A couple of days ago, Sony and AMD posted a video on YouTube, the former briefly explaining the future of console gaming, while the latter spoke about how it plans on bringing its next-gen graphics technology into what sounds a lot like the PlayStation 6 (PS6).
Mark Cerny, Lead PlayStation architect, and Jack Huyn, the current head of AMD’s graphics head, sped through the nine minutes, explaining on how it was bringing three new technologies, all from the Radeon graphics: Neural Arrays, Radiance Cores, and Universal Compression.
If this all sounds familiar, it’s because Huyn also spoke briefly about them at Computex 2025 earlier this year, when AMD announced the Radeon RX 9060 XT. Basically, at the tail end of the presentation, Huyn spoke of FSR Redstone, the next step for its upscaling stack, which included three new technologies: Neural Radiance Caching, Machine Learning Ray Regeneration, and Machine Learning Frame Generation.
The points are similar, just in the context of the alleged PS6. Neural Arrays allow the console’s GPU compute units to work in unison, sharing data and processing things together, but refraining from linking them into a single mega unit.
Radiance Cores, based on the Neural Radiance Caching technology, will be dedicated hardware blocks, designed to handle both ray tracing and path tracing on the alleged PS6 console. According to Cerny, this was developed due to the technical limitations of current hardware.
The third feature, Universal Compression, is a system that intelligently compresses and decompresses data, sending out “only the essential bytes”, dramatically reducing memory bandwidth usage.
As to when we can see the new alleged PS6 console featuring this new technology, Cerny said that Sony is still testing out these new technologies, and that any semblance of a next-gen console will only be coming either in 2027 or 2028. That’s more or less within its console lifecycle; the PS5 launched in 2020, six years after the PS4 was launched.
(Source: PlayStation)
