There is always going to be pushback with a new technology. Everyone keeps challenging the issue of latency, and I think with the new technologies being developed, nay sayers need to be aware that this is unlike anything else that is out there.
http://www.onlivetips.com

I've been analyzing OnLive for months now and I've come to the conclusion that it will not be successful. Let me tell you why.

The problem here is that gamers on the east coast will not be able to play against their online friends on the west coast. OnLive will be using separate autonomous server locations which will be limited to 1,000 mile radius. OnLive has grossly underestimated the mentality of today's online gamers. Today's online gamers have made many friends worldwide and they love that aspect of it. Someone in the USA can play against someone in Germany or Australia currently. OnLive will not have this capability. Additionally, most popular multiplayer games already have a relatively low online population and dividing them up even further is going to create an extremely low multiplayer population for OnLive which will kill it in the end.

OnLive drives me nuts due to dismissive comments like the one from poster number 1: "Everyone keeps challenging the issue of latency". I can tell from that sentence alone that you are not a programmer and have absolutely no concept of how data is sent over a network.

Latency is not something that has been invented just so people can have a bash at OnLive. It's a very real problem that exists and is a pain in the arse for developers across the globe. Moving anything from A to B, even at light speed, takes a certain amount of time. In the case of data across a network (be that local or global) it is measured in milliseconds, a seemingly small amount of time.

Now lets think about certain types of games, shall we? FPS games, music games like Guitar Hero or Rock Band, and driving games have to run at 60 frames per second in order to be playable (okay, you can argue that shooters can run at 30, but in my opinion they're sluggish). Now I'm going to be generous and say that they can all run at 30fps and they'd be fine (which they wouldn't, but this is to prove a point). So, at 30fps a game has 33.3 milliseconds in which to collect the inputs, process all of your data, and then render everything to the screen.

The system that OnLive is proposing, and I don't care what magical technology you may think they've developed, cannot possibly gather the input data, send it to the server farm, process the frame and render it, compress the video data, and then get it back to the player's TV and decompress it in 33 milliseconds. Even if the person lived next door to the OnLive server farm, it is impossible. Even if the rendering, compression, and decompression somehow took absolutely zero time at all, the transmission of the data alone would struggle to make the round trip in 33ms. And don't bring up streaming video on the internet - that is completely different - it is linear, fixed footage that can buffer a chunk before it even starts playing. Games are dynamic and cannot be predicted ahead of time. Therefore, using a bit of common sense, it can be deduced that what OnLive are proposing simply cannot work for fast-paced games that require immediate processing of input in order to play correctly.

Now, having said all that, the OnLive system may very well work for slower paced games, such as RTS or RPG games. This is, of course, ignoring the cost implications of scaling up the server farm, etc., but that's a separate issue.

What I am looking for from OnLive is an honest admission that their idea will not work for all games, as they are currently suggesting. They could put an end to the 'nay sayers' like me by just being upfront and saying that their system will never EVER be able to run a 60fps game, even if the network was running at light speed, but it is suitable for other types of games. Then people like me would shut up. I've seen OnLive in operation and it was crap (on a very, very fast network), but that isn't what is fuelling my ranting - I just want OnLive to be honest about what should be expected from them.

Blindy dismissing latency as something that knowledgable people use to bash OnLive is absolutely moronic. Latency is a huge, huge problem. Ever tried to write a multiplayer game that runs over the internet? No? Well, if you had, you'd know that simply getting a couple of Kilobytes of data across the pipe at 60fps is a massive challenge.

@1 - before you make sweeping statements, try and do some investigation so you have at least some knowledge of what you're commenting on.

No that I am certainly not a programmer, however I think it is IMPOSSIBLE for someone to make comments about a technology that they have absolutely no insight on. With hundreds of patents submitted, is it the slightest bit possible that this is a NEW technology?

At one point, it was thought impossible that there would be personal computers in the home... Now we are having companies like Cray designing High Performance Computers the size of the original desktop.

So although I may not be as techy as you (or have as much time to type a novel on a website discussing the topic), I can tell you that there are new technologies being introduced everyday that continue to push the envelope on what is being developed.

In the most simple terms for the most simple people ...

yeh canne' change the laws of physics!

And @3 obviously has some insight because he does have a concept of the subject and seen the thing for himself!!!

Good Afternoon!!! www.develop-online.net is one of the best informational websites of its kind. I enjoy reading it every day. Keep it that way.

The author of www.develop-online.net has written an excellent article. You have made your point and there is not much to argue about. It is like the following universal truth that you can not argue with: The great pleasures in life are doing what people say cannot be done. Thanks for the info.