Onlive: Video Game Cloud (PC/Console Obsolete?)
polycounter lvl 18
http://www.pcmag.com/article2/0,2817,2343717,00.asp
What do you guys think? Personally, I think the bandwidth just isn't there. That and this would depend on high quality of service lines from broadband providers. Just in example, our can go anywhere from 11mbs to 1mbs depending on time of day.
OnLive is a new gaming service, slated to launch towards the end of 2009. The core idea of OnLive is to make all modern games playable on any system. The actual heavy lifting of rendering, AI, and other gameplay is handled by big iron servers, which are loaded with multiple CPUs and high-end graphics chips (GPUs).
What do you guys think? Personally, I think the bandwidth just isn't there. That and this would depend on high quality of service lines from broadband providers. Just in example, our can go anywhere from 11mbs to 1mbs depending on time of day.
Replies
There are also times, though rare, that my internet/TV does not work, and I use things, like movies or video games, to help pass the time. This would not be possible if everything is streamed to me.
internet connections are not a constant to me
I seen this ages ago (its been knocking around for a year, surely?!) and it seems like a nice idea but not as a total replacement.
Pay-to-play fees sound like a painful inevitability, too. Screw that.
Also, there looks to be a gamepad. Where's my mouse and keyboard?
Not only that, but the idea of the internet is a decentralized network. If you do cloud computing, the server data may be spread, but its more dependent on centralized nodes. Terrorist just need to take out so many of those nodes, and poof! No net. Which could kill not only a economy, but communication and power as well depending on how interconnected these are made.
I do like my physical copies of games though, nothing like having a fresh box with the disc inside when its a new release
Besides, you are going to get the best performance on a system like this from games that are designed to stream constantly. Most games don't work that way. Only a few titles have actually been programmed to take advantage of constant and dynamic streaming. (Metroid Prime would be an example, I believe some of the Grand Theft Auto titles also work this way) For this to really work streaming would have to be efficiently integrated into every game.
They've been toying with this concept for a while. It's obviously very appealing to developers and publishers who are constantly combatting piracy and used game sales. World of Warcraft has proven that a system like this can be financiallly viable if you get enough people to buy into it. But they are forgetting that WoW is successful because of the persistent world it presents, and the social gaming experience that it provides. Providing on-line accessible games is not enough.
I'll stay skeptical til I see some more on this.
It's a neat idea but like others, I don't want to have to be constantly connected to the net to obtain anything I want access to.
edit:
ok, after watching the video of it 'in action'
..holy shit.. I'm less excited about being able to play a game from any system on pc or tv, and more excited about all the crazy community features. it sounds amazing!
LOL... the scenario you are so concerned about is the exact situation we have now. The entire internet runs through a relatively small collection of regional network access points. It's theoretically possible to take out a NAP and take down the internet for a large part of the world.
Anyhow, this is something that is inevitable. The market for audio on physical media is in it's death rattles, movies are on their way down that same path, games are next. I'm sure you will still be able to buy computers in some version of their current form for some time to come, but these guys are right, the next generation of home gaming consoles is likely to be the last.
Not really. I mean a economy would be crippled, but by no means up a creek without a paddle. You still can work with local intranets and have access to personal programs and games with the modern net even if it failed. We also haven't become so dependent on it that other forms of communication don't exist. This is changing though, examples. The closure of the PI. The loss of TV band to potential wide area wireless. The transfer of more radio unto networks versus FM.
OK YES! I ADMIT IT! I CHOOSE FOR JC TO DESTROY HELIOS!
I will state, this wont be really big time until quantum networks happen. When something can be in the exact same position no matter where that other bit is in space. Instantaneous data transfer. (Yes way oversimplification of the concept).
There is one huge bonus that we as artists are forgetting. If these are being networked rendered. No real poly limits and opens up the possibility of realtime radiosity with a few more multi core generations.
Hmmmm...so they are planning some robust community features with this thing? That is the proper direction to take a product like this. Get people invested in an on-line community, and the lack of physical media doesn't seem as important. Clearly these guys are more on-the-ball than those snake-oil salesman at Infinium. I read a bit more on their project, and apparently they are developing technology to stream the video over the internet. But that still brings up some serious issues with latency, and how it will affect gameplay, as well as how much sheer processing muscle they would need to brute force multiple thousands of users at once.
A service like this would have serious growing pains at launch, even with a limited install base. Still, I can't help but feel like they are approaching this in the correct fashion. This is like the Phantom done right.
bandwidth has never been the issue, there's 100mbit connections around, but latency will be.
I'm sure they guys at onlive are having fun playing this thing over their own LAN connection, but even with some kind of prediction there will be input latency, the worst of the kind.
pro:
* piracy made virtually impossible since all that the client gets is a "video" stream.
* developers don't need to worry about hardware incompatibilities. Consumer graphic accelerators will be a thing of the past.
* Devs just develop for the power the server can handle, instead of a zillion different PC configurations.
* pay for what you play. Its easy to just start up a game and if you don't like it. you don't pay for it.
* and in addition to that, the quality check will be high, since its all about making good content. If the consumer pays per minute
* for example games need to be good to keep the interest in it and make loads of $ signs.
On the other hand that point could lead to an insane up rise for the casual market. small fun addictive games.
con:
* monopoly position, could lead to insane per/minute ratings
* modding community will die without access to developer tools and sdk's
* video artifacts could occur. I need to see it in action live to judge how they steam the content. but if its actual video, you can bet
* on allot of artifacts when the signal is less strong then desired.
someone once said, the internet is like a highway full of trucks, you can have one truck full of packets, or a thousand trucks full of packets, so you can get a shitload of packets to your destination, but it will still take the same time as that one single truck would take.
bandwidth is no problem, thats why streaming video is no problem, it doesn't have to be realtime, latency however is limited.
The Phantom Console! And I even have a t-shirt.
Why would a company bear all the financial risk of purchasing the hardware to sell time back to the user? It makes no economic sense to me. What are you going to charge for this service?
Remember how slow the freaking terminal systems were at the library in the 80s? Those things were just sending text and there was noticable lag... why would libraries and companies switch to local hardware? Because it's faster and isn't limited to traffic or latency.
Why are video cards not all onboard video built onto the motherboard?
Why are content companies switching to swarm/torrent style download methods for large files?
Why are cpus going towards multi core?
What I'd like to know is who's funding this, and who wrote the business plan to sucker the money into coughing up the dough.
If this system of distribution is to work, look more towards stuff like peggle or puzzlequest to be offered, more than a CoD4 or a WoW.
It'd be like trying to control a satellite or space probe out near pluto from earth... the damn thing is moving faster than the commands can get to it.
and if it did the cost to set all that up for one user and charge them per month you're going to need alot of damn months to make your cash back . The amount of machinery required to run this for 10k users (which I would consider small) is crazy. And the other thing that strikes me as unworkable is that you'd need one processor and one GPU per user, so the cost for adding a user is relatively high. So the capital investment cost is high and will it be pay as you play or pay per month/year ? Presumably it would work on a flat monthly fee, but then I can imagine a fair use policy like there is on the net now.That would suck big time.
if they have a system that can service 10000 players, not all will be on at the same time allowing them to use the hardware on lets saya 2:1 ratio so they could get 20k subscribers this makes financial sense, you're halving the cost per user and making your money back quicker. If a new game is released on such a system it would overload, everyone wants to play but cant get on.
Finally tech doesnt stand still games always need more power so such hardware has a shelf life. Can you make your money back in this time? If you can you make enough for the upgrade to new tech ? Sounds really risky to me.
It kind of reminds me of the mobile telecoms industry. But in that industry cost per user is I think less and the equipment doesnt go out of date so fast.
And besides I love upgrading my PC and owning games.. I would never buy into something like that.. Its great of Intell. Property control, turning the internet more into something like TV, but I wouldnt worry at all.. not going to work anytime soon.
It's also important to note that this sort of system isn't really designed to replace PC gaming. It is designed to eat into the console gaming market. PC enthusiasts who enjoy upgrading their machines and modding games are never going to go for a system like this. But a lot of console gamers would jump on this first chance they get. All of the positive aspects of this service are the same kind that attract more casual players to consoles. The fixed hardware specs, never worrying about software upgrades, transparent patching, etc...
And you saw the little box that goes with it, right? They could ship that thing as a free bonus for signing up for the subscription service. They could charge $15 - $20 a month for the basic service, and nickel and dime the customer for bonus content and/or additional games. This could be something to watch out for.
This is going to be really interesting to watch. And I will definitely sign up for the beta, if just for the novelty of it. We have yet to see whether or not this can be made to work in a commercial capacity, but it definitely seems like they are serious about going through with it.
I think the fact that Steve Perlman is the principal behind this venture should lend enough credence to this not being vaporware. The guy was working on online gaming before the public knew the internet existed. Back when the most senior members on this forum were blowing dust out of their NES cartridges in their pajamas on a Saturday afternoon, he was trying to figure out how we could play RC Pro Am against eachother over the phone.
Look at this guy's resume... Quicktime, 3D graphics, the ubiquitous DVR, WebTV, online gaming for Coleco, Nintendo, and SEGA...
http://www.rearden.com/people/