The case against OnLive
April 08, 2009
It’s foolish to deny that the cloud will dominate PC use in the future; dumb terminals made sense when raw computing power was unnecessary, and we’ll come full-circle as soon as broadband is ubiquitous. For nearly all applications, all an end-user needs is a browser with an internet connection; there’s an easy case to be made that a majority of computer users could boot straight to a browser, relying entirely on Google applications to fulfill their computing needs. This minimization of digital clutter is intensely attractive to us, because realistically, most of our own computer use is bounded to the internet, which is a proven medium for presenting information.
And that’s why OnLive is so fascinating. It’s a gaming service that will likely leverage core web technologies to present a list of demos and full-on games for consumer consumption. And these games will not be typical Flash-based titles, but high-end games that are typically played on current-gen consoles or state-of-the-art gaming PCs. And it all works on a low-end system because the games will actually be played remotely, on an OnLive server. Rather than a monitor connected to this server, however, the visual output is streamed to the end-user over the internet through a browser plugin, so it’s like playing a console game where the console’s hardware is many miles away, but the TV is right in one’s own living room. The idea is so simple, we wonder why no one’s come up with it before.
Technically speaking, OnLive’s video streaming should be cake - high-resolution streaming isn’t new, and judging from typical HD streams available from Hulu and the major networks, there shouldn’t be any serious doubt as to OnLive’s success in this area. Unfortunately, this is only half the puzzle, because unlike typical streaming video offerings, OnLive is all about an exchange of data. That is to say, a user doesn’t just click “play” and sit back, but rather inputs data through a controller over the course of the streaming. When these inputs are received by the OnLive server, they’re processed, and the resulting video is now sent back to the user. In other words, lag is less of an issue as latency, because the video sent to the user needs to match up with what the user just pressed on his controller. The further from an OnLive server the user is, the greater concern latency becomes.
Admittedly, we don’t know much about OnLive’s server distribution yet. If OnLive takes the route of legacy dial-up companies like AOL and Prodigy, we can expect OnLive to piggyback the infrastructure at the local ISP level, and in this manner, slowly build a network of servers based on the demand for OnLive’s service. The clear advantage here is that servers will be distributed nationally, so as to alleviate latency concerns in the best way possible.
But let’s say that the latency issue is overcome with reasonable ease, and that graphics will down-sample when lag comes into play. We’re still not convinced that consumers will tolerate the hiccups that video streaming occasionally produces, but even with that sorted out, we’re now left with balancing OnLive’s subscription costs with the initial expenses of a typical console purchase. In short, end-users will have to grapple with the idea that they can shell out $300 or so dollars plus game costs, else some small amount for OnLive hardware plus a recurring monthly expense. At what point does the average user go with the monthly subscription to stay cost-effective? For a fresh XBox 360 purchase, plus a year of XBox Live!, and a game purchase once a month for the life of that Live! subscription, we’re looking at a rough cost of $950. Assuming a hardware cost of $100 and a reasonable $20/month subscription, that $950 gives us unlimited game time for about 3.5 years using OnLive. Best part is that the XBox will ultimately be replaced by the next big thing (requiring another hardware purchase), while OnLive’s hardware upgrades should be transparent to the user. Sounds great, doesn’t it?
Yeah, it sounds great initially, but that’s before the realization that a $240 annual investment by the end-user ($340 for the first year, assuming the $100 hardware cost) arguably won’t cover the hardware costs and maintenance fees for the OnLive service. An XBox 360 is designed to run one copy of a game, and while one could argue that an OnLive server’s configuration will be able to serve up quite a few more people, we’re still likely talking one dedicated graphics card per customer, and those don’t come cheap. This isn’t even taking into account the licenses per game that OnLive will be running, and with this comes another initial obstacle for OnLive: the success of a console at launch is all about the game library. If OnLive ships with a small library that doesn’t expand steadily, then OnLive won’t attract a large enough audience to build up the needed momentum.
And with game libraries, we’re talking clear limitations on what OnLive can offer on account of their setup. Many folks are enamored with the idea that OnLive will be a stand-in for PC gaming, because OnLive’s servers will be upgraded automagically over time. That’s a faux pas, however, because PC gaming’s strengths are in add-ons, conversions, and other customizations. OnLive severely limits PC gaming because all software, including any add-ons, are controlled entirely by OnLive. This means no playing MMORPGs with custom UI enhancements, limited flexibility in controller options, no beta conversions/mods, et al. That’s why OnLive, for all intents and purposes, is a more likely console replacement than a gaming PC replacement. It simply happens that OnLive works with PCs and Macs, but this is just a cheap way to get out of using a TV, not at cheaply proving a PC gaming experience.
OnLive’s gaming library pretty much acknowledges the notion that the service is aimed to compete against consoles and not PCs. Burnout, Tomb Raider, World of Goo, Crysis, LEGO Batman… we’ve seen these before. Where’s the die-hard PC games that don’t have great console ports, like RTS games, MMORPGs, etc? We’re looking at rehashed console games here, which OnLive wants to use as a springboard before developers start looking at OnLive as a serious competitor. That means that OnLive users shouldn’t expect any worthwhile exclusives for some time. And a second question begs our attention: since these titles are all available already in the console battlescape, why not just get them for the consoles we already own?
The math we mused over before changes some when we’re looking at the current state of console gaming. Assuming a gamer even buys 12 titles in a year, they spent about $600. An OnLive subscriber, on the other hand, will spend $240 that same year. But at only six titles per year, the numbers are already looking more similar, which means that casual gamers will see less benefits from OnLive, especially since their purchases are more likely to be blockbuster titles that OnLive won’t even see for at least a year. When all is said and done, OnLive looks to sport a formula that would be great for hardcore gamers, if only OnLive was mature enough to offer a larger gaming library of blockbuster titles. Asking a gamer to invest in OnLive’s offerings at release, however, is about as bad as proposing someone buy the Wii but stay away from first-party titles.
What it comes down to is this: early adapter aren’t going to get anything from OnLive that they can’t already get on current-gen consoles, but it’s these early adapters that will be necessary to grow OnLive’s offerings to a point where the average gamer will be attracted to the service. For PC gamers, OnLive is a complete bust, because the closest thing to playing PC games on the cheap that we’ll see anytime soon is in something like NVidia’s ION, not OnLive. If OnLive actually manages to survive such that it’s around when the next console refresh happens, then it may be a worthwhile investment for gamers so as to avoid expensive hardware costs.
That all said, OnLive could have a couple tricks up its sleeve, like licensing copies of popular MMORPGs, and figuring out a secure way to let customers upload, or activate via OnLive’s web interface, popular add-ons. Realistically, though, we’re not holding our breathe.