OnLive, Not Very Live: Only 1600 Concurrent Users of Cloud Service, Company Tumult Reveals

If you hope that cloud-streaming of 3D interactive content will transform MMOs and virtual worlds and make them much more accessible (and I definitely did, and do), then this is pretty bad news: Leading cloud service OnLive just went through some painful company layoffs and restructuring, and in this painful tumult, we learned just how little people are actually using OnLive:

CEO Steve Perlman told workers the company was averaging 1,600 concurrent users on a service with 8,000 servers requiring maintenance.

Emphasis is so mine. 1,600 concurrent users probably translates to about 16,000 monthly users, which is a tiny fraction of the 2 million registered users it was estimated to have. "You could design the next Second Life on this!", Perlman told me three years ago, but ironically, SL has 30,000 concurrent users at its lowest point. I haven't followed OnLive closely in the last couple years, so I'm not immediately clear why the service has failed to gain a wider audience, but my strong guess is the continued growth of Steam, in terms of users and game/feature offerings, made OnLive seem less appealing to gamers. (So good thing SL is going on Steam, rather than on OnLive.)

Does this mean cloud-based gaming is dead? Not necessarily:

Onlive competitor Gaikai, of course, was recently bought by Sony as a Playstation service, and I know of two or more other cloud-deployment services currently in development that could rise where OnLive has fallen. Trouble is, with OnLive sputtering so spectacularly, I think investors will worry about playing in this space for quite awhile.

Tweet

Hat tip: Howard Berkey.

Comments

[Joker]

just another hyped piece of tech making a very few 6 figure salaries for 5 years....

[Adeon Writer]

As a gamer, I passed on OnLive as their library of titles simply didn't interest me. Nice tech, lack of interesting games that matched my tastes. (Mostly grunge shooters from what I saw)

[Scarp Godenot]

Any cloud streaming service needs a business plan that will MAKE MONEY.

Why is this so hard for people to understand?

Here is what happens every single time. Great idea leads to application showing up and bragging about how it will take over the world.

Investment capital runs out. Company goes down.

[Ezra]

"I haven't followed OnLive closely in the last couple years, so I'm not immediately clear why the service has failed to gain a wider audience, but my strong guess is the continued growth of Steam, in terms of users and game/feature offerings, made OnLive seem less appealing to gamers."

This is exactly why it failed. You aren't alone in not having had any reason to check it out.

I loved OnLive for the very few games that it had that were somewhat close to still being any kind of popular. For me, that was Borderlands when it came out on OnLive. It was just plain awesome to be able to spectate other's progress in such a dynamic game and even be able to drop in for multiplayer.

OnLive is two major things. First, an as-good-as-desktop gameplay experience. Secondly, a great social gaming experience that allows instantly seeing what one's friends and others are playing in actual realtime.

The issue is however OnLive has a terrible selection of games. So despite making jumping in games easier than Steam, it fails in comparison to Steam because there's no Skyrim and plethora of indie games on OnLive.

Even though it has this great spectate and socialize in a dozen different friend's gameplay streams at a time it fails against disjointed recorded/live experiences like gameplay on YouTube, twitch.tv, ustream, etc. due to a lack of good games.

Content is everything. The platform and tech is brilliant. But OnLive if it's going to survive this second rebirth is going to have to find a way to get good, interesting games. It doesn't need triple As, it can take the GOG approach of systematically getting older games up to par for playing today. It can appeal to indie game developers by supplying an SDK and way to make money off of OnLive.

[Jessi]

OnLive looked like a failure from the get-go, and it appears this is becoming increasingly true.

Requires a large dataplan for broadband, and even then, it still would be laggy.

Didn't give the full HD experience on all games, even if you did pay to get the required minimum graphics card.

Either one of those is bad enough, but you also had to rent the game at anywhere from a couple bucks, all the way to the full price of the game - Though you'd never own the game...And they were actually planning on charging users a monthly subscription.

It would've been a fantastic almost-scam-like service if it worked.

[Ezra]

@Jessi

I'd suggest trying it if you were interested in whether or not the tech works. It does.

The only reason to assume lag, latency, video quality degradation or other problems common to video is if infact you already have those problems with streamed video. The client is littlemore than a glorified video player that sends mouse/keyboard/gamepad input over network. If you don't have video streaming problems with Netflix or YouTube you shouldn't with OnLive. It should be as smooth. For me, it always has been.

The bandwidth though. Yeah. For someone who games for 6 hours a day that's akin to watching Netflix for 6 hours a day. It can have a big impact on data caps.

[Archangel Mortenwold]

I suspect the same will hold true for this Valve-Steam deal as well: the promised or expected surge in users won't materialize because Second Life never was and never will be a gaming platform. It's not World of Warcraft. Minecraft, Steam, OnLive, The Sims, or any other game or gaming environment. Yes, there are dedicated roleplay regions, but they're set up more like D&D than anything else and they rely on the creative talent and imaginations of the people involved to realize their full potential. Anyone expecting to come into SL and find a gaming experience like what you find on Valve is going to be disappointed, and upset about being lied to.

As people keep saying, Hamlet: the only thing that can save SL is lowering the outrageous tier prices. Every time Linden Lab tries to market the grid as some kind of gaming platform it fails spectacularly. The tools for making it into a gaming platform simply are not there, nor is the demand for such a thing from the vast majority of users. I know you want SL to become the next WoW or whatever it is you'd like it to be, but it's not gonna happen. The price in real money is simply too high and the demand isn't there.

Now, you do make some very good points when exposing the backward thinking of Linden Lab. That's to be commended. You've pointed out how the grid has been losing regions to the ridiculous tier prices — something that's been happening for years now — and that needs to keep being said until LL pulls its collective head out of its ass and brings the prices down to something reasonable. But really, what point is there in continuing to demand and expect SL to become something few want it to be anyway and that is priced to the point of being prohibitively expensive for most users?

What Linden Lan needs to do is go back to the basics of what made Second Life enjoyable in the first place. When I joined in October of 2008, SL was still a place where educators, nonprofits, businesses, and others could network with people all across the globe. There was a lot more freedom for creativity, especially with the teen grid there to allow minors to enjoy what SL had to offer without coming across content they shouldn't. The viewer GUI was usable and easy to learn, in contrast to the clunky, inexcusably bad interface in place now, with its bugs and lag.

Three things should be done to start getting SL back in top form. It's not a complete or comprehensive list, but it's a good start:

1.) Lower land tier prices by one half to two thirds what they are now. it's ridiculous to charge someone a thousand dollars he or she doesn't have and then another $295 a month for server maintenance fees when competing grids like OpenSim and InWorldz offer far more in their regions (up to 45,000 prims, for example) at $75 a month. That's why OpenSim and Inworldz have been growing in size and membership while SL has been shrinking, which is the real reason Linden Lab blocked access through its own and approved TPVs. Lower region prices to something competitive with the other grids and at least double the prim allowance, and invite people back. The upsurge in returning region owners and new buyers will more than make up for any temporary revenue loss and spread out the revenue source over a wider user base so that future losses won't make such a huge impact.

2.) Bring back the teen grid and do away with the stupid rating system that resulted in the adult continent. If the average SL user is between the ages of twenty-five and forty, it is just plain stupid to force us to constantly be on guard against the entry of underage users into our areas for fear of being banned for violating laws. Give the kids their own grid again and let us adults use SL the way we used to.

3.) Do away with the V2-V3 GUI and bring back the V1 GUI. This does not mean getting rid of mesh or pathfinding or any of the other features Linden lab wants to implement — although those need serious retooling such as making mesh imports the same price as image uploads and doing away with extra prim cost for mesh objects. What it does mean is going back to a user interface that works and is easy for people to figure out. If TPVDs want to develop something to their liking, let them, but don't turn people off with something no one wants to use and that they get so frustrated trying to learn that they either leave SL or end up on TPVs that retain the V1 GUI. When the most-used viewer is one that has the V1 GUI and LL's default viewer is a distant third, that should be sending the message to LL that it needs to bring back the workable GUI.

[Hamlet Au]

"Lower land tier prices by one half to two thirds what they are now. "

Alas that just can't and won't happen:

http://nwn.blogs.com/nwn/2012/08/linden-lab-post-sl-era.html

[Archangel Mortenwold]

I know it probably won't happen, but that doesn't mean it shouldn't. At this point lowering tier prices is the only viable way to make Second Life competitive again. As you pointed out, the vast bulk of Linden Lab's revenue comes from the renting of virtual land and the land barons are leaving because people can no longer afford the ridiculously high tier prices. Virtual land — practically a requirement in order to realize the full potential of one's SL experience — has become prohibitively expensive. With no alternative sources of revenue visible Linden Lab has to lower tier prices to something reasonable. If they don't, no amount of marketing SL as something it wasn't designed to be and never will be is going to attract and keep new users. Meanwhile, OpenSim and other competing grids are experiencing steady growth as their performance improves and virtual land tiers are priced within people's range of disposable income.

Linden Lab can do the smart thing and lower tiers (and initial buying prices) to something affordable, or it can go out of business within the next few years. That's the reality of the situation.

[RexxHuet]

First time giving my 2 cents here. Hamlet enjoy your blog, come here a lot.

Wanted to comment on Archangel response. Point 1 ok, point 2 ok, point 3…. Ouch. Hope they do not do away with V3.3 UI. Whish people would not group V2 and V3.3 together. V3.3 is very different from V2. For perspective I started on V1, moved to snowglobe, then Phoenix. V2 came out and I tired it. Found it to be confusing and slow. Felt lost when using it, so I uninstalled it. Then some one tipped me off to Singularity and found it to be faster and smoother running then Phoenix. Have looked at Firestorm based on V3.3 but they have added so many options it looks confusing and runs slower. Watched a short video on Firestorm options, my head was spinning at the end. Also see they have in world glasses but do not care about spending one or two hours in glass learning a viewer. Couple of months ago I downloaded and installed V3.3 for fun. Wanted to check out want changes they had made. Surprise surprise, I like it. In ten minutes I felt comfortable with it. Changed the buttons so now have few on the bottom that I use; also gives more screen space for in world. The menus are intuitive and found most every thing right away. It is faster then Singularity and never crashes. (probably just jinks myself, will probably crash next time I use it) Did make two changes in preferences; double mouse click instead of one mouse click to move avatar and changed im floating windows to one im window with tabs. Tried the new update to Singularity and found it to be slower then V3.3 and the V1 UI seems clunky to me now. Think its cool we have TPV’s and gives us choose which is good. Only thing I can say is try V3.3, not v2 or Firestorm, but V3.3 and see if you like it.

[Archangel Mortenwold]

That's exactly my point about the V1 GUI. It's pretty straightforward and easy to learn. There's no need to spend hours taking classes just to figure out how to turn off the media players or locate the chat windows. I've tried the latest Firestorm and found it severely lacking. V3.3 frustrated me to no end as it has everything in the wrong places and is basically V2 made worse (in my humble opinion). As Phoenix, Singularity, and Cool VL Viewer show, putting the V1 GUI onto a V3 viewer base isn't impossible, nor is it as difficult as some would have us believe, and I'll happily stay with those. At the end of the day the viewer preference is moot since there are options available for people who have made their choices.