AMD RX 6600XT review: A sad trombone noise of a “$379” 2021 GPU

In a traditional PC hardware cycle, AMD’s new RX 6600XT could have been a welcome stopgap for a budget audience. Over the years, we’ve regularly seen this kind of GPU from both major GPU manufacturers. Those companies regularly turn down some specs, repurpose sub-optimal chips, and get a moderately priced option to follow their biggest kahunas for anybody tiptoeing into solid 1080p or 1440p gaming options on PC.

Unfortunately, there’s nothing traditional about the latest traditional PC hardware cycle. Today’s supply-and-demand ecosystem of computer GPUs looks like something out of a terrifying Dario Argento film. The horrors that lurk in every shadow include chip shortages and bot-fueled scalper waves.

And that context really helps us frame the $379 RX 6600XT—an underpowered, overpriced, and downright disappointing GPU whose primary sales pitch is 1080p gaming. That category is famously CPU-limited, not GPU-limited, so this GPU’s mileage will truly vary based on your rig. In general, Nvidia’s RTX 3060 Ti (at only $20 more MSRP) wins handily, while Nvidia’s RTX 2060 Super (which launched for $399 in July 2019) is within shouting distance of this brand-new card. That latter yardstick in particular makes AMD’s newest product a hard GPU to recommend.

Any changes in cabin pressure yet?

This is arguably the loudest example yet of GPU manufacturers getting in on the sector’s recent price inflation, as opposed to letting the market do so. A $379 price tag is not a great cost for the 6600XT, but it’s not clear that a more appropriate MSRP, like $299 or $319, would lead to a fairer price reaching average customers. “Street value” is a real thing. When calculating actual market prices, for example, we’re closer to the $500 range for both aforementioned Nvidia GPUs.

As such, I once again have to smother this piece in reminders about general GPU availability. And, once again, AMD (like its competition Nvidia) isn’t forthcoming about how many of its new cards will go on sale later this week. The best data point we have is how far AMD trails behind Nvidia on the metric of installed GPUs in the wild. Steam’s publicly available hardware surveys point to Nvidia’s RTX 3000 cards landing inside far more Steam-connected PCs than AMD’s RDNA 2 cards. That could be because Nvidia has sold more cards, or at least sold more cards to active Steam gamers. Either way, I don’t have a firm indication that the 6600XT will dramatically change the availability story.

(Editor’s note: After saying so many similar things in so many recent GPU reviews, I feel like a flight attendant reading rote notes ahead of a flight. “Should the GPU market stabilize in the near future, please look to the exit doors at the front, sides, and rear of the PC marketplace.”)

A note on 6600XT’s 1080p emphasis

When it comes down to it, the RX 6600 is a peculiar GPU option for one main reason: its pledge of solid PC performance at 1080p resolution, as opposed to 1440p or higher. AMD swears that this differential bears out in its own cherry picked test results, promising statistically significant 1080p gains over Nvidia’s RTX 3060 (MSRP: $329). Its charts include tests run on a Ryzen 5 5600X system, where the RX 6600 beat the cheaper Nvidia card with games set to 1080p resolution and identical “maxed out” visual settings at anything from 5 percent to 30 percent.

However, Ars favors 4K resolutions for GPU tests, as anything lower can contribute more CPU-bound limitations to resulting frame rate charts. Our measurements are meant to offer you percentages of difference that you can then expect to scale to your own PC. When AMD’s RDNA 2 line began promising significantly better performance at 1440p over Nvidia’s competition—and claimed this was an intentional engineering direction on their part—we ran resulting tests. Sure enough, we saw significant percentage gains that weren’t necessarily CPU bound.

But dropping all the way to 1080p with a testing rig that maxes out at an Intel i7-8700K GPU means we can’t reliably duplicate AMD’s own 1080p performance-boost sales pitch. Our percentages in those tests vary with both higher and lower results based on the game, and these games are stretching into 120 fps territory and above at that point, which is typically the threshold where a CPU begins to top out. Our 1080p charts are therefore, in our opinion, misleading enough for average customers to not be worth factoring into our usual comparison matrix.

I’m not printing AMD’s exact numbers because we cannot verify them, and because their own 1080p tests weren’t also conducted on the $399 RTX 3060 Ti, nor on 2019’s RTX 2060 Super. Having already written my own negative review of the RTX 3060, I understand why AMD might be tempted to have the press directly compare the RX 6800 to that card. However, that card’s MSRP is roughly 15 percent cheaper—and AMD’s list of 1080p results for that card are, whaddya know, roughly 15 percent worse.