Thursday, November 3rd 2022
AMD Announces the $999 Radeon RX 7900 XTX and $899 RX 7900 XT, 5nm RDNA3, DisplayPort 2.1, FSR 3.0 FluidMotion
AMD today announced the Radeon RX 7900 XTX and Radeon RX 7900 XT gaming graphics cards debuting its next-generation RDNA3 graphics architecture. The two new cards come at $999 and $899—basically targeting the $1000 high-end premium price point.
Both cards will be available on December 13th, not only the AMD reference design, which is sold through AMD.com, but also custom-design variants from the many board partners on the same day. AIBs are expected to announce their products in the coming weeks.
The RX 7900 XTX is priced at USD $999, and the RX 7900 XT is $899, which is a surprisingly small difference of only $100, for a performance difference that will certainly be larger, probably in the 20% range. Both Radeon RX 7900 XTX and RX 7900 XT are using the PCI-Express 4.0 interface, Gen 5 is not supported with this generation. The RX 7900 XTX has a typical board power of 355 W, or about 95 W less than that of the GeForce RTX 4090. The reference-design RX 7900 XTX uses conventional 8-pin PCIe power connectors, as would custom-design cards, when they come out. AMD's board partners will create units with three 8-pin power connectors, for higher out of the box performance and better OC potential. The decision to not use the 16-pin power connector that NVIDIA uses was made "well over a year ago", mostly because of cost, complexity and the fact that these Radeons don't require that much power anyway.
The reference RX 7900-series board design has the same card height as the RX 6950 XT, but is just 1 cm longer, at 28.7 cm. It is also strictly 2.5 slots thick. There's some white illuminated elements, which are controllable, using the same software as on the Radeon RX 6000 Series. Both cards feature two DisplayPort 2.1 outputs, one HDMI 2.1a and one USB-C.
This is AMD's first attempt at a gaming GPU made of chiplets (multiple logic dies on a multi-chip module). The company has built MCM GPUs in the past, but those have essentially been the GPU die surrounded by HBM stacks. The new "Navi 31" GPU at the heart of the RX 7900 XTX and RX 7900 XT features seven chiplets—a central large graphics compute die (GCD), surrounded by six memory control-cache dies (MCDs). The GCD is built on the TSMC 5 nm EUV silicon fabrication process—the same one on which AMD builds its "Zen 4" CCDs—while the MCDs are each fabricated on the TSMC 6 nm process.
The GCD contains the GPU's main graphics rendering machinery, including the front-end, the RDNA3 compute units, the Ray Accelerators, the display controllers, the media engine and the render backends. The GCD physically features 96 RDNA3 Unified Compute Units (CUs), for 6,144 stream processors. All 96 of these are enabled on the RX 7900 XTX. The RX 7900 XT has 84 out of 96 unified compute units enabled, which works out to 5,376 stream processors. The new RDNA3 next-generation compute unit introduces dual-issue stream processors, which essentially double their throughput generation-over-generation. This is a VLIW approach, AMD does not double the rated shader count though, so it's 6144 for the full GPU (96 CU x 64 shaders per CU, not 128 shaders per CU).
Each of the six MCDs contains a 64-bit wide GDDR6 memory interface, and 16 MB of Infinity Cache memory. Six of these MCDs add up to the GPU's 384-bit wide memory interface, and 96 MB of total Infinity Cache memory. The GCD addresses the 384-bit wide memory interface as a contiguous addressable block, and not 6x 64-bit. Most modern GPUs for the past decade have had multiple on-die memory controllers making up a larger memory interface, "Navi 31" moves these to separate chiplets. This approach reduces the size of the main GCD tile, which will help with yield rates. The Radeon RX 7900 XTX is configured with 24 GB of GDDR6 memory across the chip's entire 384-bit wide memory bus, while the RX 7900 XT gets 20 GB of GDDR6 memory across a 320-bit wide memory bus (one of the MCDs is disabled). The disabled MCD isn't not "missing", but there's some dummy silicon dies there to provide stability for the cooler mounting.
Each CU also features two AI acceleration components that provide a 2.7x uplift in AI inference performance over SIMD, and a second-generation RT accelerator that provides new dedicated instructions, and a 50% performance uplift in ray tracing performance. The AI cores are not exposed through software, software developers cannot use them directly (unlike NVIDIA's Tensor Cores), they are used exclusively by the GPU internal engines. Later today AMD will give us a more technical breakdown of the RDNA3 architecture.
For the RX 7900 XTX, AMD is broadly claiming an up to 70% increase in traditional raster 3D graphics performance over the previous-generation flagship RX 6950 XT at 4K Ultra HD native resolution; and an up to 60% increase in ray tracing performance. These gains should be good to catch RTX 4080, but AMD was clear that they are not targeting RTX 4090 performance, which comes at a much higher price point, too.
AMD is attributing its big 54% performance/Watt generational gains to a revolutionary asynchronous clock domain technology that runs the various components on the GCD at different frequencies, to minimize power draw. This seems similar in concept to the "shader clock" on some older NVIDIA architectures.
AMD also announced FSR 3.0, the latest generation of its performance enhancement, featuring Fluid Motion technology. This is functionally similar to DLSS 3 Frame Generation, promising a 100% uplift in performance at comparable quality—essentially because the GPU is generating every alternate frame without involving its graphics rendering pipeline.
The new dual-independent media-acceleration engines enable simultaneous encode and decode for AVC and HEVC formats; hardware-accelerated encode and decode for AV1, and AI-accelerated enhancements. The new AMD Radiance Display Engine introduces native support for DisplayPort 2.1, with 54 Gbps display link bandwidth, and 12 bpc color. This enables resolutions of up to 8K @ 165 Hz with a single cable; or 4K @ 480 Hz with a single cable.
The "Navi 31" GPU in its full configuration has a raw compute throughput of 61 TFLOPs, compared to 23 TFLOPs of the RDNA2-based Navi 21 (a 165% increase). The shader and front-end of the GPU operate at different clock speeds, with the shaders running at up to 2.30 GHz, and the front-end at up to 2.50 GHz. This decoupling has a big impact on power-savings, with AMD claiming a 25% power-saving as opposed to running both domains at the same 2.50 GHz clock.
AMD claims the Radeon RX 7900 XTX to offer a 70% performance increase over the RX 6950 XT.
The complete slide-deck follows.
The complete slide-deck follows.
336 Comments on AMD Announces the $999 Radeon RX 7900 XTX and $899 RX 7900 XT, 5nm RDNA3, DisplayPort 2.1, FSR 3.0 FluidMotion
RX 6800 XT is a very old card which now supports only old standards and it's not worth it.Frank Azor doesn't know what he is talking about. Fake news, speculations and nonsense.
Frank, how do you design something with something else in mind which you don't even know what it will be? :D
Don't get me wrong I wish AMD all the best. This round they seem to have great perf/$ AND perf/watt, but a couple years back I had to replace the old monitor and got a G-Sync one, so I'm kinda stuck on team green, so if they're "bullied" into reducing prices I'm all in! It's insane that a 3070 nearly TWO YEARS after launch still costs over 600€.
Despite all the talk about 7900XT vs 4090, AMDs midrange looks to have been eviscerated and only the 7900XT + seem to show upgrades, on paper anyway.
Maybe the numbers aren't giving a clear indicator of performance, but I wouldn't bet on it.
i.e. this - just looks lame to me. Same thing with the 7700 XT vs 6700 XT, they actually went from a 192 bit bus to a 128 bit bus.
The upgrade seems to start with the 7900XT, lower end looks like downgrades :
For those that didn't see it these "performance numbers" were with the RX 7900 XTX only running at 300 Watts no where near it's 355TDP. expect higher clocks and better performance at 355. Why would they do that you ask? Simple AMD wanted you to focus on performance per watt upgrade and rightfully so and this was probably the best PEAK PPW area so the cherry picked it and thats almost 100% why they didn't show FPS. Thats the big improvement you see almost every year and speaks loud and clear. To many of you just don't remember because you are to young to remember lol. NVidia fans would bash ATi/AMD because of their Power hog cards that weren't when compared to what we see now from NVidia. 125W X1950XTX ( space heater Power hog screams from Nvidia fans ) Whats the 4090 pull again? oh thats right 450+ Watts, so about the same as a small window AC unit. Think about that for a minute! Those things can cost you over $100 a month to run if they are running 8+ hours a day. The 4090 pulls 12 kWh A DAY... WOW in most places thats $0.25 ( a quarter ) per Hour that you play games. I play 20 games a hour a week some times more that would be $5-$7.50+ a week in electricity $260-$390 a year!
CPU's and GPU's power draws are getting down right stupid! For those in the USA you do know that the average power cost is going to be raised about 70% right because of inflation and the rising cost of power with the removal of the gasoline surplus? Or did you all miss that also?those AMD RX 7800 XT specs are place holders and is from techpowerup and it SAYS BELOW where you trimmed it out PLACE HOLDER. That's everything not just the drawn picture of a GPU lmao
- Such heavy GPUs have existed before. I've not heard of any mass-reports about any issues related to its weight.
- It's not. The FE edition runs around 67C.
- Capped to ~315W (70% of TDP) it loses less than 5% of performance.
- The 90 series cards have always been super expensive. They are not for the average gamer.
- I don't give a .... about its UI. It works. At the same time I get lost in AMD's UI. Which tab do I need? Which place the option I'm looking for? Where's digital vibrance? People have been asking AMD for years for this option. Competive CSGO players do not touch AMD cards because of that.
- EVGA, what? Who the .... cares? 95% of the world have never seen EVGA cards.
- Out of literally tens of thousands of sold cards, fewer than a few dozen people have reported issues. And it looks likely all of them have done something wrong, including bending the cable too much or not properly inserting the adapter. Again, a card for the rich or professionals.
Literally not a single argument.
It's a free market. AMD is all yours. Remember Radeon 9800. Should I remind you about its release price? It was $400. Corporations are not your friend even though you want to love AMD so much.
There will be a 4MCD N32 part for the 7800XT with a 256 bit bus and 7.6k shaders.
The 192 bit 12GB is 7700XT.
Performance extrapolation for 7800XT is probably between 4080 12G and 4080 16G raster perf so maybe 3090Ti to 3090Ti + 10%.
Interesting thing here though is that unlike 7900XTX vs 4090 where the former has 96 RT cores and the latter 128 RT cores the 7800XT will have 60 RT cores as will the part formally known as the 4080 12G so there may be a smaller RT delta at this tier than there is between the 4090 and the 7900XTX.
The way I see it is that Nvidia is slowly separating itself from normal GPU market conditions, and is trying to establish itself as a niche brand, like Rolls Royce - a brand that people don't buy for its usefulness, but more as a status symbol and luxury item. AMD is trying to grow out of their shadow and form an identity that is not "the little ugly brother of Nvidia", but something that is their own, and I think they demonstrated that well in this launch presentation by establishing the 7900 series as a more efficient, normal size, and more affordable product - a buy that makes more sense (and by not mentioning Nvidia). For years, AMD's message was "we're not that far from Nvidia, please buy our products". Now the message is more like "we don't give a damn about Nvidia anymore - we go our separate path", which is what I was happy to see.
Just name the "RX 7900 XTX 24 GB" as Radeon RX 7800 XT and call it a day.
You should already know by now that infinity cache is a game changer and that is in fact a multiplication of the (traditional) bit bus.
Nice try...(?)
PS
More estimates, based on AMD claims of 1.5-1.7 faster than 6950XT
Cyberpunk 2077 (4K)
RTX 4090: 71.2 FPS
7900 XTX: 66.3 FPS
Watch Dogs: Legion (4K)
RTX 4090: 105.2 FPS
7900 XTX: 95.8 FPS
Call of Duty: Modern Warfare 2 (4K)
RTX 4090: 108 FPS
7900 XTX: 139 FPS
God of War (4K)
RTX 4090: 130.2 FPS
7900 XTX: 98 FPS
Red Dead Redemption 2 (4K)
RTX 4090: 104.2 FPS
7900 XTX: 93 FPS
Assassin's Creed Valhalla (4K)
RTX 4090: 106 FPS
7900 XTX: 109 FPS
Resident Evil Village (4K, Ray Tracing)
RTX 4090: 175 FPS
7900 XTX: 138 FPS
We're slowly coming to an age when nobody cares if a GPU is 2% faster than the competition. Size, power efficiency and affordability start to matter to more and more gamers, and it looks like AMD realizes that.
Edit: As far as I recall, Nvidia never talks about AMD during their product launches. Are they hiding something too?
But now both Intel and Nvidia are getting serious. The milking of 14nm++++++ is over and Nvidia returned to cutting-edge TSMC process.
I have a solution for AMD thought. Lisa Su can always spend some more billions on share buybacks.
Like I said, I don't recall Nvidia showing more numbers and AMD comparisons during their product launches, yet no one complains.
Show off is for juveniles and school kids.
A direct comparison to competition is not a show off. They didnt do it because they've decided to take the efficiency road this time. Less cost on power delivery, coolers... etc... And they probably going to come close to 4090 without braking the bank.
Anything wrong about that?
Their design (from RDNA2 and on) is cheap and probably they kept their profit margins high since they've made it even more production efficient with the cache chiplets.
nVidia does not have a direct product on the market right now for that price range so there is no point to show slides where they are below the competition even if they priced it at 60% of the 4090.
Prices are forgotten more easily than raw performance on slides.
I guess marketing is not for everybody...