Is this the end of the traditional console generation?
This is where we suspect there is a difference in opinion between Sony and Microsoft. Last year,
Mark Cerny drew a line in the sand - PS4 and Pro were of the same generation and we'd require a radical upgrade in CPU, GPU, memory and storage to justify a prospective PlayStation 5.
Mike Ybarra's comments about users wanting the latest technology faster suggest something different.
Our take? Mark Cerny's outlook is possibly the ideal, and more in line with the core gamer's expectations, but Mike Ybarra hinting at more frequent console refreshes more closely fits the realities of console manufacturing. Moore's Law is slowing down. A 6x to 8x leap in console power every five years is looking less and less viable, so we should expect to see more frequent, more iterative upgrades. However, likening this to a mobile phone upgrade cycle isn't the way I see things going. A new console with a 2x-3x boost every three to four years seems more likely, based on current trends.
Typically, two aspects define improvements in console power - innovative design and the ability to shrink transistors down using the latest semiconductor fabrication technologies. By my reckoning, there's plenty of scope to see performance improvements via the former, but the latter is going to be a struggle. PS4 and Xbox One launched in 2013 and used 28nm chip fabrication technology - first used on PC graphics cards way back in 2011. The replacement 16nm FinFET process used on both Pro and Scorpio only became viable five years later - and I'd venture to suggest that both consoles would not be viable without it. 10nm is starting to roll out now, but may not be suitable for consoles, while 7nm is some years away. With that in mind, more iterative console launches may be the only way to get better hardware out to tech-hungry gamers, but even so, yearly or biannual updates are highly unlikely.
