OnLive Announced, Subscription-based Service Plans to Change PC Gaming with 720p Cloud Streaming

By Nick Breckon, Mar 24, 2009 1:12am PDT OnLive today announced plans to launch a new service that will stream PC games with negligible lag to televisions and PCs, eliminating the need for hardware upgrades.

Developed in secret for the past seven years, OnLive makes use of cloud computing, streaming video to your monitor at 720p resolution and 60 frames per second.

Because the video is rendered in a "cloud" of servers, games such as Crysis are delivered at full graphics settings, the only bottleneck being the user's internet connection. 1.5 megabits per second will get you to standard definition, while a 5 megabit connection bumps the resolution to 720p.

OnLive's proprietary compression technology boasts a lag time of less than one millisecond in testing, and early reports indicate that the service delivers on this front. Gamasutra notes that OnLive has worked directly with cable and internet providers to eliminate packet loss.

Users will require only a small device (pictured above, left) to connect to the subscription-based service to televisions. Two USB ports and Bluetooth capability service game controllers, while an ethernet connection grabs the signal, and HDMI and audio output jacks output to displays.

The service will also work on any PC, allowing users to play Crysis on a $200 netbook. Users need only subscribe to OnLive and purchase individual games from the library, with the option of buying the "MicroConsole" device for television functionality.

OnLive will feature friends lists and and online profiles in Xbox Live fashion. Brag Clips, 10-second replays that can be captured from any game at any time, will be viewable and shareable through the community interface.

Ubisoft, Take-Two, Electronic Arts, THQ, Atari, Warner Bros., and Epic Games have all signed on to the product, with Activision Blizzard being the major missing name. Publishers will have the option of both selling and renting their games.

The product is similar to a system announced by AMD last year, but OnLive has been demonstrated with success, has the backing of several major publishers, and plans to launch the service later this year.

We'll have more impressions of OnLive following a demonstration of the technology at GDC tomorrow.

Click here to comment...


70 Threads* | 278 Comments

  • Right now if i wanted to I can pull my atari 2600 out of my closet, or my ps1, or my NES, hook it up and enjoy the games I grew up with.. I would just hate for my kids to grow up with games and to not actually own them, to not be able to open that one game they wanted so bad on their birthday, when they are older to not be able to pull out their fav system when they were a kid and relive the nostalgia.. so this type of system may work for some people, but no way in hell I would want something like this to take over gaming completely.. no thanks

  • The creators say their compressions scheme adds a tiny 1ms of latency on top of whatever latency numbers your network connection suffers from. They have to be intelligent guys so they must realize that while their proposal may be feasible on a high quality LAN this isn't going to work on the WAN/internet. So...why make claims that this service will be available to the consumer later this year and promote the hell out of it at GDC? Why try to get yourself noticed on the industry radar now?

    Are they trying to stake their claim/mark their territory/establish a brand so that a Microsoft or Sony exec gets interested and buys them out in the assumption that OnLive will just drop into Live/PSN? Does OnLive really have any intention of delivering a product or are they just trying to make themselves into an acquisition target?

  • Seriously, are people that stupid believing this crap? This is not revolutionary, this technology is around for 30 year if anyone knows how terminal to host systems work, like Airline Reservations Systems. Web servers/ Web applications works same way. This is step backwards in gaming, not forward IMO, here is why:

    1) It relies on internet, which is prone to downtime, reliability and security issues.

    2) High cost for no ownership!. After paying probably hundreds of $$ a year on subscription fees, you don't physically own hardware or game.

    3) It relies on single entity to handle subscriptions and games. If you cancel your subscription, you won't get to play your games.

    People calling this cheap need to do some math. Would you rather pay hundreds or $$ each year and end up owning nothing or would rather pay few hundreds bucks every 5 years and own both games and console?

  • 1.5 megabits for "standard resolution?" WTF IS THAT, 480p with a bunch of artifacts? I'd like to see the magical compressor fairies that can deliver that in realtime, not that I'd even use it then since I'm used to running my game on 1440x900 minimum.

    Less than a millisecond of lag? Unless they've figured out a way to purposefully manipulate quantum entanglement there's simply no chance of that. The latency alone would be simply unacceptable even with massive investment. They'd need massive datacenters everywhere to get something like a reliable 50ms response time, which would not be acceptable at all for anything other than slow paced games.

    This a sad joke, and will be until perhaps 2018 when the next NEXT generation of consoles comes out and cloud computing applications have become common on the pc. No company has ever pulled a magical instant "game changer" out of their ass in the history of business, and it's doubtful it will start now. Revolutions take time, from Ford's assembly line to Microsoft's easy to use OS things might give very big advantages, but they only pay off over time with proper development and capitalization. From these guys rosy descriptions it appears they're far overextending themselves in a field that will strip whatever useful thing they come up from their dead hides and leave them to rot as a broken failure.

  • I agree with S1ing B1ade and a few others, I think this idea is brilliant and legitimate game changer. I've been a fan of this idea for a while. See below:

    A key point here is that this isn't likely a solution for hardcore gaming fans. Its just not. What it is, is convenient and more than sufficient for the vast majority of new and untapped gamers out there. While we may scoff at 720p, most just aren't going to care. Those that do will go and buy expensive video cards and computers to play. Latency issues, they will get worked out.

    What we are seeing is the birth of gaming networks. To be frank, I expected Valve to be in this position, not some unknown start up. Someone else pointed out the Netflix online viewing model and I think that's entirely correct. I LOVE watching streaming movies from Netflix on my HDTV. Is it full 1080i. No. Is it good enough. yes. Cable companies actually have the leg up here. They are the ones with the infrastructure, the bandwidth. Look for a cable company to snatch up Onlive.

  • Why is this concept so hard to believe? you've been able to do something like this via a remote console on the 'nix machines for forever. Yes the performance sucked and was limited but so was the idea of having a freakin 10Mb connection in your house at one point. The idea for this is just the old mainframe/terminal days yet again.

    It wasn't all that long ago that the 3d accelerator card was introduced and look how far it's come in that short time. I remember having to dial into a game service to play multi. join in progress?!? no way! that'll never happen.

    I personally look forward to this. If i kept track of all of the money i spent in the past 12 years on PC gaming hardware i'd probably cry at the total, and that was to just get solid high/midrange machines, let alone be able to play every AAA title at max settings all of the time in HD.

  • ok well this is inevitable technology, cpu cycles, bandwidth, and everything else will improve to acceptable levels for a company to achieve this, but the latency is the stumbling block that can't overcome SCIENCE AND PHYSICS for this to happen

    the solution wasnt' mentioned in the interview but I have a feeling it has to do with getting you a direct connection to their servers, similar to how cable tv and phone lines work, rather than going through a bunch of routers to connect to the internet. this will mean they will have to have servers in a bunch of places, but if this is the case latency would go down to acceptable levels, osmething like 30-60ms, and its not horrible i have a dell2408wfp that has 50ms more latency than average LCDs and I can play games on it fine. sniping is a bit harder in tf2 but it still makes this onlive technology at least feasible.

    XBAND failed but it worked for some people, it had 50ms lag, and just the fact that that worked for some people , maybe onlive will work for some people? and that would be huge. im very optimistic that this will happen but maybe onlive is a decade too early

  • Can someone explain to me how I'm going to be able to instantaneously stream 720 content to my machine when I currently have to wait 15 seconds for non-HD netflix movies to buffer?

    Obviously with a game you can't have a buffer, it would destroy the control mechanics, I'm just curious as to why we haven't seen instant video streaming if it's so possible with today internet connections. Are they making the point that you can have stutters here and there in a game which are more accepted in the gaming world where as with a video people would be outraged?

    And for those of you saying "Look at the video that proves it works", please don't be so gullible as to believe setup at a demonstration, wait until somebody at Shack gets one of these things to test from home and can give a real review.