purchasing one of the new Ryzen 5000 series CPUs from AMD anyway, then it would seem the additional performance synergies offered by Smart Access Technology would make it well worth considering upgrading to one of the new Radeon RX 6000 series, assuming your wallet can stretch that far.

More uncertain at this stage however is if you are not really interested in 4K performance or if you are looking to upgrade from a medium or lower-range AMD card (which includes basically the entire 5000 series, even the RX 5700 XT) to one at a similar price point. The three AMD 6000 cards currently available are geared towards challenging Nvidia’s most powerful cards, not the middle market, and the cheapest of them (the RX 6800) has an expected RRP of $579. If this is out of your price range then we would advise waiting until early 2021, when we can expect to have a better idea as to AMD’s plans (if any) for cheaper entries in the 6000 series. As ever, we will keep you updated with the latest news on any new releases for the AMD 6000 series and independent benchmarking as it comes available for each of the cards, as well as the best Black Friday deals. Watch this space!