GDC: OnLive to stream games to TVs, PCs News

How would this work properly?

Any lag and the game would suffer something worse than framerate drops, something like the game would freeze up momentarily while it caught up with itself. How can you hope to get a solid 60 fps for something like Call of Duty 4 for example?

Sounds like a terrible idea to me, especially in this country where internet bandwidth is shared with other users nearby so you can't even guarantee someone will get a constant 8 Mbps for example.

It is a terrible idea in any country or area that doesn't have stable broadband... call me when you find that mythical place where latency is not an issue and then maybe this might work.

The input lag on this will be horrible, and there's no way around it.

It doesn't work properly in any kind of real or usable terms. Looks great in the controlled environment of a trade show though.
They pimp this kind of stuff every now and again but it never sees the light of day.

If they've managed to get around the problem of lag then they've done a very good job.

this wont work for games like call of duty 4 yet. lag is introduced in getting the user interaction to the server, performing the action, encoding it into video, and the user seeing that video, even on a closed network this lag is noticable to the end user. relying on isp's isnt going to be great.

It's not a terrible idea - it's a fantastic idea. Whether it turns out to be fantastic in practice is another matter.

If it was rolled out today, I doubt many people would get much benefit from it at the highest resolutions (needs 5Mb) but with BT's rollout of faster BB, it could be feasible in a few years. However, some lucky folks certainly could use it, and if it worked for them the publishers would continue to support it while the infrastructure caught up.

Speaking of publishers, if it didn't work, I doubt so many big ones would have registered their interest in it.

IGN says they were sceptical, but tried it in an admittedly controlled environment, and it worked. The open beta in the summer will be the test of true functionality, IMO.

The probable nail in the coffin, though, could be download limits. It would be easy to eat up a month's limit in a few hours.

Incredibly odd idea, but very cunning.

60 fps is one frame every 16.6ms, if I'm not mistaken. They claim to be able to receive and calculate the user interaction per frame, and that takes about 1ms, then another 1ms to encode and compress that frame, leaving 14ms for the data (only 1 frame, remember) to shoot down the wires. I suppose that is do-able if the connection is sufficiently speedy and their compression algorithm is as good as they claim.

This will never work for so many reasons.

Its probably too early for it right now but I do believe that this is the future and only possible way we could have a future convergence to a single gaming format.

This will be the biggest thing ever. The "One console" future has come.

btw, How is connection-speed coming along in UK/US/rest of the world?
In Sweden and Japan, the norm is getting close to 100mbs. Its what someone would get if they sign up for internet today.

someone mentioned download limits. lol, havnt heard about that since 2001-ish

Speaking of publishers, if it didn't work, I doubt so many big ones would have registered their interest in it.

Publishers showing interest doesn't mean squat. Remember the Phantom Game Service?

Old idea, really, they've been doing this stuff for desktop applications for more than ten years.

Lag is indeed the big problem here, and I suppose they will have to deply quite a lot of servers in lokal markets. I don't think I'll be able to play on a server in the states, for instance. Local WiFi is probably out for this kind of setup, reintroducing the cabled home.

But seriously, think about it: this eliminates piracy, reselling and (potentially) even the need to develop for several platforms. Server hardware could be upgraded seamlessly, unnoticed by users, and it would give telecom companies more products to sell: both hosting services, maybe the price integrated into broadband subscriptions, and even create increased demand for advanced broadband products such as SHDSL and "ADSL 2.0". Oh, and this delivery system would obviously work great for fims and music too.

The question is: will it work lag-free - and how about sound quality? 7.1 channel surround sound eats up a fair bit of bandwidth - possibly more than the sceen image feed.

This is the future. When we have flying cars and take all our meals in tablet form, this will be how we play games. Until then, I'd rather have a box under my TV or on my desk, thanks.

"If they've managed to get around the problem of lag then they've done a very good job. "

By definition you cannot get around lag. The speed of information is not infinite. Like somebody said... the input lag will be a big problem. Even a 10 millisecond delay on key input is noticed by the human brain.

This will never work for so many reasons.

And I see you've done us the honour of listing them all here...

Oh and that was more than 20 characters by the way.

This is perfectly feasible. If people can stream movies/HD movies from the net, then this could work too. I'm more interested in the pricing and the fact that you wont actually own a copy of the game itself. Sounds more like renting or pay-per-play...

> If people can stream movies/HD movies from the net

Totally different technology.

If they can get the servers close enough to your home they might be able to do this. Gonna cost them though.

>the norm is getting close to 100mbs

On paper.

In reality it is the sustained speed that is important for this kind of service (all realtime, you can't buffer anything). Very few broadband providers can guarantee sustained speeds like that (at the moment).

Of course publishers are keen on this, more $$$ for them.

We waits and sees

Streaming a movie is a one-way deal.
Hell, even remote desktopping over a gigabyte LAN connection gives unreasonable amounts of input lag for a lot of apps.
Can you imagine trying it on your average broadband connection with the contention ratios we get?

/writes letter to EA and Ubisoft about his "lead into gold" idea

Certaily great for renting games. They had a button labeled "Buy/rent" in one of the screenshots.

If it works well, I would expect that this tech will eventually be integrated in TV sets. Then the guys behind this will be rolling in money. Good for them - and good for gamers, too, and publishers, telecoms and developers. That's the real next gen of gaming, but extremely bad news for MS, Sony and Nintendo. Bad news for Intel, nVidia and AMD/ATI too, come to think of it. That's some high-powered resitance right there.

Its probably too early for it right now but I do believe that this is the future

Absolutely, and not just for games, but for all kinds of powerful software. Eventually the only computer you'll need will be a low-power terminal, while all the grunt work will be done on massively powerful servers somewhere.

It will be fricking awesome when it arrives. I'd love to try OnLive out - I suspect it hasn't got it perfect yet, but we get closer every time with each step.

This is the future...

You've got to love the "This'll never work" nay-sayers. The same kind of people who would've said, "You've invented a wheel? Aye right, that'll never work!". History has proved these kind of people wrong over and over and over and over again.

MoFo +1

Nay-sayers, read this and it will make more sense I believe. http://kotaku.com/5181300/onlive-makes-p...

I agree that we need to see it put in practice to know just how well it works, but I do believe this will be super big.

I could see this tech working for games that don't require fast responses, say, a point and click adventure or a football manager game but the slightest bit of lag would ruin the entire experience. Remember it's nothing like gaming online, where only minimal data is transmitted back and forth, this requires *everything* to be streamed including compressed audio and video. Considering I can rarely stream relatively poor-quality HD 720p videos where I live without them pausing, playing games this way would be totally out of the question.

I guess it isn't a terrible idea as such for the reasons others have stated but I don't believe the tech will be there for years yet, especially in this country, just as I don't believe digital distribution of games will become mainstream for years to come either. Limited bandwidth is a big problem, especially in areas like mine where it is shared with other nearby users. I'm on an UP TO 8 Mbps connection at the moment but even with 24 MBps I reckon this kind of service would suffer from lag.

Also how is the actual games hardware allocated? Will it be one machine per person or is one machine shared among a set number of users? If it's shared then doesn't that create problems as far as processing/work load goes? If they aren't then isn't having one machine per user going to be very expensive?

I would like this to work. Imagine never having to buy a new console or upgrade your PC ... all games available regardless of publisher, nice dream.

It's quite an elaborate venture capitalists scam. They should call it Phantom OnLive.

Like Olemak says, another reason this will never work is because the people who will need to provide the technology behind it will be the ones that would take the biggest hit.

Seriously think Nvidia and Intel will be happy supporting this kind of thing when it eliminates the need for consumers to buy their latest graphics card and CPUs to play the new games? Think it'll stop the battle between MS and Sony to hijack your livingroom?
The technology industry is mercenary, this kind of commune gaming will be killed by the people they need onboard to make it a success.

IGN (I know, I know) have a nice article about the service (sorry EG, but this article says very little).

http://uk.pc.ign.com/articles/965/965535...

From th article:

The current solution only introduces one millisecond of lag to encode the video, which alone is completely unnoticeable to you. Obviously, a fast internet connection is required on your end to stream the gameplay video. A 1.5 mbps connection (which is usually what base-level DSL is rated at) is required for standard-definition video (480p), while a 5.0 mbps connection is required for HD (720p). The actual necessary speed is a tad less than advertised, so as long as your provider says you have these speeds, you should be OK.

Also:

Do the games run at 60fps? Technically, yes, but the video stream makes it feel less so. They're still smooth, but Burnout wasn't as brisk as it is on a PS3, for instance. But make no mistake - everything we tried was completely playable (and most importantly, quite responsive), and being that you're able to play these games without any dedicated hardware, that's a huge, huge thing.

If they were happy with a fast, twitch game like Burnout, even if was over a LAN, the potential is clearly there.

It's got respectable names behind it with experience in the field, so I don't think it's likely to be a scam.

http://en.wikipedia.org/wiki/Steve_Perlman
http://www.mobygames.com/developer/sheet...

Pah, playing PC games at 720p? What century is this?!? :O

720p is so 1995... ;)

What's with the hate? This looks fantastic. Big publishers signed up too!

That IGN write up doesn't sound very encouraging... games run at 60 fps technically but don't feel like it... games play quite responsive? Sorry but that isn't selling the idea to me at all.

Well, if you already own a quality PC then obviously you can still play at whatever resolution you'd like. In fact with the way these things work, it might actually be cheaper to maintain a good PC and buy your games than subscribe to OnLive's service.

However, for people without the knowhow to do this, or for people who want a laptop, or for people who only run Macs or netbooks, I'm sure 720p is acceptable.

It's an iteration. An evolution. We didn't instantaneously stumble upon quad-core processors, or 20Mb broadband. If it's 720p for now...fine. It's just as good as most people get from their PS3 or 360 and most of those find that perfectly acceptable. As the tech is refined, it'll get better.

I'm not sure why people are against this (from a gamer's point of view). What are the real downsides to us?

Hmmmm clouds are going to be the future of PCs full stop, just have a monitor and a network connection and go to a server cluster to browse tinternet, run all your apps, store your files etc.

I dont see how it would work for gaming though as people on here have already pointed out. Lag will always be an issue and until someone breaks the speed of light it always will.

@Darren but it's in closed Beta, i.e. not quite finished yet.

Krelle: btw, How is connection-speed coming along in UK/US/rest of the world?
In Sweden and Japan, the norm is getting close to 100mbs. Its what someone would get if they sign up for internet today.

UK is variable, but cable modem is mostly pretty good: 20Mb/s or 50Mb/s depending on where you live in UK with 100Mb/s in next year and 150Mb/s was announced today as being available within two years. ADSL is a few years behind though.

US is my experience is terrible, especially if you are outside the cities. I have friends who live in CA, not even that far from San Francisco and Silicon Valley, and they cannot get broadband at all.

>Nay-sayers, read this and it will make more sense I believe.

We know how it works dude (VM ware has been around for ages), a lot of us are saying it will be hard to make it work good enough for a lot of games. The low resolution being one of the problems together with the input lag. The streaming lag is a non-issue, that is just a question of having enough horsepower to encode and enough bandwidth to stream. The real test is the input lag on an internet link.

Will people play Starcraft 2 in 480???

It will probably work for quite a few games...

Ofc this is probably the future of all applications eventually, but I don't think we are there. It has not really worked for normal applications and games are even more demanding but somebody needs to roll this out to get the ball rolling. Maybe next gen consoles will have stuff like this build in, it will only get better if the client can process some server stuff by itself.

I read a preview on IGN, apparently lag isn't an issue. They are playing Crysis on a notebook already. I truly hope this succeeds, and I never have to buy another console again or upgrade my PC. Like someone else said, you just need a monitor/TV and fast internet connection.

No more fekking clutter, no more fekking disc cases everywhere. This news made my day.

This is a fantastic idea. Once the kinks are worked out, I don't see any reason why it couldn't replace the quaint notion of buying a frail, expensive, hot, power hungry space hogging console unit under your television.

If I were Sony or Microsoft or dare I say it, google, I would be investing real money into cargo crate style server racks filled with cheap, fast, reliable blades with commodity graphics cards mounted that could be installed in or near DSLAMs in exchanges, to minimise ping time and network congestion, and looking at partnering with ISPs to provide XBL style gaming services and game rental.

No more expensive console lifecycle changes, much much less direct customer support and one hell of a simplified development model.

I love the notion that technology companies are interested in making your life easy and convenient.
They want to make money. MS and Sony are seriously not going to go for a model where that can't sell you as much of their branded bits of plastic as they can manufacture.

I think this could potentially lead to much better looking games more quickly.

I'm not sure why people are against this (from a gamer's point of view). What are the real downsides to us?

I prefer my data to be fully under my control and not on some cloud somewhere as long as I do have a choice.

"I love the notion that technology companies are interested in making your life easy and convenient."

I will give them my money because they can make my life more convenient.

Agreed that it is probably the future. It will make games expensive for the consumer though, as (a lot more so than now), the entire hardware will have to be financed with game sales and rentals or subscription fees.

Wastelander: Tech companies would adopt this precisely because they would no longer have to sell you their expensive bits of plastic for under your TV. Bits of plastic that cost them billions to develop and that then sell at a loss in purely manufacturing terms for years. Using commodity tech allows them to drive the rich seam of profits in service provision.

Unless we're talking Nintendo, of course, who love to overcharge you for their bits of plastic.