While OnLive and Otoy are the tip-of-the tongue names when it comes to server-side rendering approaches to delivering games, Gaikai promises something else: a business model in which affiliate Web sites can host games (or virtual worlds) and earn money by doing so. Develop a Steampunk game, embed it in your Steampunk site, and make money from the users who access it.

Now, I’ve been thinking about this because I’ve been thinking about Second Life in the browser, about whether Linden Lab is truly for sale, and all the rest of the speculation from the past few weeks. I set myself up with some Google alerts for things like ‘streaming games’ or ‘games + cloud’ just to see what I’d see.

And in an article earlier today from the Game Developer’s Conference alerted me to Gaikai, a server-side/cloud-based way to deliver games without a download, and their imminent launch:

OnLive’s been going through some growing pains, but awkward teenager or not, it’s still available for all to try. Its main competitor Gaikai, however, has definitely subscribed to the mantra of “slow and steady wins the race.” When, though, will it finally cross the finish line? Speaking with VG247 during an interview at GDC Online in Austin, David Perry said that the cloud gaming service is on track to go live in mid-December.

“We said we’d be done by the end of September, and we are. We’re feature complete. You see it running from Dallas. That’s the experience that people are gonna have. So the problem is we have not had the mass market real gamers come and play this. We’ve had publishers playing, but we haven’t had real gamers,” Perry explained.

“The minute that announcement comes out of who we’re gonna partner with, we’ll start sending out invites immediately. And we’re gonna do that for 60 days. So we are 60 days from the start of those invites to launch,” he added.“So that’s means, at some point in December – probably mid-December – we will be live. There will be no ‘you’re in a beta.’ It’ll just be ‘go ahead and play.’”

Server-Side Rendering and Second Life
Now, I won’t get into the technical background about server-side rendering of games. The truth is, I know nothing about how it works technically, and by some accounts it doesn’t always work as advertised.

But the idea is simple: when you download a game you do so because you need to ‘render’ the game on your computer. How else could you wander around a 3D game environment if it didn’t ‘build’ that environment as you interact with it? With PC processing power, graphic and video cards making significant advances over the years, your home PC can render out rich and immersive environments, but at a cost: you still need to install what is often a massive multi-GB program, because you need all of the game assets (3D environments) on your machine to render them.

Even MMOs like World of Warcraft install the content on your machine. When you access the server, what you’re accessing is the position data of other users and ‘asset’ data (like where a monster is or where a herb is that’s ready to farm). But the visuals are still rendered from your machine.

The breakthrough with Second Life is also, to a degree, its challenge: Second Life is a virtual world that doesn’t require you to install the ‘assets’ (buildings, avatars, clothes, etc) and because of this, it can support dynamic building within the world. A chair is created by another user, and you see the chair at the moment of creation.

But by not downloading the assets to your machine, it downloads them dynamically. What this means visually to a user is that the objects slowly “rez” – you might stand in a store for 5 minutes in the virtual world while the signs slowly come into view. The speed at which they do so is based on a bunch of factors, among them the processing power of your machine and the bandwidth of your Internet connection.

Second Life thus faces two challenges: one, similar to games, is that not everyone will download something before trying it, and in Second Life’s case you need to download a ‘viewer’, which is the software that allows it to ‘render’ the world which is delivered dynamically (as you experience it); and the second is the ‘lag time’ by which the assets arrive into view.

Games don’t face the latter problem, because the assets are stored on your hard drive, but they do face the former problem, especially since the asset files that are stored on your hard drive count in the Gigabytes (World of Warcraft, for example, can take hours to download – and that’s just the patches which are additional to the DVDs you buy in the store).

And then comes server-side rendering. And I think of it like this:

- All of the assets are stored on a cloud-based server, so you don’t need to download any assets (Second Life already stores assets in a ‘cloud’)
- And the ‘rendering’ of those assets is handled by the cloud as well, so your computer doesn’t need to do any heavy-lifting when it comes to rendering the content.

Otoy
My interest in cloud-based rendering started with a video by Otoy that showed up on youTube a few years ago and which generated a storm of discussion and controversy:

And the controversy seemed warranted, to a degree. Otoy had no Web site at the time, didn’t respond to queries, and the whole thing came across, frankly, as vaporware.

More recently, they showed another demo video, this time running Crysis on an iPad:

The Otoy Web site claims that it’s coming, um, this summer. So I guess they’re running a little late. But that was no reason to stop Blue Mars from selecting Otoy and deciding to provide access to its virtual worlds through the ‘cloud’. While details about Otoy remain, it seems to me, sketchy, they clearly have something and with the Blue Mars deal that something must be reasonably imminent.

OnLive
OnLive, on the other hand, is live – but it’s also only available in the U.S. OnLive is a “big game” portal allowing you to play titles like Borderlands or Just Cause without a console, without a download, and on both networks, WiFi, PCs, Macs and other devices.

At launch, OnLive charged a base monthly fee for access, which they have since waived. A video from the summer shows their version of a Crysis game on an iPad:

Gaikai
And then there’s Gaikai. And I have to admit that I’ve fallen down the rabbit hole a bit with this one as you’ll see.

An article from last summer gives some insight into how the service will work, although the technical details are, well, kind of vague. But what I find interesting is the way that Gaikai is positioned for game publishers (emphasis added):

Gaikai isn’t sold to you; it’s sold to the Activisions and EAs of the world. But what are they buying?

“So, selling to publishers is to sell new gamers,” Perry tells Eurogamer after his Develop Conference speech last month. Perry believes that the number of gamers clicking to play using Gaikai and then actually following through on that and playing will be significantly higher than on existing game portals and download services, which often lose potential customers to technical problems or boredom – games not launching, games taking too long to launch, and so on. “If I told [the publisher] for every gamer, it’s currently costing you $5, and I can get them for $1, and it’s like a fraction of what you’re currently paying, then it becomes commercially viable and makes sense to a publisher, and I don’t have to explain any more.”

So how does it work from a publisher’s perspective? “They put the game on the service. They come to a dashboard, they create an account, and they add their game to the service. So if you wanted to publish a game tomorrow, you could do that. You come on, you register, you add the game to the service – it will have to be tested of course to make sure it’s not breaking any rules – but as long as it’s not breaking any rules you can drive people to your games.” The way publishers use it is analogous to iTunes, he says. “Apple doesn’t go, ‘Please can we have your games?’ They just make the service and if you use it, great, if you don’t, that’s fine.” Just remember: the way you use it will be very different.

That’s also the reason Perry is constantly correcting people who liken Gaikai to OnLive. They have some common technological genes – or at least philosophies – but their commodities are considerably different. OnLive is closer to something like Steam, with a different delivery mechanism and attendant cost concerns, whereas Gaikai is more like an embeddable YouTube window showing licensed content with a range of possible access levels determined by the licensor. In other words, publishers use Gaikai to expose you to their wares, but how they do that is up to them. Nintendo might embed a trial version of Mario Kart on Eurogamer as part of a marketing campaign, for example, or Blizzard might let you play full-blown World of Warcraft on its website. And by selling volumes of gamers to publishers, Gaikai is in a strong position to react to demand rather than having to anticipate, say, one million people turning up at midnight to play Grand Theft Auto V.

And frankly, with language like that, it may sound attractive to game developers, but it probably sounds even more attractive to virtual worlds (even World of Warcraft, which recently broke 12 million users, must face the challenge of convincing people to install a massive program on their machines or to wait for a 5-hour download).

But imagine Gaikai for virtual worlds like Second Life or, even, OpenSim. Theoretically, a virtual environment like Reaction Grid or Heritage Key could ‘subscribe’ to the Gaikai service.

Consider this quote from their Web site:

As a game publisher or developer, what’s the best weapon in your online marketing arsenal? Banner ads? Nope. Game trailers? Sorry. Gameplay videos? Guess again.

The answer is: your game. The best way to sell your game is let players play your game, as simply and conveniently as possible. But how?

“Gaikai brings new opportunities and capabilities that will improve both our craft and products, including secure beta-testing and the ability to instantaneously bring the latest games into the hands of our waiting audience.”

That’s where Gaikai comes in.

Gaikai’s streaming video game technology allows gamers to try your games online, instantly, with nothing more than a standard web browser running Flash or Java. There are no forms to fill out. There’s nothing to download or install. There’s no special hardware to buy. It’s just your game, available to try, with a single click.

But equally intriguing is their notion of an affiliate program:

Imagine being able to let your visitors try premium 3D games — some of which haven’t even been released yet — right on your site. Think of the buzz you’ll generate from players who are browsing your site and are suddenly given the chance to play the latest and greatest game, right there on the spot. Consider how many more pageviews you’ll accrue as players click around your site, hoping to be offered a hot new demo.

These are the benefits Gaikai can offer your site. Oh, and another thing: we pay you for it.

This isn’t just another affiliate program that pays you a tiny percentage while the other guy rakes in money hand over fist. As a Gaikai affiliate, you’ll make exactly as much as we do on most games. If we’re making a penny a minute for demos running on your site, so are you: we split the revenue right down the middle, 50/50. The difference is, we invented the technology, bought the servers, paid for the bandwidth, and hired a staff … all you need to do is add a single line of code to your site.
Gaikai’s game demos run directly on your site. There’s nothing to download and nothing to install — it runs right in a web browser, using Flash or Java. Just a single click, and users can immediately start playing.

Because we want players to have the best possible gaming experience on your site, our code tests the user’s connection invisibly in the background. If we can’t guarantee that your user will be satisfied with the experience, we won’t offer the demo — nobody will even know it’s there. But if our code tells us that the user has what it takes to run a great demo, then we display an offer to play, and we’re on our way.

When Philip recently removed orientation from Second Life so that new users would be brought directly to places within the virtual world, you couldn’t be doing a much better job of setting up an affiliate network. Say you’re a not-for-profit holding a weekly seminar or fund-raising event, or you have a concert venue in Second Life – now, the owner of that sim could theoretically install a single line of code on a Web site and through something like Gaikai earn a new stream of revenue.

I can’t help thinking of the number of people who watch Metanomics from the Web who would now be able to attend with “one click access”.

Here’s a recent news clip on Gaikai:

Gaikai Second Life?
Linden Lab should consider Gaikai as a way to provide the virtual world experience to new users and to avoid major barriers to entry and retention: the download of the viewer; the ‘lag’ from streaming content; and the need for a minimum spec machine to properly access and ‘see’ the world.

This approach would be a boon for both consumer use and other markets. Enterprise and education users often have remarkably low spec machines and you run into all sorts of issues related to firewalls, proxies and software updates. With Gaikai, you’d merely need to make sure that the site wasn’t on the ‘blacklist’ of sites which some organizations maintain but not have to worry about convincing the IT guys to open up some new port.

By also subscribing to the type of affiliate service that Gaikai seems to offer, the Lab would be generating a completely new layer to the virtual economy. I can’t help sort of brainstorming a little and thinking what something like Gaikai (or Otoy, I suppose, although it’s a lot more obscure) could lead to:

- Web sites on everything from music to Steampunk having ‘embedded’ Second Life streams – especially if it could be handled in a way where a user could land directly at a SLURL chosen by the site owner (Metanomics could embed a Gaikai stream and land users right at center stage)
- The Second Life Marketplace having embedded streams so that users could browse the virtual goods and then, with a single click, visit the merchant’s store
- Being able to monetize events like concerts or dances and not have to rely on tips anymore
- Landowners being able to offer companion Web pages to their tenants – “buy land, get a free Web page, and embed your home location so your friends can come visit”

Users could then make their own decisions about the types of content they place ‘around’ the embedded world. Want links out to your Facebook profile? Go ahead, because those things would be ‘surrounding’ the entry point to the world rather than embedded in it.

Of course, these ideas open up as many questions as answers, but since this is all totally speculative anyways, I’ll leave my ranting posts about identity or choosing your avatar for the day it happens. For example, all kinds of questions about ‘casual’ users come up with the potential for griefing or a lack of deeper engagement because they can ‘surf’ the world – but maybe in the end its better to have a million people surfing the Grid than 60,000 immersed in it – because didn’t we all ‘surf’ to a degree when we first arrived?

Besides which, with the Lab supposedly on the ropes (at least according to the thin slice of the user population who’s on Twitter) it’s unlikely that this is in the cards for the near future. Although it’s intriguing that Gaikai will be announcing its partners around the same time as mesh goes into public beta in Second Life, but that’s probably one tenuous connection too many.

Having said that, I’ll leave you with an observation: one of the major investors in Linden Lab is Benchmark Capital, who are also investors in Gaikai. And a former board member of Linden Lab, Mitch Lasky, is a board member at Gaikai.

Regardless of what Linden Lab does, the game is changing for virtual worlds, and whether it’s OpenSim or some other platform, whether it’s Indie game developers or one of the big players – with the download removed, with the computing processing requirements removed – maybe the metaverse isn’t so far off as we might dream, and it will look a lot less like Farmville and a lot more like Second Life, coming soon to a Web site near you.