-
How many Radeon RX 6700XT graphics cards has AMD produced? Well, at least one.
Sam Machkovech
-
One HDMI 2.1 connection and three DisplayPort connections. Note that this lacks the USB Type-C connection found in its “Big Navi” siblings.
-
Power: one eight-pin, one six-pin.
-
Unscrew if you think you have a better idea for a back plate, I guess.
-
unpack.
-
Ready to unpack and install.
Look, I’ll agree with you: judging a GPU amid a global chip shortage is ridiculous enough to be considered dark comedy. Your ability to buy new, high-quality GPUs from Nvidia or AMD has been crippled for months — a fact confirmed by their very low ranking on Steam’s gaming PC stats collected around the world.
As of press time, AMD’s latest “Big Navi” GPUs are barely making a ripple in the Steam list. That’s arguably a matter of timing, with their launch in November 2020 two months after Nvidia started shipping its own 3000 series GPUs. But how much is that compounded by low inventories and shopping bots? AMD isn’t saying, and on the eve of the launch of the Radeon RX 6700, the first in its “Little Navi” line, the company’s guarantees aren’t entirely reassuring.
In an online press conference ahead of the launch, AMD product manager Greg Boots offered the usual platitudes: “a lot of demand out there,” “we’re doing everything we can,” that sort of thing. He mentioned a few steps from AMD that: be able to help this time. First, AMD’s “reference” GPU model is launching simultaneously with partner cards, so if the inventory is actually there (we don’t know for sure), that will at least put a higher number of GPUs in the day-one pool. Boots also emphasized that the stock will be made available specifically to brick-and-mortar retailers, though he didn’t give a ratio of the number of GPUs going to those stores compared to online retailers.
But then he offered an insurance policy that raised my eyebrows: “We’re working with retailers to prevent bots, like we did with the 6800.”
My apologies, what?!
AMD
I asked again, pointing to ample evidence that the combined launch of the 6800 and 6800XT was an online retail disaster, and Boots said he “couldn’t give too much detail about exactly what we’re doing.” Minutes later, one of his employees admitted otherwise: A memo that AMD had sent to online retailers had leaked onto Reddit. I couldn’t find the memo in question, but it turned out to contain advice on purchases by account and other standard anti-bot practices, as opposed to high-end trade secrets.
Moments later, Boots seemed to admit defeat: “At the end of the day [retail partners] are going to use it or not. That’s up to them.”
None of this makes AMD particularly special as a GPU manufacturer, mind you. Nvidia hasn’t been a glowing steward of bot prevention or GPU supply control either, and quick sell-out of its most recent GPU launch, February’s RTX 3060, proved that. In the case of both manufacturers, we’ll have to play this broken record of “can you really buy it?!” madness for any review until things change.
In particular, the typical “value for money” talk is currently out of the question. MSRP, or “manufacturer’s suggested retail price,” doesn’t mean it’s a nuisance in an eBay wasteland; your least favorite five-star retailer can’t give a damn about what AMD or Nvidia are suggesting, as long as desperate gamers and ethereum miners put the supply-demand curve in favor of scalpers.
Be sure to leave milk and CUDA cores
Despite this market reality, the MSRP at least gives us an idea of the intended range of each GPU, and the $479 RX 6700XT is the closest thing to Nvidia’s RTX 3060 Ti, “priced” at $399. Want to read a classic Sears catalog, dreaming that St. GPU Nicholas would one day fill your stocking with teraflops, this head-to-head showdown makes the most sense as a Team Red vs. Team Green – both focused on high-end 1440p performance with these products, not 4K.
AMD would probably prefer we compare the 6700XT to Nvidia’s more recent $329 offering, the RTX 3060, and if those are the only two cards in stock and we don’t measure retail prices, the conversation tilts further. in favor of AMD. As I reported last month, the 2019 RTX 3060 and RTX 2060 Super are neck-and-neck in terms of performance, and you’re more likely to have an RTX 2060 Super to compare this to, so I put it in set my benchmarks. My charts also include 2016’s GTX 1070 as a “baseline” card and last year’s AMD RX 6800. I’m using the latter to help you possibly measure the curve of what a hypothetical RX 6700 (not “XT”) might look like in terms of moving from one model to another.
-
The “1% low” bar in most of these charts represents the worst FPS you can expect 1% of the time. It’s perhaps the quickest way I can clarify whether a particular test or GPU/game combination comes with frametime spikes. The closer that 1% value gets to the average, the more stable the refresh rate.
-
This gallery is mostly captioned to make it easier to read the charts; if a caption gets in the way, click the gallery to make it full screen.
-
This “opening battle” is a very variable test where I play through an opening sequence. It’s less reliable than a solidly repeatable benchmark, but it’s here for your consideration of how the GPUs handle real-time gameplay.
-
-
-
-
-
-
-
-
-
-
New to our benchmark suite: Hitman 3, a slightly updated version of IO Interactive’s long-running Hitman engine. The new “Dubai” test mainly pushes GPUs with reflection effects.
-
-
Hitman 3’s Dartmoor test, meanwhile, is a beast of particle and tissue effects, and each GPU struggled with the workload in terms of holding a solid frame rate.
-
-
-
-
As a very, very unoptimized game, Cyberpunk 2077 lands here as a curio, and I chose not to bend the ray tracing or DLSS modes.
-
This is a brisk walk through the Novigrad section of the game. The 6700XT for some reason really struggled to lock in a consistent frame rate here.
All the benchmarks in this article were run on my standard Ars test rig, which uses an i7-8700K CPU (overclocked to 4.6GHz), 32GB DDR4-3000 RAM, and a mix of a PCI-e 3.0 NVMe drive and standard SSDs. has.
In general, tests of gaming software at 4K resolutions are best for demonstrating the full potential of a GPU for your system, even if you don’t plan on playing games at 4K resolutions. By setting a game’s resolution to 4K, it’s better for these measurements to take a CPU’s impact on a game out of the equation, then check the percentage differences between my tests and guess accordingly. The same goes for the overkill I hit on each benchmark’s settings, which I usually set to the highest possible or second-highest setting – and in your actual gameplay it’s generally better to turn off certain expensive effects, resulting in higher frame rates than seen here (often with negligible hits for how your favorite games look in action).
Still, AMD recently insisted that its GPUs are tuned to run at 1440p resolution, so I took the company’s reps at their word and included 1440p testing for every benchmark above (albeit with exaggerated graphics settings). . Sure enough, that sales pitch makes sense to some extent, especially when the 6700XT is directly compared to the 3060 Ti.