OnLive is the darling of this year’s Game Developer’s Conference. It’s making the rounds and has more or less overshadowed all of the other GDC news this year. Wednesday’s Penny Arcade talked about it. Sort of.
The idea of the service is this: It lets you play console games or PC games without needing to own the console or required PC in question. It runs in a browser window or on a special micro-console, but the actual code is run on the server. You inputs are sent to the server, and the rendering is streamed to you realtime. In effect, the console is thousands of miles away in some warehouse that probably looks like a Borg Cube that just assimilated EB Games. Your game runs on that console, and you’ve got a really long extension cord on your controller and monitor that takes them all the way to your living room. An extension cord which passes through the internet.
If you’ve got a PC or a Mac, OnLive runs in a browser window. As long as your machine is capable of playing streaming movies, it’s capable of playing Crysis through OnLive. You pay a monthly fee for the service, and then you may buy or rent games for your theoretical omni-console. That is, you buy the rights to play those games. You don’t actually buy a disk. I mean, what would you do with it?
Everyone is excited about it. It apparently works well enough to have captured the interest of folks at GDC this year. But as the unofficial Community Pessimist I feel duty-bound to point out my concerns and why I don’t think it will work. This is not because I don’t want it to work. I dream of the day where we can set aside these idiotic console wars and meaningless hardware balkanization and simply focus on the games. I will happily feast on these words if it all works out and leads us to that bright and pure future of platform-agnostic gaming.
The big reason I don’t think it will work is that we’ve seen this technology before, decades ago. We called them dummy terminals. Big Iron died for a reason, and I don’t think the expense of modern consoles is enough to reverse the trend that has taken us to ever more distributed computing. I don’t think the potential market for OnLive is as big as it might seem at first. A customer of OnLive must:
- Want to play these games, yet not own the console in question.
- Have access to a high bandwidth connection. Streaming video that’s sharp enough for you to read what’s going on is going to have to be at least equivalent to high-definition YouTube movies. How many people can watch high-def YouTube without buffering? I don’t know, but I’ll bet it’s easy for someone living in Silicon Valley to greatly overestimate the size of this group.
- Have a low-latency connection. This means that some types of DSL and satellite subscribers will be left out even though they have the bandwidth to handle it.
- Be open to the unconventional marketplace that OnLive presents, where you buy the “right” to play a game without getting any physical media. People tend to be cool to these sorts of transactions, and it’s not because of price.
- Be willing to forego the benefits of owning the console directly.
Are there enough people who meet these criteria to support the service? Keep in mind, this service is not like World of Warcraft where they can fire up a single server that can support thousands of people. If you’re going to play Crysis, then they need to have the equivalent of a Crysis-ready PC on their end. One for every player currently playing. They need multiple console farms sprinkled around the country to avoid the game-killing lag. They’ll also need enough of each particular machine so that there are enough Xbox seats to go around when Gears of War 3 comes out and half the user base all sits down on a Saturday night to play it at the same time.
Let me address some of these issues in more detail:
1. Potential Market
What happens if I own an Xbox 360 but not a PS3? I have a pile of Xbox 360 games already. Will I be able to play them over OnLive? Or will I need to buy them again? If the latter, then the service is only useful to me as a means of accessing the PS3 library.
The service doesn’t really eliminate the platform divisions we’ve come to hate so much as trade them for different sort of divisions. There will games you own in the sense of having a disc you can take to a friend’s house, and games you own in the sense that you can play them anywhere you have a good connection with nothing more than your OnLive login.
The usability of the service will be directly related to how far you are from the nearest Borg Nexus and how many network hops it takes to get there. This is to say: It will be very variable.
Note that the latency we’re talking about here is the very worst sort. A game like Unreal Tournament or Quake deals with lag with client-side prediction. If you move or shoot, your local copy of the game shows that happening instantly, and then sends that information off to the server where it is processed. The results of your actions are subjected to a delay, but you can at least control your avatar without experiencing temporal anomalies. But client-side prediction isn’t possible on a dummy terminal. If you’re lagging, you will feel it in the controls. If you’ve ever had a telephone connection that echoes your words back at you a second after you say them, then you know how impossible it is to talk with your feedback being subjected to that sort of delay. Input / visual lag is the same way, and it can make walking down a corridor a challenge, much less trying to engage in combat.
You press forward, and nothing happens. Surprised, you let go of the key and begin moving forward. This disconnect makes you want to step back to where you were, so you hit the back button and stop. You let go of the button and find yourself backing up.
But even if your latency is tolerable enough that it doesn’t interfere with gameplay – say 100ms or so – it’s still a constant overhead. Assuming the service supports multiplayer gaming (and I’m not clear how that would work since the console isn’t bound to an Xbox live or PSN account) then your OnLive latency will be added to any additional latency you experience within the game itself.
3. Community and Account Features
It’s not so true for the PC or the PS3, but the community features of the Xbox are fairly rich, and I’m pretty sure you won’t have access to them through OnLive. The descriptions I’ve read suggest that you never see the out-of-game interface. No gamertag. No avatars. No achievements. No trophies. No friends list. No making saves and taking them to a friend’s house.
Perversely, you’re going to need a high-speed, low-latency connection to play a strictly single-player game.
4. Buying & Renting
I’m very skeptical of how many people will want to buy games for OnLive. If I buy Little Big Planet for my OnLive account, and then I later buy my own PS3 because I want access to trophies and friend’s lists and such, will I have to buy an additional physical copy of the game? I think the market will reject buying games for OnLive for the same reasons it rejected DIVX players. When people buy things, they expect to own them. I can’t imagine buying a copy of a game which is locked to a subscription-based service. That’s like buying carpeting for the apartment you’re renting.
(At the same time, renting could be very attractive. It’s nice to be able to rent a game and not have to return it later.)
5. Publishers, Microsoft, and Sony
I can’t help but wonder how much they will oppose this system. OnLive is letting people play on a console without owning it and play games without buying them. They have already expressed frustration with the second-hand and rental market, and the more successful OnLive is the worse it will be for them.
Something which obliterates the walls of exclusive content they’ve built around their platforms is not going to sit well with Microsoft and Sony. I’m sure Microsoft would much rather you buy a console, pay for Xbox Live, and then buy your games. OnLive users will be parasitical to the existing revenue model they have set up, and it’s within their power to make life very difficult for OnLive.
Again, I’d be happy to see OnLive to take off. As someone who has been evangelizing videogames for years (my column at the Escapist later today is on that very subject) I’d love to see holdouts join us in the hobby. But I’m skeptical of the plan as presented.
Still, once it goes live I’ll be in line to check it out just like everyone else.
Quakecon Keynote 2013 Annotated
An interesting but technically dense talk about gaming technology. I translate it for the non-coders.
A stream-of-gameplay review of Dead Island. This game is a cavalcade of bugs and bad design choices.
Batman v. Superman Wasn't All Bad
It's not a good movie, but it was made with good intentions and if you look closely you can find a few interesting ideas.
id Software Coding Style
When the source code for Doom 3 was released, we got a look at some of the style conventions used by the developers. Here I analyze this style and explain what it all means.
Why Batman Can't Kill
His problem isn't that he's dumb, the problem is that he bends the world he inhabits.