Welcome to the Red Team. That’s the message that initially welcomes you when you raise the cover on among AMD’s brand-new Radeon 6000 series graphics cards, and it’s fitting – after all, AMD’s fans are a few of the most die-hard in video gaming, and these ‘Big Navi’ GPUs have actually been expected with nearly messianic fervour by the neighborhood at big. Thankfully, all of that favorable, memeable energy has actually discovered a worthwhile target in the RX 6800 and RX 6800 XT, 2 of the greatest developments by AMD’s Radeon graphics group in a long time.

Before we talk about how well these cards carry out in a large choice of video games – and do not stress, there will be plenty of that – let’s briefly set some expectations. At $580 (£530) and $650 (£600), respectively, the RX 6800 and RX 6800 XT slot in above the $500 (£450) RTX 3070 and listed below the $700 (£650) RTX 3080. That recommends we’ll see an RTX 3070-beating efficiency out of the RX 6800, while the RX 6800 will succeed to draw with the higher-priced Nvidia card. More notably, it likewise suggests that AMD is lastly completing at the luxury of the graphics market as soon as again, something the business hasn’t done because the unfortunate Radeon 7 nearly 2 years ago – and something it hasn’t done effectively for much more years than that.

As well as much enhanced out-and-out efficiency, these Big Navi cards likewise mark the PC launching of AMD’s RDNA2 architecture, which likewise forms the basis of both the PlayStation 5 and Xbox Series X|S consoles. Like those next-gen devices, the RX 6000 series supports hardware-accelerated ray tracing, lastly ending Nvidia’s RT monopoly, plus a raft of other DirectX 12 Ultimate functions like DirectStorage, variable rate shading and mesh shaders that can enhance efficiency in myriad methods. And speaking of efficiency improving, AMD’s engineers have actually likewise included functions like ‘Smart Memory Access’, ‘Rage Mode’ overclocking and an ‘Infinity Cache’ that ought to all conspire to press these cards to their limitations.

From this overwelming description alone you can see that there’s plenty to cover here, so let’s start with a better take a look at each of these cards – how are they constructed, how does the hardware accumulate and how does it connect into the brand-new functions available here?

AMD GPU Specs CUs Boost Clock VRAM Mem Interface TDP Price
RX 6900 XT 80 2250MHz 16GB GDDR6 256-bit + 128MB IC 300W $999
RX 6800 XT 72 2250MHz 16GB GDDR6 256-bit + 128MB IC 300W $649
RX 6800 60 2105MHz 16GB GDDR6 256-bit + 128MB IC 250W $579
RX 5700 XT 40 1905MHz 8GB GDDR6 256-bit 225W $399
RX 5700 36 1725MHz 8GB GDDR6 256-bit 180W $349
RX 5600 XT 36 1560MHz 6GB GDDR6 192-bit 150W $279
RX 5500 XT 22 1845MHz 8GB GDDR6 128-bit 130W $199

While specifications sheets can be a bit dull, taking a look at this one informs you rather a lot this time around. While AMD are running with the very same 7nm TSMC procedure as the initial ‘Little’ Navi style, their brand-new architecture handles to wring out a lot more efficiency. This is driven by considerably greater clock speeds throughout the board and a bigger enhance of more separately performant calculate systems. There’s likewise devoted ray tracing hardware backed into each CU, which is 10 times much faster at computing ray crossways than a simply software application method – and it’ll be remarkable to see how it compares to Nvidia’s second-generation RT hardware.

These calculate enhancements are backed with an ingenious brand-new memory option. By matching the 16GB of GDDR6 video memory with a 128MB cache, comparable in style to the L3 cache discovered on AMD’s Ryzen processors, AMD have the ability to attain some blistering arise from the card’s reasonably pedestrian 256-bit bus – in their slides, the system supplies 2.17x the bandwidth you’d get out of a basic 256-bit user interface. This increase in memory bandwidth assists balance out the higher calculate system (CU) count and greater clock frequencies, while keeping power usage under control. It’s a smart relocation that suggests that Nvidia’s benefit from its early adoption of (pricey) GDDR6X memory on the RTX 3080 is at least partly ameliorated. It likewise suggests that AMD can pay for to use 16GB of VRAM on both designs, something that will no doubt salve the nerves of players fretted about lacking video memory with a raft of next-generation video games in the pipeline. This 16GB of VRAM can likewise be accessed quickly on Ryzen 5000 systems by AMD’s Smart Access Memory tech, a performance-enhancing function we’ll cover in higher depth on page 6 of this evaluation.

That almost finishes up the architectural enhancements inside, so what about the commercial style outside?

While customized RX 6800 and RX 6800 XT cards will be produced by a vast array of AMD partners, today’s evaluation centres on AMD’s recommendation designs. While these GPUs are set to retail at the base RRP of each card – $580 for the RX 6800 and $650 for the RX 6800 XT – they seem like anything however entry-level styles, with an excellent three-fan axial cooler, lots of metal fins to dissipate heat and a more intricate aluminium shroud than the blocky Radeon 7. Overall, AMD’s Radeon 6000 cards are weighty and feel extremely well built, without any provide or bend apparent anywhere. The vanilla RX 6800 feels especially thick, with a more compact two-slot style, while the 2.5-slot style of the RX 6800 XT supplies more space for the higher-powered card to breathe. Both styles are the very same length and height, at 267mm (10.5″) long by 120mm (4.7″) high, so they ought to suit any basic ATX case without problems.

These styles are a substantial advance from the blower-style RX 5700 series cards, and ought to carry out far better in basic PC cases with appropriate air flow. I especially delighted in the no RPM fan mode, which appears to engage from around 52 degrees celsius and cuts the fans totally for utter silence throughout regular daily computing jobs. This is far from the very first GPU to support such a function obviously, however it is uncommon to see it on a first-party style and it speaks with the quality of AMD’s cooling option that it’s even useful to consist of. Throughout my screening, the cards ran peaceful, with the ‘well balanced’ power profile we utilized for screening topping the fan speed at simply 50 percent of their ~2000RPM optimum. We’ll leave it to the Buildzoids and Gamers Nexuses of the world to do a much deeper dive into the internals, however the 15 power stages and 14-layer PCB sure noise outstanding – and we’ll take a better take a look at power use later this extremely page.

In regards to ports, we’re taking a look at a somewhat various clothing as Nvidia’s RTX 30-series cards: one HDMI 2.1 port (ideal for those next-gen 4K 120Hz or 8K 60Hz Televisions you may have gotten to utilize with a brand-new console) plus 2 DisplayPort 1.4 ports and one USB-C port (which obviously can be utilized as a 3rd DisplayPort, HDMI and even DVI-D with the best adapter). As with the Nvidia cards, AV1 decoding is consisted of which ought to enable websites like Netflix and Twitch to stream high-quality, high frame-rate video material to you while needing considerably lower bandwidth than present codecs. Finally, the card is PCIe 4.0 suitable however works simply great in PCIe 3.0 slots too. For power, you get 2 eight-pin inputs on either card, with AMD advising a 650W PSU for the 6800 and a 750W PSU for the 6800 XT – rather affordable for mid-range to high-end graphics hardware.

To get a much better judgement of Big Navi’s power effectiveness while video gaming, we utilized Nvidia’s (agnostic) PCAT system – an interposer board that sits in between the graphics card and its power inputs to determine real power usage. When twinned with in-game efficiency metrics, like frame-time and frame-rate, we can compute precisely the number of watts are needed to render each frame, and thus get a step of real-world effectiveness. AMD point out a 54 percent enhancement in efficiency per watt in between RDNA and RDNA2, so is that substantiated by screening? And how does Nvidia’s Ampere architecture, our present effectiveness champ, compare?

Joules Per Frame RX 6800 XT RX 6800 RTX 3080 FE RTX 3070 FE RTX 2070 RX 5700XT
Death Stranding 2.933 2.644 3.349 2.915 3.651 3.846
Percentage Diff 110.9% 100% 126.6% 110.2% 138.1% 145.5%
Joules Per Frame RX 6800 XT RX 6800 RTX 3080 FE RTX 3070 FE RTX 2070 RX 5700XT
Gears 5 4.384 3.792 4.156 3.603 4.734 5.403
Percentage Diff 121.7% 105.2% 115.3% 100% 131.4% 150.0%

We see that in Death Stranding, AMD’s RDNA2 architecture is somewhat more effective than Nvidia’s Ampere, and considerably much better than first-gen RDNA (represented by the RX 5700 XT), coming close to AMD’s declared effectiveness gains. However, this video game is extremely kind to AMD hardware, so what about a more well balanced choice?

In Gears 5, the 3070 supplies much better competitors versus the RX 6800 and shows to be the somewhat more effective card, although the margin is simply 5 percent in Nvidia’s favour. Clearly, both brand-new architectures are considerably more power-efficient than their predecessors, which speaks with the engineering that has actually entered into each choice.

Interestingly, the RX 6800 XT needs in between 15 and 20 percent more joules per frame in the 2 video games we checked, so the more compact 6800 is absolutely the series leader in regards to power effectiveness. We saw a comparable margin in between the 3070 and 3080, so this is barely unforeseen.

With our short power screening concluded, it’s time to move onto the video game criteria. You’ll see a couple of brand-new titles here if you have actually not taken a look at our RTX 30-series examines yet, along with a brand-new test platform. If you have actually checked out those evaluations – thank you! – then do not hesitate to avoid to the next page as there’s absolutely nothing brand-new here.

We checked AMD’s Ryzen 3950X as a prospect for our GPU test bench construct, as this permits access to the complete PCIe 4.0 bandwidth supported by AMD’s 6000 series and Nvidia’s 30-series, however we ultimately decided on an Intel Core i9 10900K for our test system, changing the Core i7 8700K we utilized for our previous generation tests. Though the Intel system is technically constrained to PCIe 3.0 bandwidth, hence possibly restricting efficiency in GPU-bound circumstances, we still got greater 4K frame-rates with the Intel chip than with the fastest Ryzen processor offered at the time. The Intel system revealed a more substantial frame-rate lead at 1080p and 1440p, hence making it our best option for preventing CPU bottlenecking as much as possible to enable various graphics cards to separate themselves. We’ll think about Ryzen 5000 or its followers the next time we upgrade our test bench, however for now it makes good sense to stay with it so we do not need to retest lots of graphics cards in the most hectic duration of the year (and throughout a brand-new console cycle).

To guarantee our outcomes are similar, the Core i9 10900K is locked to a 5GHz all-core frequency and cooled by a 240mm Alpacool Eisbaer Aurora AiO which keeps the overclocked system at around 75C under complete load. The 10900K is backed by an Asus Maximus 12 Extreme Z490 motherboard and 2 8GB sticks of G.Skill Trident Z Royal 3600MHz CL16. Our video games are ranged from a capacious 2TB Samsung 970 Evo Plus NVMe drive provided by Box. The entire rig is powered by a 850W gold-rated Gamer Storm power supply.

Now that you recognize with our test bed and have an excellent introduction of the cards themselves, let’s enter into some video gaming tests to see how they carry out.

AMD Radeon RX 6800 and RX 6800 XT Analysis