Steve Perlman and Model in fron fo Mova Contour reality capture system

Mova CONTOUR: The Hidden Potential Of OnLive


When I finally made the decision to take the plunge back into PC gaming late last year, I took some time to actively consider OnLive. The concept was appealing at first blush. No need to worry about a new graphics card, or even a new monitor. Just plug any PC (or netbook, or Mac) into your HD TV, sit back, and play PC games. All the hard work is done in their servers remotely, and all I'd need is the right size internet connection and, whiz bang, I'm back in the saddle again. Hands down, this would have been the least expensive option for me, but I just couldn’t do it.

The problems for me were two-fold. First, there was a critical mass of users on my primary platform, the Xbox 360. I knew damn well there was a group of users just as large, if not larger, on the PC. These were my friends and GWJrs alike, and to enter the OnLive space would have been to leave them behind. Part of the joy I have of gaming is multi-player, and I was not confident that I would have the same experience inside such a small community as OnLive.

Second, there was the catalog. While the 360 has many exclusives, there were many games I couldn't play because they were on the PS3 or the PC. To take the leap into OnLive would have given me access to dozens of PC-exclusive titles, but as far as limiting my library it would only have exacerbated the situation.

And so I made the decision to build my own PC, because the best gaming experience lay there and there alone. But I can look into the future and see a time when OnLive and its partners can contribute to gaming in real and surprising ways.

You've made it this far into my diatribe, so a little full disclosure. Once upon a time ago I did business with OnLive. Nothing I say here will have anything to do with the NDAs I may or may not have signed, and know that what I say here is entirely my own opinion and not that of my former employer. In fact, every bit of background knowledge I reference here can be found in OnLive’s press conference from GDC in 2009.

One thing that OnLive has going for it right now is momentum. Even before their service went live, they had most of the major publishers, from EA to Take Two, on board. Why? Two reasons. The chief reason a publisher would want into OnLive is the elimination of the piracy tax. If your game never leaves a proprietary data center, if the only element of gameplay forced down to users is a video feed, then you can’t pirate the game. I can also envision a scenario where porting the game code into the OnLive cloud is cheap, if not entirely free, in terms of development costs.

Another sign of momentum is the increased size of their catalog. OnLive now offers 100 games for sale or rent on their cloud system, and regularly has tournaments, contests, and Steam-like sales. If this isn’t actual momentum, at least it’s a good approximation of momentum. It appears just as attractive to some gamers as Steam or a new PC, and the multi-platform availability (TV, PC, Mac, netbook, and eventually phone) has a target market out there in the wild.

So what is this momentum building towards? Well, making OnLive the fourth console. 360, PS3, Wii, and OnLive. What I’m saying is that OnLive intends to leave traditional PC gaming in the dust, to relegate it to obscurity. And here’s how they might be able to do it.

If you dig a bit into the structure of the OnLive family of companies you see that they are a subsidiary of a company called Rearden Labs. Rearden was founded, and is currently owned by, one Steve Perlman. He’s not some gray-haired Silicon Valley investor, he’s an old-school computer engineer of the first order. His CV goes all the way back to Atari, and he happened to create WebTV (the design team behind which ultimately birthed the Xbox360), had a hand in developing QuickTime for the Macintosh as well as modems for Sega and Nintendo. He has all the knowledge and experience to have real, hands-on input into most of the aspects of the technology that makes OnLive work. This makes him fairly unique among platform owners.

Another Rearden subsidiary is Mova. This is the same company that did CGI work for The Curious Case of Benjamin Button and, to quote a recent article, the latest entries in the Pirates of the Caribbean, Harry Potter and Transformers series. Mova seems fairly secretive, but if you hit their site you can learn what you need to know about their motion capture technology.

You’re all probably familiar with motion capture. An actor, usually complaining of both heat and embarrassment, dons a black leotard studded with golf-ball sized dots, or has a series of dots plotted directly on their face for close-ups. They then proceed to pantomime a scene in front of at least three cameras. The computers triangulate the dots, and a virtual 3D skeleton appears on the computer screen. Animators then “skin” this wire frame with customized art. Mova's CONTOUR does it differently. From their whitepaper, available on their Web site:

"Rather than capturing a sparse set of dots in a scene, CONTOUR captures entire 3D surfaces. For example, with markers it might be feasible to capture a 3D 'constellation' of up to 200 points on a human face. With CONTOUR we capture over 100,000 3D points on a face with 0.1mm precision, and while we are doing it, we also capture the visual image of the face as it is lit in the scene. With so many points captured, the face no longer looks like a constellation of dots; it looks like a photoreal face, just like one you’d see captured by a POV [point of view] motion picture camera. And there is good reason for this: We are capturing surfaces in 3D volumetrically (i.e. in the round), with similar resolution as a conventional motion picture camera records scenes in 2D from a single POV. Effectively, this makes it possible to shoot in 3D without compromising the realism that we expect when we shoot in 2D.”

From clues scattered throughout this paper it seems to have been written in 2006. I can only imagine that their technology has improved somewhat since then.

Reading into recent press releases, interviews with industry publications, and yes even their GDC 2009 unveiling press conference, you can begin to see that Mova’s plan is to remove some or all of the need for traditional digital-art workflows. Imagine an entire character design department that can be replaced by a Mova device. The personnel costs, which make up an inordinately large part of AAA game design teams, are cut dramatically. Scaled up and used to its imagined potential, what you have in a Mova device is the digitizer from Tron. Push a button, and you are in the game. Push a button and a chair, table, sword, machine gun are in the game. Modeling items, as well as people, becomes a trivial matter.

Imagine a game where modeling and animation are trivial, where photorealistic faces, deformable surfaces like cloth and skin, are taken for granted. This would free game designers to focus their attention on all the other elements of game design, like physics, artificial intelligence, or the overall size and scale of the game world itself. Imagine what a company like Bethesda could do for Elder Scrolls 9 if all they had to do to create characters and items was scan in some actors and their clothing using a Mova device.

Now, imagine the kind of computer you would need to run a game like this. If this future comes to pass, then it’s possible that the only way to do it affordably would be to run that game remotely inside the cloud, where multiple high-end devices would simultaneously pound out the polygons you’d need to be able to see and experience the game. And, using the OnLive technology, the only piece of information you would receive would be the audio and video feed.

It’s about now that you’re rolling your eyes at me. That’s OK, because I have every expectation that this will never happen. It would be a giant shift from the way that games are made and could, potentially, merely lead to the “full motion video” debacle of the nineties, where every game devolved into a grainy video stream with choose-your-own-adventure gameplay. I don’t want to see that happen again. But understand that even if OnLive fails, or is bought and reduced to an add-on for AT&T cable subscribers or hotel chains, Mova is a separate entity. It seems to be playing a much longer game than even OnLive is.

So, it’s time to stop debating if OnLive will be “the one console” or the one console in every Motel 6. The real company I’m interested in is Mova. They’re the only people I can think of that might prevent me from buying my next high-end gaming PC.


Thanks for this. I had no idea about the existemce of MOVA. I'm wondering if CONTOUR uses an IR pattern system for its plotting (a la Kinect).

I'm always wary of photorealistic technologies. The Uncanny Valley makes the particular art style problematic for me, as the tiny flaws detract from the actual design presented. I also tend to recoil when game characters are directly modeled on their VAs...

OnLive as a "hotel supply point" would actually be a damn interesting direction to go in. It would mean an investment in the IT infrastructure, but the ability to play a game you're working through while on vacation, without having to lug your game equipment? That's pretty novel.

I don't have a proper understanding of the North American gaming marketplace, and its online infrastructure, but OnLive could work a lot better in East Asia, particularly because a substantial portion of East and Southeast Asians already do their gaming and socialization in internet cafes.

That said, the big problem with OnLive is that you can't game off the grid, and the upgrade costs would be absorbed by the gaming masses eventually. In the sense that users are sharing computer units, it's a lot like a system of internet cafes, but you additionally have to distribute a lot of terminals to a lot of homes, and you have connectivity issues.

Finally, the most damning thing about OnLive is that you can't pirate the games. Companies right now think that this is a good thing, but once they can actually play on a truly piracy-free platform, I think they'll eventually realize that piracy has a positive effect on markets.

A lot of games in East Asia have had to evolve under the pressure of rampant piracy. Their online-only software infrastructure insures that they can't be pirated in any realistic way. You know how much they cost to play? Free. That's right. Pirate-proof games in a rampant piracy environment still have to ply themselves aggressively, because in a pirate-rampant market, you have to compete with the "free" mindshare of pirated games just to have a chance of getting any money from a consumer.

Mova and onlive seem like two separate issues, sharing a parent company (if my skimming is correct).

It also seems to me that Mova isn't really a magical cure-all for the high costs of high-def on modern games, it's a tool that can be used, but that tool must fit into how games are made by a particular studio. Unless you're making a fully contemporary game, you still have to create the art it's capturing and bring it into a game.

Piracy also seems like it needs another description when you're talking about markets like E Asia, it just seems like a whole different game than just people getting stuff for free. companies whining about it aren't really going to change anything, bringing something like onlive to the table is something that might work, but it still might fail if the options are pirating a game, or paying a subscription to stream the same game.

Onlive is apparently looking into implementing the free-to-play model into their infrastructure (with micro-transactions, et al).

... or paying a subscription to stream the same game

Onlive doesn't have a mandatory subscription fee anymore. You can buy a single game for a set price (just like buying retail or steam), or you can subscribe to the "Playpack" for $10/month and get full access to ~80+ games (they add 2 or 3 a week at this point).

So if/when they release free-to-play games you won't have to pay a subscription fee to play them on Onlive. And in a few years, if Onlive takes off, those free-to-play games could look pretty darn good if developers start pushing Onlive's supposed superior technology.

OnLive as a "hotel supply point" would actually be a damn interesting direction to go in.

They're starting to build Onlive into certain HDTVs, so there's a good chance that even if Onlive doesn't get bought out, you could find it in hotel tvs in the next couple of years...

LarryC wrote:

That said, the big problem with OnLive is that you can't game off the grid, and the upgrade costs would be absorbed by the gaming masses eventually. In the sense that users are sharing computer units, it's a lot like a system of internet cafes, but you additionally have to distribute a lot of terminals to a lot of homes, and you have connectivity issues.

I think that pooling together the hardware is likely to save money, just as it's more efficient to have a power plant than having individual generators at every home and building. That said, there will likely be similar peak-hour problems.

That said, there will likely be similar peak-hour problems.

Not if Pearlman has anything to do with it!! Apparently Reardon is also working on some super advanced wireless system that would make peak-hour congestion a thing of the past..

Ariskany Evan wrote:
That said, there will likely be similar peak-hour problems.

Not if Pearlman has anything to do with it!! Apparently Reardon is also working on some super advanced wireless system that would make peak-hour congestion a thing of the past..

Peak-hour problems in electricity are caused by the relatively low use during the day and late at night, but with a sharp jump in everyone's electricity use from the time they get home to the time they go to sleep. I think you had something else in mind.

Here's the thing, modelling items is already a trivial matter. That work is all outsourced to art studios overseas that pump assets out far cheaper and higher quality than this device would. This technology has its uses (it sounds similar to what was used for facial capture in L.A. Noire, and that's where this tech shines), but using it to model in-game assets is like using a state-of-the-art military helicopter to deliver pizza. Modelling assets is a fraction of what the art team on a game does, and it's the cheapest fraction.

As far as not needing a character art department; well, replacing modellers and animators with a cast of actors, a wardrobe and make-up department is not a decision that will save you money.

Mova sounds interesting.

That said, I'll never ever ever ever in my life use onlive. The consequences of putting all that control in a set of hands that can just switch your games off and leave you with nothing are just unacceptable to me. If there was ever a service that could be akin to renting games, this is it.

Steam is as high as I will allow myself to submit, and only because they compensated their DRM by giving back a lot of useful features and they have sane prices. Onlive?. Nope. And even less with the flakey quality of my bandwidth.

Speaking of bandwidth...

Maybe Steve just seems so calm about all this because he's busily solving all these problenms on his own.

"As such, it is ideally suited for Perlman's other business, the gaming service known as OnLive... OnLive and DIDO can share the same data centers... 'The unveiling od DIDO is the latest piece in the puzzle'... 'There are more coming.'"