OnLive rebooted: What went wrong?

By Evan Narcisse on August 21, 2012, 6:30 PM

I remember the first time I saw OnLive demoed for me. Three years ago, I was ushered into a conference room on Manhattan's Midtown East neighborhood and saw Crysis running of the cloud gaming service's network. It looked impressive, sure, but there could have been all sorts of tomfoolery going on to make the streaming look that good. But when I demoed it at my own desk some months later, I had to admit that the experience was better than expected. Damn if the thing didn't work pretty well. Damn if they didn't invent something that really didn't exist before.

Then it all went pear-shaped.

The saga of OnLive took surprising turns over the weekend as the company went from suddenly not existing anymore to a hollowed-out version announcing that it was still operating. While you can't have a funeral for OnLive since the service is continuing, it's still arguably the end of an era. Some small part of me always rooted for OnLive, despite experiencing firsthand the challenges that the cloud gaming service faced. The key component of OnLive—streaming games hosted on servers elsewhere—worked but it was met by obstacles on nearly every side.

When I reviewed the company's micro-console more than a year ago, I was locked out because my cable bandwidth was so terrible. This was despite the fact that it exceeded the minimum standards set by OnLive. Those hiccups got ironed out but the microconsole was still limited to a hardwired Ethernet connection. The lack of WiFi in a gaming device felt annoyingly backward. A year later, that hasn't changed.

Nevertheless, once you got onto OnLive, the catalog of games was dwarfed by what you could access in a game store or, worse, Steam. The offerings did get better with time and it was more common to see big releases like NBA 2K12, Deus Ex: Human Revolution and Darksiders II launch on OnLive on their dates of release. But it still seemed like a perpetual game of catch-up.

Another problem with OnLive was the way that they tried to position themselves. No exec or PR person I ever spoke to at the old OnLive ever fooled themselves that they'd replace the Xbox 360, PS3 or Wii. They saw themselves as additive, a place where you could sample games before you bought them or rent titles without having to wait on the mail.

OnLive always seemed like an alternative and not like a main destination, which was sort of refreshingly blunt, but it also made them easy to ignore or forget. The smart-TV partnerships that OnLive struck up with Vizio, LG and other manufacturers gave OnLive another entry point into the living room, but those deals would also be vulnerable to the quality of broadband. And in each of those deals, OnLive was just another bullet point in a long list of similar features. Hard to muster a reaction stronger than, "Oh yeah, OnLive…" when it's presented like that.

There was slightly more wow associated with the company's moves on tablet platforms, though. When OnLive's controller-and-app combo launched last year, it looked like a great way to get the kinds of deep, rich AAA game experiences that were sorely lacking on bigger-screened handhelds. I tried it on iPad and really liked it.

But while an Android launch spread out amongst that operating system's various devices, an iOS offering never materialized. OnLive never commented on the delay and what might be behind it but I think it's reasonable to speculate that they ran afoul of Apple's restrictions on in-app purchases. Apple wants all in-app purchases to run through their App Store infrastructure and the ultimate stumbling point may have been a failure to get OnLive's in-client economy to fall along those strict Apple guidelines. Whatever the reason, OnLive found themselves with no way into the ecosystem of Apple's incredibly popular iDevices. Not a great place to be.

A well-received foray in cloud-based productivity called OnLive Desktop followed, using OnLive's streaming tech to host access to Microsoft Word, PowerPoint and other programs. But games were the company's main focus. And the recent announcement that OnLive would be appearing on the Ouya console seemed like another chance for them to grab the precious eyeballs they needed to stay viable.

But then the crumbling and rebirth happened.

OnLive's latest mutation isn't the end of the affair by any means. What Lauder Partners has invested in is a company that reportedly still owns the servers, patents and partnerships that the old OnLive did, with a fraction of the operating costs. That makes them even more attractive to potential buyers than they were before. Picking up OnLive would be a no-brainer for Microsoft, especially if they want to be on equal footing after Sony's purchase of rival cloud-gaming outfit Gaikai.

But a juicier fantasy would be an acquisition by Google. The search giant's server infrastructure would probably help mitigate OnLive's network problems and Google would get a distribution pipeline for Google Play content that's already optimized for gaming. The two companies already have a pact in place and OnLive might even thrive as part of the Google Fiber initiative. Valve's already working on its Big Picture offering, so it seems unlikely that they'd want to acquire OnLive to integrate it into something that's already well underway.

Someone somewhere might wind up buying OnLive in its newer, leaner form. But if the service is ever going to reach its full potential, it's going to have to be a buyer that can address the problems—building a catalog, optimizing bandwidth, establishing a real foothold in living rooms and mobile platforms—that plagued OnLive's previous incarnation. OnLive still probably points the way to video games' future, but it's going to an extremely bumpy ride for the company to get there at all.

Republished with permission. Evan Narcisse is a contributing editor at Kotaku.




User Comments: 14

Got something to say? Post a comment
tonylukac said:

Who will maintain such servers? You essentially need a $500 video card for each user and how much memory can servers handle? The same is true of the new windows server that acts like a mainframe. Would you need 4 terabytes of memory for 1000 users? No such servers are built and even if split up think of the expense.

TekGun TekGun said:

Totally agree with tony, on top of all that there's the bandwidth. This idea was/is never going to be worth it.

wastedkill said:

What games are you playing I would love to know as my 460GTX is worth what $100 and it plays every single game out at max except BF but it plays it 1 below ultra! and 4TB of RAM for 1000 users... you crazy bro? what games require 50GB of ram haha.

The servers required for onlive wouldn't need to be as high cost as $100 billion dollars like you think tony actually far from it I think they would only need $50million dollars in servers for around 1 million users, the main cost of all this would be electric and bandwidth.

ikesmasher said:

you are annoyed but the lack of wifi, but do you realize the ping that it would cause? the ping required by onlive practically requires a wired connection.

Everything else I agree with though.

TomSEA TomSEA, TechSpot Chancellor, said:

I remember when OnLive was first announced and I published my "I dunno about this" comments. Too many hardware issues and publisher contracts to overcome. On the surface a great idea, but just too much, too fast. Not surprised at this at all....

Timonius Timonius said:

Yeah it still reminds me of Sega Channel that our cable company provided way back. Only I believe that actually worked for it's time. OnLive still has a long way to go to catch up with this 'nothing new' concept.

Guest said:

The premise of streaming games is still perfectly sound. All the naysayers will use this as "proof" of the model being flawed, but the problem is really just that disruptive technologies take a long time and lots of investment before they start turning a profit, that puts a lot of strain on a small startup like OnLive was.

Guest said:

This streaming crap will be the DEATH of gaming as we know it! I wanna own what I pay for! I don't want to rent and subscribe to every little thing in life. I'm sick of the money-milking! Don't fool yourselves for a second, guys. If game-developers could get away with it, games on future consoles, computers and whatever else would become a subscription service. After all, it's the only way to really stop piracy, isn't it? So you better believe it, this is what they want the future of gaming to be like. And it's not just games! Soon enough you'll have to subscribe for your O/S, and everything else.

Sure, I play MMOs, but I'm sick of being at the mercy of some EULA saying I can't even talk the way I talk "in public" because it's "bad language". Who the hell is Blizzard or Bioware to tell me how to talk and how to act? We have free speech in my country! That's the law! And then some unimportant little rabble of nerds decide "To hell with your laws, we have our own laws and you will follow them or else!". Imagine a world where all your savegames, friendslists and everything associated with your games is at the mercy of the sensibilities of some pimple-faced squint and his fat uncle! I don't like it!

rgreen said:

What a stupid comment, to say cloud gaming will never be worth it when in fact one day it will be the only way to play.

rgreen said:

you are annoyed but the lack of wifi, but do you realize the ping that it would cause? the ping required by onlive practically requires a wired connection.

Everything else I agree with though.

I get perfectly good results on wifi.

ikesmasher said:

I did too, but I had noticable delays occasionally because of ping.

Guest said:

wastedkill wrote: ""and 4TB of RAM for 1000 users... you crazy bro? what games require 50GB of ram haha."

Math has never been your strongest point, now has it son?

Gumpngreen Gumpngreen said:

I think cloud gaming will only come into its own when paired with a MMO which is designed from the ground up to not incorporate the client-server model. What do I mean? Dramatic 10,000+ strong war engagements with real-time physics-based combat that's not based upon the client-server limitations. Heck, you could have the landscape being blasted into bits with debris flying everywhere.

Guest said:

4TB or ram would be about 50 grand. So wheres the issue here when were talking about a multi million dollar data center?

Load all comments...

Add New Comment

TechSpot Members
Login or sign up for free,
it takes about 30 seconds.
You may also...
Get complete access to the TechSpot community. Join thousands of technology enthusiasts that contribute and share knowledge in our forum. Get a private inbox, upload your own photo gallery and more.