When Epic Games' Samaritan demo dropped at GDC in March, you could hear the gasps all the way to Frankfurt.
A cyberpunk vision of a twisted reality bathed in glorious neon shades, it was perhaps the closest the mainstream interactive industry had ever come to a 'photo real' graphical showcase. And it wasn't even a real game.
Although Epic admits it was boosted by a beast of a PC - pushing Unreal's DX-11 capabilities - the studio called the footage its "message" to modern day console manufacturers. It was a very clear one, too: Sony, Microsoft - you better be able to pull this off with your next machine, or else.
That's some threat from a company which boasts 65 per cent of the licensed engine market with Unreal - technology that has memorably powered Batman: Arkham Asylum, BioShock and Epic's own Gears Of War amongst countless others.
We caught up with the firm's European boss, Mike Gamble and its senior technical artist and level designer Alan Willard to talk about the future of gaming, Unreal and interactive devices.
The first part of our chinwag is below, in which we discuss Samaritan, the next-next generation - and what video gaming will look like in years to come...
Is there a particular vision you have for how a hardcore gamer will play in five years?
Mike: To make that bet would be a foolish thing. It's just so rapidly changing, in the space of three years or so we can play on devices that we never could before so game styles are changing. I think anyone that places their money on that number is not good.
Looking at what you can do now with Samaritan, how far away are we from what we used to call photo-realistic gaming?
Alan: It's too problematic to make something that's...well, nobody wants to go home, sit down and play their own life, I think that's one of the mistakes that a lot of people assume. You want it too look good, you don't want it to look real. Yes we want to make characters look really good but if you look at BioShock Infinite those characters look fantastic, Gears of War's characters look great, they're not photo-realistic, they're not meant to be. They're meant to be interesting, fun and great game content.
Mike: There's a style.
Alan: Yes exactly. There will always be people that are there for technical fun or because they can strike photo-realism but I don't think any games that will go for that because the hurdles are fairly significant for not a whole lot of return in terms of making the game better and I think that should always be the focus of technologies. If it doesn't make the game better, if the particle system look great and characters look great but if it does that at the expense of making a fun game that's a trade-off that shouldn't necessarily be made.
Mike: I think that really is the key thing. The ultimate stress test for anything we do is whether it is fun. If it's not fun there's no point in doing it.
The scariest part, from your competitors' points of view, is that it took three people 12 months to make it...
Mike: The real point to take away is that it wasn't built using a special toolset, it was built with what is now in the marketplace. The only thing about it was the hardware it ran on was a very expensive PC. But even so you could go out today, buy all of those components, build that PC and have that capability. You'll have no content to run on it but you have the capability. So there's no trickery, it's not like we created some hyper-tool or whatever you want to call it. It's what we can do.