 |
|
|
| |
DICE 2010: OnLive's Perlman Showcases Cloud Gaming Service Improvements
by Simon Carless [PC, Console/PC, Exclusive]
|
|
| |
|
February 18, 2010
|
| |
Game streaming technology developer OnLive showed the progress of its upcoming service by way of a session by founder Steve Perlman at the DICE Summit 2010 in Las Vegas, with demonstrated features including a new version of the beta, an iPhone version of the client, and new social features.
OnLive, responding to what it perceives as the demands of modern consumers, allows users to play high-quality video games by rendering graphics remotely and streaming the frames as compressed video -- meaning no expensive game consoles or advanced PC graphics hardware is necessary on the user's end. Perlman cited a Sandvine study released late last year, revealing that real-time entertainment internet traffic is now 26.6 percent of total traffic, up from 12.6 percent in 2008.
In fact, he said, the given delivery medium is becoming less and less important. People no longer say that they're going to "watch a DVD," for example; they simply say they are going to watch a movie, which could be delivered in any number of ways, including streaming.
But, worryingly, Perlman noted that many of these industries outside of games don't actually make a big profit on streaming media. Even major sites like YouTube have seen major profit challenges.
And video games are a different beast entirely. A large file is generally needed before a game can start loading, games are tied to specific devices like consoles, and online games can't be pirated.
Still, Perlman believes that, even though lower-footprint Flash and iPhone games are much closer to being delivered instantaneously than their larger-scale cousins, their monetization model is much worse. The answer? There's a need for "instant gameplay of high-quality games," he said.
To achieve that, the consumer must be focused on the game, not the client device, Perlman said, while the Internet mitigates issues with piracy and used console game sales. OnLive is predicated on that model.
"If the money [from used games] is not going back to the people making games," Perlman offered pointedly, "in the end the games are not going to be as good."
Perlman demonstrated the new beta version of the cloud-based system, which includes number of social elements and will roll out next week. While booting it up, he joked that the service ran smoothly "if you are using a legitimate copy of OnLive" -- a reference to an unauthorized OnLive beta review that surfaced, apparently outside the beta's geographical catchment area.
OnLive has been doing years of testing in thousands of locations and homes across the U.S., and claims it has created the "first [video] compression algorithm in the world which is adaptive to the nature of the connection" -- meaning it changes based on the user's type of internet connection and other factors, such as the degree of packet loss.
The firm currently data centers set up in the San Francisco Bay Area, Dallas, Atlanta, Chicago, and the Washington, D.C. area, It routes its streaming traffic through multiple internet providers to find the quickest way to each individual user. The DICE demo was streaming from the Bay Area data center, roughly 500 miles away.
Showcasing a "major progression of social features" on the platform, Perlman demonstrated that users can now see friends and friends of friends playing games in real time through video streams, and send them messages while they play. Players can easily capture the last 15 seconds of any game and share them with friends by way of an Xbox Live-like social interface.
All OnLive games load swiftly since they boot off RAID arrays in the data centers, but Perlman pointed out that games incorporating the OnLive SDK load even more quickly. He backed up the claim by booting up Unreal Tournament III, whose title screen appeared after only about a few seconds of waiting.
As a final party trick, he showed Crytek's notoriously-demanding Crysis running on an iPhone version of the client. Like other OnLive games, it was actually a full high-definition version of the game running on the cloud, but resized to fit the iPhone's screen resolution. Perlman did admit that some combinations of games and platforms might demand considerable control redesign to ensure practical playability.
Perlman said "hundreds of thousands of people" have signed up for the OnLive beta so far, and the company is ramping up new server deployment, with new servers going in this week. Declining to state what exactly is coming next for the service, Perlman concluded with, "We're going to have some things to announce soon."
|
| |
|
|
I can't wait for this, so long as i can afford it. Awesome idea, awesome.
And this still seems to be the problem to me, it's not, that I think, the connection couldn't be fast enough or responsive enough, it is more, that I can't imagine, that OnLive is able to run servers für "hundereds of thousands of people". Even with specialized versions of the game, you need an awful big server park, to compute "hundereds of thousands" Crysis games simultaniously.
Perlman is absolutely right on this point. As I point out in my most recent blog post, for most game developers, the iPhone will be a money loser.
@John
I don't think OnLive will be as revolutionary as you think - at least not right away. But in the long run, the technology behind OnLive will change the way games are consumed. Eventually, cloud gaming will be deployed on a large scale and we will probably see TVs designed with OnLive-type integrated gaming capabilities. Until that happens, I see OnLive's market as hotels, airports, and other public venues. It may also enjoy some success as a smartphone client, but I see that market as very limited. Most smartphone users will be satisfied with low demand casual games that are already in plentiful supply. Most iPhone users don't even know what Crysis is.
"Even major sites like YouTube have seen major profit challenges."
Youtube is something of a special case - there's more than likely a little bit of creative accounting there to minimise the risk of lawsuits from the media companies: don't bother suing us, we're broke. However I'll happily agree that the general streaming landscape isn't particularly profitable: bandwidth costs money, advertising revenues aren't great and people are generally resistant to pay-for content.
Which begs the question for OnLive: how are they going to handle this challenge? People may be more willing to pay a subscription fee for streamed gaming, but they also tend to play games for longer (Youtube videos are generally limited to ten minutes apiece; the average video game has 10-15 hours of content - or far more if you're dealing with MMORPGs) and there's higher up-front costs (game licences, high-end, 3D-capable hardware, etc).
"If the money [from used games] is not going back to the people making games in the end the games are not going to be as good."
At least some of the money does go back. Trade-ins are used to subsidise the purchase of new games and the profits from reselling these titles helps to keep game shops open and is used to subsidise special offers and promotions of new games. It can be argued that there's profiteering on used games - especially when it comes to recently released, high profile titles - but there is an ecological cycle there: new games turn into old games, which are then used to buy new games.
I can't help but think that if the games industry ever does get to the point of a "perfect" lock-down (i.e. no piracy, no secondhand market), they're going to be sorely disappointed at the levels of revenue which don't materialise - there's actually a good chance that overall revenue will drop.
Finally, beyond the Ars Technica article (which was on a borrowed key and in uncontrolled conditions), I've not yet seen anyone discussing their experience with the OnLive client. Hundreds of thousands may have signed up for the beta (if I was in the USA, I would too - out of curiosity) but how many are actually *playing*?
But a service like OnLive requires you to live in a rather well connected area with a great connection and preferably close to an OnLive "cloud center" which will likely be located in big metropolitan areas.
Yet one day, when the global network is truly all fiber optic, then it will definitely rock!
"It not just piracy that is hurting the game industry. it is stores like Gamestop. Every time they sell a used game, that money never makes it back to the developers, so basically like pirates one game can be shared by multiple people."
Strange argumentation, as long as I can think, there were stores for used books, records, CDs, comic books, etc. Where is the difference?
No author, musician or comic book artist ever complained about the fact, that they don't profit from used sales. It didn't hurt the book industry or the comic book industry or the music industry. Only in the video game business people can't stop complaining about used games sales.
In the past, I bought many new games, I bought many used games. I bought many new books, comics, CDs, etc. I sold many of them over the years.
You say, that's like piracy, cause "one game can be shared by multiple people", so I think you belong to the category of people, that don't lend their friends books or DVDs? Cause that is just like piracy. Where is the limit? Do you let your colleague read your newspaper in the office? That's just like piracy, cause one newspaper can be shared be multiple people.
The bashing of used games sales is just a red herring of the industry, cause they are in trouble, caused by not enough creativity and the focussing on a to small customer group, that starts turning away from games.
Agree that major uptake will likely take significant time, but there's another major impact (vis-a-vis your comment @John). And that is the fact that OnLive will give gamers the ability to *really* try games at a much deeper level (vs a trial version) and as a result it'll be the engagement of the game in the first 2-20 minutes, rather than the marketing campaign or reviews, that determine whether someone sticks with it.
I'm still skeptical about the service's ability to scale (during peak evening & weekend hours that get high traffic already) as well as the ability for OnLive to deliver the service for lower than the cost of the bandwidth to the consumer (plus the amortized cost of the datacenter build-out).
But - all of that being said - digital delivery is the wave of the future. The best move EA could make is buying OnLive, or cloning it once successful. Same for Microsoft - buying or cloning it as the core value premise for Live on the Windows platform.
The ability to play Crisis and other games with system requirements much higher than their computers can currently tackle with nothing more than a browser and the system they're using is gonna be the big thing that changes everything.
My thought is, all those people who can't or won't upgrade their computers to the system requirements for said games won't need to, but yet they'll be able to play games they couldn't. That's huge.
That means people don't have to upgrade their computers in any way. That's phenomenal, and that will get more people playing more and different types of games.
My fear is that some giant company/s buys the company out and the public never see's it come to light. Sort of like the oil companies buying out a person for creating a 200 mpg carburetor so they can sell more oil.
ŹŹ It is annoying me. Really.
Anyway, let's write again..
If you people all hate Gamestop, convince them to move to my country, we would love it.
And you can full expect that RAID structure of the servers will be dynamic so that that there will be servers that are set up in a hive structure so that when additional feeds of Crysis are need as demand grows the game code will be copied and fed to end users instantaneously.
In order to be profitable they have tie demand to cost of meeting that demand. So, the more games you consume and longer you play them will cost more. Just like I said that more resource intensive games will cost more too.
I think it is concerning that the Data centers have to be within 500 miles of end users. This means that the cost of the service will be also controlled by only allowing people who are in close proximity and are located in densely populated areas to have access to Onlive. This is is similar to the behavior of phone, DSL and cable companies who don't want to serve rural areas due the high cost of having to provide service per user.
I doubt that Onlive will be available to anyone who sports and internet connection. They have to approach the ability to make a profit on this service with common sense. Otherwise they will collapse very quickly due to the high cost of setting up the service. They won't be setting up Service centers for a long while or if ever in Oklahoma for example.
I don't think that the question is whether or not Onlive cloud service can be done. The question is whether the model can self sustain. That is why the beta program is so important. The beta program shows them the level of interest within a certain geographic area that can actually sustain itself. It gives them a ready made list of end users and their geographic locations.
I am sure that they will be mining the beta sign up for virtually all of their start up users. This is no different than the process on which MMO's like Global Agenda are built upon. Globa Agenda's beta program was really just a giant marketing machine. However, Global Agenda's beta sign up users quickly realized that they really weren't beta testers but they were merely members of a client list to be tapped at rollout. Many people who thought they were signing up for an opportunity to participate in the beta realized that they never would get to test the game at all and suffered sticker shock when the details of the MMO's pricing structure was revealed and how limited the game world really was.
I think that Onlive's beta pool will be similarly affected. Which could ultimately be Onlive's undoing. I mean think about it. Do you really think that Onlive can afford to provide free access to hundreds of thousands of people who signed up for the beta? This means that Onlive is even on more precarious ground than Global Agenda.
Who knows though, maybe it will catch on. maybe someday I'll be complaining to customer service because my introductory rate to OnLive went up... More than likely though I'll just move onto something else to do with my free time.
It was funny though that as soon as Gamestop admitted that about 75% of their profit came from used games, Best Buy, Amazon, Wal-mart, and Toys-R-US announced starting of programs to buy and trade used games, luckily for the industry most of these programs fell through.
I don't know how it is in the USA, but here in europe, you find for every CD shop, that sells new CDs 2, that only sell used CDs. Second Hand book shops play a strong role in the book selling market and almost every comic shop I know sells more used comics, then new ones. Some of them for ridiculous high prices, from which the creators or the publishers get nothing.
In fact, I am happy, that there is a thing like used games sales, cause otherwise, I wouldn't be able to buy a game for a console generation, that isn't produced any more. Nintendos Virtual Console is fine and I find some interesting old games on XBox Live too, but what, if I want to play a game, that isn't available on these platforms?
Without used game sales, video games wouldn't have a memory at all, we would always live in the digital present hell, the industry calls Current Generation.
Just like second hand book stores or used record shops, I can pick up games, that are long out of production via used games sales.
I don't think what they consider acceptable responsiveness and visual fidelity loss is going to be acceptable to most people playing the games.
I think other means for playing games online will win over... I see good potential in Unity3D and of course Flash etc.
But let's only take the 500€ for two years. That makes around 20€ per month. Sounds like a good fee per month to me on that perspective.
Let's calc more. What about the prices for up to date games. Lets say 2 games a 60€ a year, makes 10€ per month.
Now taking this together we have 30€ fee that sounds reasonable, not to mention the saved electricity bill.
In addition you save a noisy big pc. When you use the small box for playing they offer, you have even a device that looks good in the living room.
More to think about: someone who lives in a rural area can still buy it's own high end pc and games in a boxed version to play it with a low latency, the same is true for high end gamers.
The only point that is nagging on me is online gaming, since this will add even more latency to the one you already have because of streaming. So for a good playing experience in CS you have to either play only with people who are using OnLive too (since they are literally in the same data center), or you have to buy your own pc.
On the other hand, when you play Games like WoW, does the lag really matter that much?
I bet my 50 pence on OnLives success!
I don't think OnLive will offer a flatrate service, I expect, that you have to pay for every game you play. So your calculation is missing the price of the OnLive games.
Plus, it isn't necessary to buy a new PC every 2 years for 500. In the last years, the PC hardware got faster and faster, as usual, but there were no games, to take advantage of this. Still Crysis is the hardware reference game and no other title demanding more hardware is in sight this year.
You write:
"someone who lives in a rural area can still buy it's own high end pc and games in a boxed version to play it with a low latency, the same is true for high end gamers. "
But why should this person use OnLive at first?
And the electricity bill isn' as high, as you may think, here in Berlin, a PC, that consumes 400W costs around 5 Eurocent per hour. If you play 365 days a year for 4 hours a day, it costs you 73 per year, or 6 per month. And this calculation is much to high, cause I don't know of any PC game, that let the PC consume 400W every minute, so the costs would be more around 50 a year. But that isn't the point. These costs exist on OnLive's side as well and they have to pay them as well, so you can be sure, energy costs will also affect OnLive's price.
With the e.bill I may have been a little bit to quick, but a 500€ pc every 2 years for a gamer who wants to play up to date games in a good quality (not high end) I think is not unrealistic and what a lot of kids are willing to pay. Of course this depends on what kind of games you are interested in, but the best sellers (shooters) are demanding.
This means, a high end PC of 2007 will be enough to play games at least throughout 2010 in high quality and, as it looks, throughout 2011 at least in good quality.
My PC has cost over 2000€ in the year 2004 and could be considered as high end, most of the up to date titles I can play only with low resolution and with low details, nothing to enjoy on my 21" display.
Some other factors that came to my mind. It can happen that you have a hardware fault, then you have additional costs.
Another thing is that there are still a lot of compatibility issues between games and hardware components. With this kind of problems you don't have to fight when you use OnLive. Some easy clicks and the game runs smoothly. Nothing to worry about!
That may be even the most striking factor for a lot of people who don't want to bother them self with administrative tasks but simply want to play. You don't even need administrator privileges, so you can play at work every game!
Today, you can play every new game on a 500 PC in High Details and you will be able to play every game in the coming years on this PC, cause the only hardware hungry games, that are published on this platform are XBox360 ports and even those ports getting fewer and fewer.
Apart from that only some east european and russian teams are working on original PC IPs, but these games don't have enough mass appeal to appear on a platform like OnLive. I am pretty sure, no Stalker or Metro2033 will be in the games list of OnLive.
I really don't see, how OnLive can be profitbale by supporting a completely doomed games platform, that only exists cause of 2 games, The Sims and World of Warcraft, titles, that aren't interesting for OnLive. Apart from this, I am still very sceptic about the technology.
Personally, I would use it when it comes out for a fair price (whereas of course fair is a very stretchable term).
Thank you for the nice talk! :)
http://www.eurogamer.net/articles/gaikai-cloud-computing-gameplay-that-works-blo
g-entry
Last post for this thread. Good day
I got the luck, although being in Germany, to try it out and I must say even though the data center is in US I was even able to play the shooter games. Yes there was some major latency in this case, but it was impressive how playable it was even though (and I used wlan). Strategy games where playable absolutely fine!
I would have wished for a little bit better video quality, but I guess this will improve as soon as the servers have reached good ol' Europe.
Guys I'm really impressed by it and this makes me even more confident that my forecast was right! *yes!*
Most impressive is that they announced that no monthly fee will be needed and a flat rate for a big collection of games only costs 10 bucks! (hm, maybe I should apply as a consultant for onlive...) ;)
Cheers!
Peter