Thoughts on OnLive

Re:Thoughts on OnLive by thunk on Friday June 18, 2010 @11:34PM (#1444101)

I wonder if you could eventually sell cycles back to OnLive, since you aren't doing the heavy lifting on your machine ... Maybe offset the cost of the service? Or maybe OnLive could do an ElectricSheep-like screensaver for distributed processing. That could mitigate some of the latency issues as well, for those near you. Social rendering, comrade!

Edit: The more I think about this, the more it seems this is precisely (or at least should be) where they're going with this.

[ Parent | on HN ]

Re:Thoughts on OnLive by po on Saturday June 19, 2010 @10:22AM (#1444676)

The added complexity and lack of centralized control isn't worth the cost of cycles.

[ Parent | on HN ]

Re:Thoughts on OnLive by thunk on Saturday June 19, 2010 @12:46PM (#1444901)

I don't know. OnLive can't scale very well as is. A liberal estimate would be four game instances per server, which means 250,000 servers for a million players. But if you're spreading that around, especially on players' machines that aren't being used at the time and may be fewer hops from the destination, it may be worth it. Plus, it'd be a good first step towards a mostly decentralized gaming infrastructure.

[ Parent | on HN ]