OnLive will never be successful, especially with ISPs looking to cap everyone's bandwidth usage. | |
By 2013 there will be no ISP capping. Broadband access will be unlimited due to better technology. | |
Right...onlivefans.com...Have you looked into your magic ball and saw it become reality? | |
Well... damn. This is one year too late. Don't they know we'll all be dead from 2012 by then!? D: | |
It might not excel right away, as with all new technology, but it will work. That's all it takes. | |
what, you say that just because they have better technology the companies wont bother to capitalise on a capping system that will help them rake in the millions? P.S. | |
That's a bit pessimistic don't you think? I can understand how in your limited experience world seems predictable and simple but if you study ways in which business works you can see how strongly competition and free market influence the policies. Just because all you ever knew is bandwidth limitations is not reason enough to proclaim that is how it will remain for the foreseeable future. | |
It is very simpple logic technology will be better, but what do the users needed if the company lock that technology. I live in EU and don't have this kind of problems as such an act in our ISP would be terrible. But in america they will have to eventualy drop this cap or users won't upgrade to better connection witch mean less profit. So they have two options less or more money to make. And it is pretty clear that they won't go with the cap as this mean less money (lower connection>users pay less>less money for isp). | |
onlive will never work. no company will ever have enough servers to allow for smooth gameplay that has low lag for every body. the idea that you simply plug a controller device and tv into the internet to play a game is absurd. if i were you i would stop talking. now. | |
Well not really as I do know basic (maybe more) of ISP, video streaming (lag free streaming of course), and of course the hardware. This is where you think that can't work. Look at this most game nowdays use 2 cores max, but what to to with other 2 cores, in this case simpl just use the other 2 cores for some other game. And let put in perspective: I have a ISP tv (tv chanells via internet) and for a full HD it has 1 second lag and it use only 1mb internet line and it streams via mpeg2 standart. So if onlive is completly different it can have lag free (video standart). So maybe you think it is imposiblle but I know some thing about this type of deal and I am sure it is possible (worked in ISP) the only problem in onlive is one thing resolution. I don't think that people will go from HD resolution to SDTV. Ow and if you think that you know more than I that give me evidence that it is imposible. | |
I ran some calculations and came to the conclusion that a 20mbit connection allows you to play on 1024x768... at 1.05 fps. Movies can be encoded for smaller file size but games, due to their unpredictable nature, can't so you are stuck transmitting the entire image in a lump or encoding it after generating before sending and then decoding it before displaying, lagging everything and crippling the server(encoding video takes about as much cpu usage as the average game). Yerli is talking bullshit again, nothing new to see here, move along. This is the guy who loudly proclaimed that the ammount of pirate copies of crysis exeeded the number of computers able to run said game at that time by a 150% margin. | |
Asehujiko Your caculation is falls (if you know some basic of this thing you would si that your caculation is more BS than entire world). I worked in ISP and the isp has 1920x1080 resolution for tv program (no upscaling on the costumer end) and it has 1s lag and uses only 1mb connection. And this is done with mpeg2 (or mpeg4 it depends what is the connection max on the costumer side). Like I said before give proof that this is imposible ow and not your "calculation" but proven facts. | |
And, in 2013, when they are still failing, everyone will have forgotten that they predicted success by then. The perfect plan. | |
But... Will they accomidate controllers? I know that all my experiences of any game with a controller on a PC were horrible. I'm sorry, but the level of fun with most games on the PC, rather than a console, drops immensely. If consoles die off, I'm not going to be very happy. | |
You can still use mouse and keyboard, and if you want to play with mouse&keyboard you just need browser plug-in and no console. | |
You misunderstand- | |
Hang on... if you're streaming a game, that means you have no physical copy of it, right? Or at the least, you don't have all the necessary data on your hard drive to run it? So, what's to stop this OnLive simply becoming the next step in the 'games as rentals, not as possessions' business model that all the publishers seem to be following these days? | |
Well they will rent games, but they will pay devolopers for each rent they give. | |
I'm sure this will be Great! Just as soon as we get fast, RELIABLE non-capped internet. -_- | |
Speaking as someone with almost a decade's worth of IT and tech infrastructure experience, to say that this could work in 2013 is actually fairly feasible. The infrastructure to provide such a connection is expanding at an exponential rate. With services like FIOS out there that promise the next step in Internet connectivity and speed, it's not really out of the realm of possibility that a service like OnLive could thrive in 4 years time. However, the main problem is the ISPs. Cable and DSL providers like Comcast and AT&T have taken the place of AOL back in the days of dial-up modem connections. They're confident and have become a household name for the necessity of interent connections, so they are able to do things like bandwidth cap with impunity. But as new, faster, better services end-run them, we might see a relaxation of those restrictions, or have them be overtaken entirely as the demand for streaming services like this and Hulu and more increase. Basically, the potential is there, but right now? I'm extremely skeptical of OnLive's claims of immediate success. Their presentation at GDC was venture marketing 101, and as many gamers know, hype rarely translates into actual reality. | |
I just wanna play Crysis 2 or whatever, not Warhead | |
There's a large difference between encoding-transmitting-decoding pre-recorded images(tv services) and trying to cram an entire screen worth of graphics directly through the internet(as is required for gaming). I know that encoding can make things smaller a couple hundred times but the resulting lag like your 1k ms would make the game unplayable. | |
I am amazed! It looks like something that will actually be difficult to pirate! | |
I want to see someone do realtime video encoding of resolutions above 1080p while running crysis on maximum settings 500,000 times over on hardware made this century. You see, you can compress TV programs, and they do not require much server-side graphics processing, but a game like crysis at maximum settings, being broadcasted to 500,000 people would require unfathomably huge amounts of both processing and bandwith. | |
Look onlive doesn't mean to give above 720p (1280x720) and you must have 5mb connection (witch I don't think has a lot of people in america), so games like crysis doesn't need that resources. And about resources it still doesn't give that a problem. From what I understand they devolped a program that use the entire procesing power of each gpu, cpu, ram... So that mean if one game uses only 60% of power the other 40% will go for different user (so if you have a god hardware with enought for everyone that this is no problem). And the bandwith is not a problem for this kind of company they can just order a few 1gb connection and it is solved. From now on I won't post same thing over and over, so when you give me proof that it is imposible I will explane the whole thing to every detail. | |
I've been following the OnLive story from the start. While I agree that it is still too early to predict success, I'm amazed by the veritable deluge of nay-sayers predicting its failure. The theory behind OnLive is sound. Whether it will work in practice, be it this year or later, remains to be seen. However, it seems people are falling over themselves to criticize it. Why might that be? As with any new idea, OnLive comes with a proposition to change the rules. The change it proposes is so major that, if it works, will deserve to be called a "revolution". And, as in all revolutions, there are those who do not want it to happen, namely everyone who stands to lose profits by people switching to OnLive. In case you don't see where I'm going with this, OnLive's proposed revolution could spell doom for console makers, as well as companies such as nVidia, ATI etc. I'm not surprised that Crytek, whose game Crysis is single-handedly responsible for a hefty amount of system upgrades, are keen to criticize OnLive. "Do the maths!" | |
"A few 1gb connections" don't solve a bandwidth problem, because the communication is two-way - from the source, to the destination, and then back again in acknowledgment. Basic networking. Delivery from the source can be extremely fast, but if your machine has a dog slow network card, or less than ideal infrastrucuture, guess where the bottleneck is? Streaming content is currently a bandwidth hog, and is exponentially going to be more of an issue when it requires input from the player. The infrastructure out there right now can handle some of this, but in large, massive chunks? Color me skeptical. Whatever "new protocol" they claim to have developed in order to reduce the lag still doesn't take into account that the client side, given that they might have an underpowered machine, might have a NIC that only supports 10 and not even 100Mbps. If you're looking for "facts", look no further than the proven way that networking works, not at whatever they are claiming will make the server shoulder the burden of the networking communication. I see OnLive as working similarly to streaming video - the server hosts the file and delivers the file and plays it, and the client just asks for it. But the client still needs to have some means of communication that can handle it. As I said before - latency and lag is a two-way street, so I'm extremely curious to see how they figure on handling this Pentium IV with an old Kingston card I have sitting at home. Delivering the same or relatively playable quality to a machine like that? I'll believe it when I see it. Like Crytek, I believe it's possible. Just not now, or any time soon. | |
Yes you have a point (in a way), but why in the world do you need 100mb NIC if you have a 1-5mb connection? You don't benefite anything if you have a 1gb NIC if you have a 1mb connection. And still onlive clearly stated that for a 720p resolution 90% of the tame it only uses 2mb of the 5mb recomended the 5mb is only needed in few moments. But about infrastructure I can't really say if it will handle as I only know europe ISP how it works. But I really doubt that even in america ISP give more speed than the bandwith can handle. Becouse if this would be true than yours internet connection has to be break daily, becouse users uses already to much bandwith. And the information the users send to onlive is only a few kb/s. Ow and I already stated that I worked at ISP (in europe) and I know how network works. | |
I thought that one of the selling points for OnLive was that people with awful hardware could play on very high graphical settings. That conflicts with your claim that "they devolped a program that use the entire procesing power of each gpu, cpu, ram... So that mean if one game uses only 60% of power the other 40% will go for different user (so if you have a god hardware with enought for everyone that this is no problem)." That statement says that graphics processing would be done by the end-user's hardware ("God hardware"?), not the serverside hardware, which makes absolutely no sense at all. If you mean that the graphics processing was done serverside on multiple graphics cards, that's been around for ages (SLi and Crossfire) and not exactly efficient or new. Also, you did not address the issue of realtime video encoding of (what is apparently) 720p. In order to have a relatively low-lag stream of video, it must be encoded to reduce size. However, as we are dealing with video that is generated in realtime, we cannot encode it beforehand, so we must either have a massive internet bandwith, or the server has hardware that makes the i7 look like a pre-pentium CPU. | |
i've said this before and i'll say it again, this has "The Phantom" written all over it. Which is oh so definetely not a good thing. | |
I mean that they dovolped for there server to utilize all power. About realtime video encoding you should know that this kind of thing in the last 3 months it was a lot of progres now with the help of cuda you can do 1080p realtime video encoding. And yes i7 is like a pre-pentium cpu in this thing as if I remeber it right the best procesor has a little more than 100gigaflops power, but the gpu they have almoast 2 teraflops of power (a single card) so they are about 15x more powerful than cpu. So if they utilize gpu power thay can have 10-15x more power rather than using a cpu. | |
Extra Punctuation
MovieBob: Intermission
Experienced Points
First Person
Miracle of Sound: Encore
Exploding Barrels
Loading Time
No Right Explanation
Backchat
Drinking Games
Escape to the Movies
Escapecraft
Escapist Podcast
Feed Dump
GinxTV
Jimquisition
LoadingReadyRun
Mahalo
Meatbag and Co
Media Sandwich
Miracle of Sound
No Right Answer
Pro-Gamer Gauntlet
Reviews
Space Janitors
Tales from the Table
The Big Picture
The Escapist Presents
The Retro Show
Trailers
Unskippable
Zero Punctuation
Max Payne 3 Review
Company of Heroes 2
XCOM Enemy Unknown
Black Ops 2
Diablo 3 Review
TERA Review
Firefall Beta Preview
Starhawk Review
Critical Miss
Name Game
Crytek Predicts OnLive Success, in 2013
German developer Crytek has done some research into the technology behind streaming videogame services like OnLive and says that such systems could in fact work - in about, oh, 2013 or so.
OnLive, a "cloud computing-based" system that will make use of centralized servers and streaming video to offer on-demand gaming without the need for high-end hardware, was revealed in late March. The system will work with conventional mid-range PCs, including laptops and netbooks, or through a "MicroConsole" provided by OnLive, prompting co-creator Steve Perlman to predict that the current console generation would be the last one.
The announcement of OnLive and similar services has prompted a great deal of discussion about streaming games, with most observers expressing strong skepticism that such systems are viable. Crytek CEO Cevat Yerli, whose game Crysis was used to demonstrate the technology at the 2009 Game Developers Conference, said a service like OnLive will be technologically possible, but not anytime soon.
"We had our research in 2005 on this subject but we stopped around 2007 because we had doubts about economics of scale," he told GamesIndustry. "We saw that by 2013 - 2015 with the development of bandwidths and household connections worldwide that it might become more viable then."
The weak link, according to Yerli, is the broadband technology necessary to provide smooth, uninterrupted gameplay, which he claimed isn't yet in place. "It doesn't take a lot to make a video-based renderer, but what you need is the right infrastructure that is beyond the technology we have, it's more like cable net providers and communication networks," he said. "They have to provide fast bandwidths and connectivity in order to allow such technology to excel."
Yerli admitted that despite Crysis Warhead being one of OnLive's headline titles, he hadn't actually seen the system up close, and said that despite his reservations he hoped it worked out. "I want to see it myself. I don't want to say it's either 'top or flop'. I hope it works for them because it could improve gamers lives," he said. "The technology of video-based rendering is not actually a very new concept but they do some things that others didn't do before so it will be interesting to see."
Permalink