Radeon 7900 XTX and XT review: faster, faster and cheaper than the RTX 4080

NVIDIA RTX 4080 And the 4090 GPUs are an amazing performer. It’s also startlingly expensive, starting at $1,200 and $1,500 and going on road Even for cards from partners like MSI, Gigabyte, and Asus. The 4080 costs almost twice as much as the original $699 MSRP for the RTX 3080.

These price increases are due in part to pandemic-era concerns like a supply chain crisis and inflation and in part to a cryptocurrency-fueled boom (now over, thankfully) that has emboldened a network of speculators to snap up every high-end GPU. Also at play was competition’s lack Increased cost and complexity To build giant monolithic wafers on cutting edge manufacturing processes. Today, AMD is trying to solve the latter two problems by releasing the Radeon RX 7900 series of GPUs.

At $899 and $999, the RX 7900 XT and RX 7900 XTX are still objectively expensive — but since they’re not a further step up over the RX 6900 XT’s starting price, both cards are your passes for a good deal in the GPU market. today. If you’re looking for cards that can consistently handle 4K games at 60fps or higher, these GPUs do it for less than Nvidia’s latest cards, and they’re good enough and fast enough that hopefully they’ll start lowering Nvidia’s prices a bit, too. .

But Nvidia still retains some key advantages that complicate an easy David-and-Goliath narrative. These GPUs don’t seem like a Ryzen moment for AMD’s graphics division — a tipping point where a fickle AMD manages to make a huge dent in the market share of an established, satisfied competitor. But if you can actually find it for its starting prices, it’s the first sign we’ve had in a while that some relief is coming for high-end but price-conscious PC gamers.

Akbar Navi

The RX 7000 series is the third version of the RDNA GPU architecture, sometimes referred to as “Navi”, after the codenames of the GPU chips themselves. RDNA 3 doesn’t add anything that feels as significant as ray tracing support in RDNA 2, but AMD added a lot of additional hardware and made important changes under the hood.

Most important is a new chip-based approach, similar in concept to which AMD uses for its Ryzen CPUs. Instead of building the entire GPU die in a single manufacturing process—increasing die size and thus the chances of some or all of it being defective—AMD builds the main Navi 3 GPU die on TSMC’s 5nm manufacturing process and a series of smaller memory controller dies (MCDs) on the 6 process. nm. All of these chips are linked with a high-speed internal connection, which AMD says can transfer data at speeds of up to 5.3 terabytes per second.

The Graphics Compute Master Block (GCD) contains most of the hardware you think of when you think of a GPU – computation units, shaders, ray tracing devices, a media encoder and decoder block, and output rendering. The 7900 XTX and XT both use the same Navi 31 GCD, but the XTX runs at higher clocks and has more CUs and stream processors. The XTX has 96 CUs and 6144 stream processors, while the XT has 84 CUs and 5376 stream processors. Both cards represent a jump above the Navi 21 die used in the RX 6900 series, which maxed out at 80 CU and 5120 stream processors (and that’s before accounting for other performance-boosting improvements).

All MCDs include a single 64-bit memory controller and 16MB of AMD’s Infinity Cache, and they show the advantages of the chiplet-based approach. The 7900 XTX has a 384-bit memory bus and 96MB of Infinity cache, where the 7900 XT has a 320-bit bus and 80MB of cache; To do this, all AMD has to do is remove the MCD. The same MCDs can be reused up and down the stack with all the different RDNA 3 GCDs AMD chooses to launch, from low-end products with one MCD up to mid-range GPUs that use between two and four. Defects in MCD dies will not require larger and more complex GCDs to be scrapped or discarded, and vice versa.

In the world of RDNA 3’s all-new features, there are three things worth noting. First, the GPUs include new AI accelerators, which could be useful for the slew of AI-assisted content creation that came out last year and for AI-assisted scaling (if AMD chooses to implement it in some future version of the FSR upgrade algorithm). DLSS and XeSS both use AI for upscaling, but FSR 2.0 does not.

Secondly, the video encoding and decoding block supports hardware-accelerated encoding of AV1 video codecs, just like the RTX 4000 series and Intel Arc GPUs. This should come in handy for both content creators and streamers who want to stream HD video or video at the same resolution while using less bandwidth.

And third, the Radiance Display Engine adds DisplayPort 2.1 support for GPUs. Monitors that take full advantage of DisplayPort 2.1’s extra bandwidth don’t really exist as of this writing, but when they do, RDNA 3 GPUs will be able to power 4K displays at up to 480Hz and 8K displays at up to 165Hz.