• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Discussion RDNA 5 / UDNA (CDNA Next) speculation

Page 105 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Saylick said:
Hey, when gaming is <10% of Nvidia's revenue, it's natural it only gets <10% of the their effort.
Click to expand...
The extraction team is top notch though. Bad driver? Card in PR trouble? Unlaunch in less than 24 hours. It's as if it wasn't there to begin with. /joke

On a mores serious note, that 10% of Nvidia effort is still updating the Nvidia Shield, more than 10 years later. They probably forgot to re-allocate the dev, or they can't find who's in charge of updates anymore 😀
 
<circular news alert>

AMD’s RDNA 5 “AT0”: Enthusiast GPU with the handbrake on ???​

27. February 2026 06:00
Samir Bashir

Realistic assessment​

  • Rumor status: Plausible, but unconfirmed
  • Market strategy: Image cultivation instead of mass market
  • Competitive pressure: High, especially in the >$1,500 segment
  • Timeframe: Realistic in 2027 at the earliest
If AMD launches AT0, it will likely be as a prestige project with limited quantities. If not, RDNA 5 will remain positioned in the high-end performance segment, with prices between $700 and $1,000 as the likely focus.

In short: AT0 is currently more of a strategic option than a finished product. And as long as AMD does not officially confirm anything, it remains just that, a rumor with medium plausibility.

https://www.igorslab.de/en/amds-rdna-5-at0-enthusiast-gpu-with-the-handbrake-on/
 
marees said:
<circular news alert>

AMD’s RDNA 5 “AT0”: Enthusiast GPU with the handbrake on ???​

27. February 2026 06:00
Samir Bashir

Realistic assessment​

  • Rumor status: Plausible, but unconfirmed
  • Market strategy: Image cultivation instead of mass market
  • Competitive pressure: High, especially in the >$1,500 segment
  • Timeframe: Realistic in 2027 at the earliest
If AMD launches AT0, it will likely be as a prestige project with limited quantities. If not, RDNA 5 will remain positioned in the high-end performance segment, with prices between $700 and $1,000 as the likely focus.

In short: AT0 is currently more of a strategic option than a finished product. And as long as AMD does not officially confirm anything, it remains just that, a rumor with medium plausibility.

https://www.igorslab.de/en/amds-rdna-5-at0-enthusiast-gpu-with-the-handbrake-on/
Click to expand...
I don't think AMD intends to compete with Nvidia in the gaming segment; they're simply not interested in it because it´s not woth of efforts. They are only interested in AI/HP Compute and high margin products. The gaming sector, they are content with a position just below Nvidia's high/mid offering and they are happy with their - $50 price strategy. Offer enthusiast GPU doesn´t make sense anymore
 
adroc_thurston said:
$2k is the segment where price is not a consideration and people pay top dollar for the best.
Click to expand...
Yes, that's why if AMD releases AT0 this side of Xmas then it will fly off the shelves, I am certainly ready for upgrade from 4090, but it has to be better than 5090 by at least 30% - raster, and RT should be at least matching it.
del42sa said:
AMD intends to compete with Nvidia in the gaming segment; they're simply not interested in it because it´s not woth of efforts.
Click to expand...
AMD won't break huge laptop market with sloppy RDNA3.5 - this keeps Intel in the game and will backfire badly.
 
Win2012R2 said:
Yes, that's why if AMD releases AT0 this side of Xmas then it will fly off the shelves, I am certainly ready for upgrade from 4090, but it has to be better than 5090 by at least 30% - raster, and RT should be at least matching it.
Click to expand...
No Geforce no DLSS no RTX. DOA.
Win2012R2 said:
AMD won't break huge laptop market with sloppy RDNA3.5 - this keeps Intel in the game and will backfire badly.
Click to expand...
GFX does not matter in non-discrete compete segments.
 
adroc_thurston said:
No Geforce no DLSS no RTX. DOA.
Click to expand...
And you think people (like me) who buy high end like 4090 wants to use DLSS? No - we expect to render in full native 4k, that's the reason for premium paid for high end GPUs.

Plus pretty much all new games (that can run slow even on 4090) support FSR.

Heavy RTX stuff is only pretty much still CyberPunk - I played it with 3090 and 4090, found way too slow with RTX and fine with raster.
adroc_thurston said:
GFX does not matter in non-discrete compete segments.
Click to expand...
Intel does not seem to think so and they are the market leader.
 
Win2012R2 said:
你觉得像我这样的高端用户（比如4090）会想用DLSS？不——我们期望以完整的原生4K渲染，这也是高端GPU付费的原因。
Click to expand...
Most modern games use garbage TAA by default. You should really embrace upscaling tech like DLSS, as it actually provides a superior anti-aliasing solution.
 
marees said:
<circular news alert>

AMD’s RDNA 5 “AT0”: Enthusiast GPU with the handbrake on ???​

27. February 2026 06:00
Samir Bashir

Realistic assessment​

  • Rumor status: Plausible, but unconfirmed
  • Market strategy: Image cultivation instead of mass market
  • Competitive pressure: High, especially in the >$1,500 segment
  • Timeframe: Realistic in 2027 at the earliest
If AMD launches AT0, it will likely be as a prestige project with limited quantities. If not, RDNA 5 will remain positioned in the high-end performance segment, with prices between $700 and $1,000 as the likely focus.

In short: AT0 is currently more of a strategic option than a finished product. And as long as AMD does not officially confirm anything, it remains just that, a rumor with medium plausibility.

https://www.igorslab.de/en/amds-rdna-5-at0-enthusiast-gpu-with-the-handbrake-on/
Click to expand...
Developing an expensive chip to ship it in in limited quantities is a sure thing to loose money. That will not happen.

To get a fast GPU, you not only have to install lots of shaders, you also have to utilise them to their full capacity. And that doesn't look good when comparing the RX9700XT and RX9700, for example. So it's very questionable what kind of performance you'll actually get from a RDNA4 with twice the number of shaders.
 
ETI4711 said:
Developing an expensive chip to ship it in in limited quantities is a sure thing to loose money. That will not happen.

To get a fast GPU, you not only have to install lots of shaders, you also have to utilise them to their full capacity. And that doesn't look good when comparing the RX9700XT and RX9700, for example. So it's very questionable what kind of performance you'll actually get from a RDNA4 with twice the number of shaders.
Click to expand...
It's 3 times the shaders and it's a different microarchitecture.
 
Booth: ) it's a technological leap for AMD to make core dies from three companies compatible. Will AMD promote HBM or keep it a secret?

Nvidia has each company redesign their core die to fit base die. E.g. socamm?
 
Last edited:
Thunder 57 said:
So who's the bozo? AMD for not having a halo part which you just said has "no real market", or me for making fun of Nvidia's drivers by calling them "poverty"?
Click to expand...
AMD can only sell $2k+ parts if they win by solid double digits and don't have glaring sw weaknesses.
AT0 ain't quite big enough and they've got a lot of work to do with the whole FSR5 suite by the time RDNA5 launches to be respectable.

Now there is another way they could sell and that is if 90 class stuff stays at extremely inflated prices and you can slot AT0 in at 80 class street pricing.
They've gotta push forward with software, set the parts you're gonna support long term and go for it. Wasting time on niche projects on obsolete products is not going to win AMD any new fans, bending over for the loud minority of chronically online nerds is why Radeon is a small die poverty brand. NV does what NV wants to do with only colossal backlash being backtracked eg. 4080 12G.
Also excuse the latest RE having DLSS performance turned on by default when PT is used, that is a far bigger controversy than AMD refusing to support something they never said they would support but uhh.

I'm very bullish on GFX13+ software support being far superior to how it has been done in the past, not having something like PTX has been a colossal oversight.
 
If, say, the additional costs of bringing AT0 to market for gamers is $200 million. AMD would need to sell 100,000 AT0s at $2000 to make up costs. Would it really be impossible to make those numbers?

There are like a hundred million PC gamers worldwide. Many of them can easily sell their previous nvidia flagship and buy the new AMD flagship for zero cost. There are at least 100,000 wealthy AMD fanboys who would buy it even if AT0 is dogsh1t.

There is no economic or logical reasons not to bring AT0 consumer gaming version. Its not being made because the market is a cartel. You will have a real market once the Chinese enter it.
 
OK, if you cannot change it then it's bad. But I would suggest a game developer issue first, not an Nvidia issue. This is the first PT game which does that.
 

AMD Says “Helios” Racks And MI400 Series GPUs On Track For 2H 2026​

Timothy Prickett Morgan
Published mon 23 Feb 2026 // 14:00 UTC

we sat in on a conference call hosted by New Street Research last week, where Forrest Norrod, general manager of the Data Center Solutions business group, and Doug Huang, one of the co-founders of ZT Systems, which was acquired by AMD last year to underpin its rackscale system engineering efforts, and now its senior vice president in charge of the Data Center Platform Engineering group.

Norrod was having none of it when asked about a delay with the Helios racks:

“I have no idea where this purported issue around thermals is coming from,” Norrod said unequivocally. “I literally have no idea. We have no significant thermal issue. The risk around the thermal design at the component level all the way through the rack level was retired quite some time ago. So, no idea where that’s coming from. I think the meta question that you are asking is when we expect to see the ramp, and are we on track. Lisa showed the first silicon, we are right on track with where we thought we would be, both in terms of the readiness of the overall solution as well as the readiness of the silicon. And we are highly confident of ramping Helios in high volume in the second half of the year.”

So that is that. The reason why AMD can be confident was explained by Huang, who showed off two charts, and explained that they use what amounts to dummy hot plates to simulate the CPUs and GPUs long before they come back from the fabs so they know the physical specs and thermals of the chips that will come back will fit into the racks and the racks will work right.


This may be AMD’s first rackscale design, just like the NVL36 and NVL72 rackscale machines that were supposed to be based on Hopper were – and never came to market except for one machine sold to AWS. But this is not the first rackscale machine that ZT Systems has developed, and it is not the first one for Meta Platforms, either, which started the Open Compute Project back in 2011 with server, rack, and datacenter designs. Getting Helios out the door on time and being manufactured in volume is why AMD spent $4.9 billion to acquire ZT Systems back in August 2024. Not wanting to be in the server manufacturing business is why AMD sold the manufacturing arm of ZT Systems to Sanmina for $3 billion last fall.


www.nextplatform.com

AMD Says “Helios” Racks And MI400 Series GPUs On Track For 2H 2026

While releasing an update to its InferenceX AI inference benchmark test, formerly known as InferenceMax and thus far only having Nvidia and AMD testing
www.nextplatform.com www.nextplatform.com
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top