AI scaling and post-processing for video games has been a hot topic recently, and PlayStation is now gearing up for another development. After Nvidia's reveal of DLSS 5 earlier this week invited plenty of criticism and controversy, Sony has just confirmed that a different setting dependent on machine learning is on its way.
Speaking to Digital Foundry, PlayStation's lead system architect Mark Cerny confirmed that frame generation "should be seen at some point on PlayStation platforms." Frame generation uses machine learning to insert additional frames in between natively rendered ones. This makes it possible to achieve a higher frame rate with a lower performance cost, but the technology isn't without its problems.
The main benefit to frame generation is simply creating a smoother overall experience while gaming, as low frame rates lead to choppy experiences. Unlike native rendering, however, the additional frames provided by frame generation are effectively just guessing at what information should come before and after the rendered frames. This can lead to visual artifacting, and the need to render the subsequent frame before displaying the generated frame can increase latency.
Cerny doesn't clarify when machine learning might come to PlayStation hardware, simply suggesting that nothing is...
