Now they want to take the rendering and move it from the client to the server. Okay, that's not a bad idea, it allows for very thin clients. This is the general premise behind web-based applications such as Salesforce.com etc. However, there are two problems that need to be addressed for this to work.

1. Rendering: Currently if 20 people are playing a game that means there are 20 videos being rendered. In order for this to work a "server cloud" would have to have the power of 20 high end video cards along with the CPU and RAM to manage all the variables. Now multiply that by the number of people playing at any one time. Now clean the brain matter off your screen from your head exploding.

2. Bandwidth: Now you are not only sending a data stream back and forth, but fully rendered HD video. The amount of bandwidth used for one 20 player game that lasts 30 minutes would be tremendous on both the server and client side. And with several major ISPs (Time Warner anyone) setting bandwidth caps you could blow through that in one game.

This is an interesting idea and will make an excellent proof-of-concept, but I simply can't imagine it is ready for prime time or will be anytime soon. Not only do I think the tech is not there, but only people who buy more than say a dozen games a year will be willing to pay for it. I only play 3-4 games a year, so it would have to be less than $10/month for me to even consider it. Oh, and they would have to carry the games I want to play.

 Reply