Couldn't see a thread anywhere...

But this is absolutely huge in regards to the Playstation 6 and it's new catchy advertising name for a hardware feature... The "Radiance Cores".
It's not a rumor either. You don't need salt.
https://www.tomshardware.com/video-games/console-gaming/sony-and-amd-tease-likely-playstation-6-gpu-upgrades-radiance-cores-and-a-new-interconnect-for-boosting-ai-rendering-performance

Normally when a GPU compute unit needs work, does work or has finished work... That data is sent back and forth to caches for that data to be used for something else in another compute unit, it's sometimes high latency and impactful to performance... Provided that work is needed straight away that is.
This literally sidesteps all of that.

This should be able to allow for some coherency for A.I applications, some RT effects will see massive improvements as it can do everything at once in a single pass. (Think Path Tracing)

This is a much larger paradigm shift for rendering next-gen games than what we saw with the jump from the PS4 to PS5 which is a very positive sign.

We should see this technology debut on PC first, so should be good to put that stuff through it's paces.

Cerny is really, really damn intelligent, he and Sony devs working in unison with AMD might indeed be huge, and as said back them, that's for both PC and PS, so pretty much the entire high end gaming market is benifiting from this collab, which is awesome.

Seems like the PS6 isn't coming before 2028 at the earliest, which is good news to me, but I hope this leads to it being more powerful than what the leaks suggest. If not, then at least cheaper than we think.

If a true paradigm shift is on the way, a 2028/2029 PS6 achieving only 5080 level performance would be quite underwhelming.

The technology will be rolling out in RDNA 5 next year, which will enable developers to start playing with the technology and build development tools around it... Which means we should see a rapid uptake when the Playstation 6 drops.

Technically Sony could drop the PS6 late 2026 in conjunction to RDNA5, but 2027/2029 is likely a better window so they can be on a better process node for a smaller and more power efficient chip.

I would expect the Playstation 6 to be about equivalent to a mid-range PC, just like the Playstation 4 and Playstation 5 to keep costs reasonable, but it will be -smarter and able to do more with less. - And that is what the industry needs to reign in costs, especially as Microsoft, Nintendo and Sony don't wish to subsidize console costs to the same degree as they used to.

But the bigger picture here is what it would mean for a potential Playstation handheld... A.I scaling and efficient RT would be transformative.
We are seeing AMD Z2 Extreme chips competing with the Switch 2 in terms of raw power, but the Switch 2 wins because its a smarter and more efficient design, despite being based on older technology, so that's super exciting.

Cerny and AMD are doing some great things that should benefit the entire industry.

The rumours have it at about 5070 Ti level, which would be underwhelming for a late 2028 console, notably worse than PS5 and even PS4 were at the time of their release. Perhaps this isn't factoring in the more advanced architecture and feature set which could be the game changers that effectively put it far above a 5080 in some ways.

PS5 ($400/Nov 2020) beat RTX 2070 Super ($500+/July 2019) and traded blows with 2080 (highend, $700/Sept 2018) in raster performance. PS6 needs to at least match a high-mid-range PC from the year prior AND be relatively cheap to achieve a comparable result. Granted RDNA2 sucked at RayTracing and ML.

Xbox Series and PS5 also introduced RDNA2 a week before PC. I wonder if PS6 will include any significant features not found in 1-2 year old highend PC's from at least AMD's side.

It makes more sense to compare Sony's consoles with AMD GPU's directly.
The Playstation 4's GPU's closest equivalent was the Radeon 7850.
The Playstation 5's closest equivalent GPU is the 6700... A mid-tier 9060XT is already 50% faster.

They roughly fall in that upper mid-range segment in AMD's product stacks.
The Playstation 6 would be roughly equivalent to a Radeon 9070 if it released today, which is already competitive with the RTX 5070. (Non-Ti) 

The Radeon 10070 or 10700 (There will probably be a new naming scheme.) should be a substantial improvement over the 9000 series due to the improvements to A.I and Ray Tracing which is the industry paradigm shift that is happening currently... And where AMD, Intel and nVidia are investing most of their transistor budgets into improving.

As for hardware release timelines... AMD is building it's GPU's alongside next-gen consoles, they release when they release... AMD, nVidia and Intel have a release cadence to coincide with their OEM contracts, which may be before console hardware or after... But these things take YEARS to design, iterate, design some more, ratify and test... Then release.

Honestly though, I hope Sony and Microsoft drop their next-gen consoles with xx60 tier hardware rather than xx70 tier hardware and opt for Zen Compact cores to optimize die-area and substantially reduce costs.
And with the new compression being built into next-gen GPU's, we should see some significant bandwidth gains.

In short... I want smarter and more efficient designs, not just large, fast and power hungry.

In AMD terms, Kepler said it would be around 9070XT level (which is comparable to 5070 Ti in raster), but be more advanced than Blackwell (which is superior to RDNA4). All in all, the implication is that it would be around or slightly weaker than a 5080, even with the more advanced architecture/feature-set in mind. Alex Battaglia thinks having a 9070XT+ of raster performance is too high, unnecessary, and supposedly contradicts Cerny's vision.

What I'm saying is that PS6 matching the upper end of midrange AMD GPU's from 2027 or later requires a greater offering than what's being implied by those rumours. But I realize that they're just rumours, and that a lot of the leaked plans and targeted specs are subject to change. Rumours claimed a 2027 release, when Cerny's wording kinda makes it clear the console(s) isn't coming before 2028. At this point I'm not sure if the handheld thing is even happening.

Estimates for Yotei by Alinea Analytics.:

-1.6M+ sold to consumers (2M+ incl. retailers).
-$100M+ in revenue. Yotei made back its ≈$60M dev budget the day after launch.
-Outselling AC Shadows by 1.5X on PS5 so far.
-93% of people playing Yotei played Tsushima, 27% AC Shadows, 20% AstroBot and 18% Death Stranding 2.

I care about sales because of the developers safety lol and it looks like Sucker Punch will be fine, they can work in Legends without worries.

Not sure I trust Alinea fully yet as their basis on estimates doesn't make much sense lol. Typically tied to PS Store reviews.

Still, given how well Tsushima sold, I wouldn't be surprised at all. Even moreso, the fact they were able to accomplish something on the scale of Yotei for only $60 million makes it seem like black magic lol. Sucker Punch almost seems like the black sheep of Sony's first party studios when you see regular figures of $200-$300 million budgets for Naughty Dog, Santa Monica, and Insomniac. I think even Bend had a much bigger budget for Days Gone lol.

It's amazing what you can accomplish when you have a set plan and keeping your budget in check.

I'm hoping to finally get to play Yotei later this month.