OnLive's Steve Perlman

Yes. If you look at Zynga, there's many things you can say about that, but the one of the things I see as complete genius is that they study the data so carefully to understand what makes the best sense for the users and what makes the best sense from a business point of view, and they come up with this happy medium where people love playing their games and they make a lot of money. What's not to like about that? It's been very, very hard to do that in the core game market because of the very high friction it takes to getting the games out there, and of course with the decline of PC gaming, there's fewer and fewer high performance PCs out there, most people have laptops if not tablets, once again its very hard for the publishers.

At least that market they could kind of directly reach instead of going through publishers. So we gave the publishers tons and tons of data, they love it. It allows them to go and make decisions, to design games better. Now in the design process we can actually go and put up games in a beta group that's closed off from the rest of the world even though it's running on the same servers but no one else can spectate that. And then the publishers are the only ones who can spectate that. They can actually watch their users testing their games. What they get stuck on, what doesn't work and so forth. They're loving it. So we have again a very, very unique way for people to look at the whole process of development in addition to the opportunities for merchandising.

If you'd asked me at E3 2010 if we would have a hot game that's two and a half months old, in a flat rate tier for $9.99 a month, I would tell you there's no possible way that's going to happen. But when you look at the economics and you look at the business and you look at the usage patterns it makes an incredible amount of sense. There's even a slight bit of genius that if you really knew what the metrics were measuring you would understand, that the version that's available in the flat rate package is only multiplayer, for Homefront. And the single and multiplayer is still available for full price. This is managing core gaming with metrics the same way that Zynga manages social gaming, that they've never been able to do before. And you're right, we've gone from thinking – because we were using NPD and Nielsen numbers to figure out usage patterns and so forth, and try to decide how many hours a month people were going to be spending, what type of patterns and such. So we were terrified about just being overloaded and not being able to provide the service. So we built in all these mechanisms for virtualisation and so forth in order to manage the load, deal with peaks and so forth. And I guess we overachieved.

Put it this way, we did such a great job we thought when we launched, we saw with all these technologies we built in to the platform was that we believed, even exiting beta, there was enough revenue coming that we could offer the service without charging anything for the basic access for demos and things like that, and sure enough that proved out in practice once we had enough people using it, enough sales, enough different types of users and so forth. And you're right. What we did is we adapted.

That's correct. At least another country, maybe two, by the end of the year, but we'll see. And I say that with my vice president of engineering grimacing! But nonetheless, we're going to try. There's no technical obstacles, we've been in Europe for well over a year. It's just to get a system to not just work for demos but to actually work for hours on end of gameplay, over any internet connection, even in Wales it's not just engineering, its a matter of testing and logs and understanding and having a record of all the different types of ISPs, all different types of connections that people have and then having them adapt.

For example if we were connected to your home, it would go and see what ISP, what region it's in, what's the route is to it and it would go and find what type of compression algorithm is the best one to use, of the over a hundred that we have available. And then you would look at the different statistic for what was happening, from the time of day and so on. It would even adapt the compression algorithm based on the type of game you were playing. It's a very, very complex thing, it took years to get that data. And Europe's similar to the US, but it's not the same.

We've done all that, it's complete. So the trials that we've done, we can't go through the details of what BT has done but let's say they were very comfortable with us going in and putting their name on OnLive.co.uk and launching this autumn.

Do you know one person who really got this early on? Chris Kingsley down at Rebellion in Oxford. When we were working with him during launch last year he was thinking about all these different economic models. I forget the words he used but it was like the diversity of economic models you can create on a platform like this are infinite, you can really adapt it to the needs of consumers and find a price point that meets their budget. I'm sure next year we'll have some new models we didn't even think of now.

Nintendo has essentially given credibility to the idea that there are going to be multiple screens in games. The difference between us and what Nintendo are is offering is that we'll work on iPad, Android, smartphones, we're completely hardware agnostic and we'll work with a TV where you can be seeing the same thing controlling, even use a smartphone to control it if you want touch, but if you want to take it with you, you just take it with you wherever you are. You can be on the train, you can be at work, you can be hanging out at a cafe, whatever it is and you still have the capability here. Or you come home and you want to pick up the controller and use that. So what Nintendo has described in old world terms, non-cloud terms, of the two screen experience is what we're showing in cloud terms with compete generality. In 2012, I think we'll be quite well established. And the publishers totally get it. They're designing stuff with the OnLive SDK that's going to do very much the same kind of thing that you saw with the Nintendo announcement. But the difference is that it will be on platforms generally rather than to one.