Great point about extrapolating game state. They are way overselling it. It's simply not possible to do as well as a game running locally, as they claim (maybe

Great point about extrapolating game state. They are way overselling it. It's simply not possible to do as well as a game running locally, as they claim (maybe in 7 more years...). But there's a continuum of playability, between ideal and unplayable. If there are other benefits (flexible, convenient, cheap), and there is an audience who is very happy with this tradeoff, then this possibly, maybe could be a disruptive innovation. It's not going to take off with hardcore twitch FPS, so they seem to be targeting the wrong market (probably just for a more exciting demo). It's niche audience might be more like for the Wii: they don't care it's less powerful and not HD, but they like it because it's more fun (and cheaper). A similar story for casual games (whose success surprised some people). I'm not saying it will find a niche; just that it's possible. And then over time, latency improves, and they're ready for it. The argument I've heard about coping with first day demand spikes is a bit silly. Of course it won't handle that - but it doesn't mean it's useless. What about all the other days? Demand spikes are a problem for physical distribution channels too.