AMD has revealed its first RDNA 3 graphics cards, the $1000 Radeon RX 7900 XTX and the $900 RX 7900 XT. As the numbering scheme and price points indicate, these high-end models are designed to compete with Nvidia’s RTX 4090 and 4080, but they come with more significant changes than you might expect – including FSR 3, designed to counter Nvidia’s DLSS 3-frame generation, and a completely new chiplet-based design.
Before we get into the features, let’s take a look at the cards themselves. The chiplet design breaks a traditional monolithic GPU into multiple interconnected sections. For the RDNA 3, it is a single 5nm graphics computer (GCD) that is 300mm² and six 6nm memory cache dies (MCDs) that are each 37mm². This design means that only the most critical areas need to be made using a cutting-edge 5nm process, helping to improve CPU yield, reduce costs and ultimately consumer prices. However, it also requires a fast interconnect between the various chips – which runs at 5.3TB/s here. The same chiplet approach worked brilliantly with Ryzen, transforming AMD from an also-ran to a growing giant, so it will be fascinating to see how it works miracles in the GPU space as well.
Each of the dies is impressive in its own right. The memory cache uses a 64-bit memory controller and a second-generation Infinity Cache, which together provide 2.7 times the peak bandwidth of RDNA2 designs. Meanwhile, the GPU offers unified RDNA 3 compute units with hardware for stream processing, AI acceleration and RT. The design also decouples shader and front-end clock speeds, with 2.3GHz for the former and 2.5GHz for the latter, which AMD says will result in a more efficient design – with up to 25 percent shader power savings and 15 percent higher front-end clock frequency.
RT has long been an AMD bugbear, so their second-generation solution that supports “1.5x more rays in flight”, “new dedicated instructions and “new ray box sorting and traversal” should result in up to 50 percent more performance per compute unit – although it doesn’t look like AMD is speeding up some parts of the RT pipeline like Nvidia is, so we might not see a huge jump in RT performance relative to raster performance that you’d otherwise expect.
The display engine built into these cards is pretty nifty, supporting DisplayPort 2.1 and up to 54 Gbps display link bandwidth, enabling 8K 165Hz (!) or 4K 480Hz (!!) with 12-bit color. Suffice it to say, we’re a long way from these types of displays, but it’s an effective counter to Nvidia’s 40-series cards, which are limited to DisplayPort 1.4. A dual-media engine, meanwhile, should shore up AMD’s lackluster reputation for streaming and media encoding, with AV1 encoding/decoding support, simultaneous encoding/decoding for AVC/HEVC and ‘AI Enhanced Video Encode’ which I look forward to hearing more about .
|Model||CUs||Game clock||VRAM||Mem. bus||Power of the board||Start MSRP|
|RX 7900 XTX||96||2.3 GHz||24 GB||384-bit||355W||999 USD|
|RX 7900 XT||84||2.0 GHz||20GB||320-bit||300W||$899 USD|
|RX 6950 XT||80||2.1GHz||16GB||256-bit||335W||$1299|
|RX 6900 XT||80||2.0 GHz||16GB||256-bit||300W||999 USD|
|RX 6800 XT||72||2.0 GHz||16GB||256-bit||300W||$649 USD|
As for the cards themselves, it looks like there is a big gap between the 7900 XTX and the 7900 XT. The amount of RAM is the most obvious difference – 24GB on the XTX and 20GB on the XT, with a correspondingly smaller memory bus (384-bit vs 320-bit). The nominal “game clock” also drops from 2.3GHz to 2GHz, while a drop from 96 to 84 compute units is also significant. However, both cards guzzle power compared to the likes of the RTX 4090, with a total card power of 355W for the XTX and 300W for the XT. Both cards support DisplayPort 2.1 and AV1 encoding/decoding.
In terms of expected performance, AMD provided frame rate data for the RX 7900 XTX, but only against the RX 6950 XT. Here they recorded a 50 to 70 percent performance improvement for the new generation card, with 1.5x in COD: MW2, Watch Dogs Legion, Resident Evil Village (RT) and Metro Exodus (RT), 1.6x in Doom Eternal ( RT), and 1.7x in Cyberpunk 2077. I expected to see more performance data than this – like comparisons with Nvidia cards, maybe? – but as always, we’ll have to wait until the cards reach reviewers to see how well these GPUs perform in real-world tests.
Finally, AMD announced FSR 3, promising up to double the frame rate of FSR 2. Based on this wording – and the presentation’s endnote referring to “Fluid Motion Frames” – it seems a pretty safe bet that this is frame generation a la DLSS 3. This technique slightly increases input lag, but greatly improves visual fluidity when AI-generated frames are inserted between “real” frames. Since it doesn’t appear that RDNA 3 hardware is used for this, the technology could be made available for older AMD GPUs as well – or even Nvidia/Intel models. That would give it a unique advantage over DLSS 3, which is exclusive to the expensive RTX 4090 (and its upcoming 40-series counterparts) for now. FSR 3’s release date was listed as ‘2023’, so presumably we’ll be seeing a lot more information about it in the coming months.
So – that was AMD’s RDNA 3 announcements! It will be fascinating to see how the new hardware fares as we approach the December 13th release date, as with an all-new architecture there’s room for significant performance improvements – and the potential for some interesting edge cases too.
AMD is certainly making the right noises to appeal to those turned off by Nvidia – significantly higher frame rates, future display standards, reasonable power targets and no 12-pin connectors – but performance and features also need to be on point.
What did you make of the announcements? Let us know in the comments below.
#AMD #announces #XTX #graphics #cards #FSR