The Sad Lesson of Online's (Near) Demise

3 Sep

I remember back in 2002-2003 looking at all those online TV concepts and (as a potential customer) finding it amazing to have so much choice and access to niche content that mainstream media would never offer. In France there was a company called Canal Web which offered really amazing stuff. They folded in 2003: the quality of content delivery over the broadband that was available at the time (and still a fair portion of the market on narrowband) simply wasn’t sufficient for such concepts to soar. I also remember back around that time, when looking at Fastweb and all the innovation going on in the Italian market that there was a precursor to Youtube in Italy (the name escapes me now). They folded also.

Two years later, Youtube was launched, and a year after that, the writing was on the wall, enough so that Google quickly grabbed the company for $1.65bn in stock. Eyeballs.

This stroll back into the early days of the internet as we know it is a necessary introduction to the point I want to discuss today, namely the demise (or as close as it gets to a demise) of Onlive, the innovative cloud gaming company based in San Francisco. A couple of weeks ago, Onlive restructured radically, got rid of their CEO of the time, laid-off half the workforce and was bought over (see Ars article for details). We don’t know much about why Onlive had to go through such a radical restructure (apart from the inescapable fact that they weren’t selling enough, I guess…)

One safe bet though, is that history is repeating itself: if you want a high quality cloud gaming experience, you need high quality broadband, and especially broadband with a decent upload rate. Addressable customers worldwide with such a good quality broadband service are limited, to say the least, and even in the US their numbers are limited. Onlive has the foresight of signing a number of deals with telcos offering such broadband (like Belgacom) but as far as I’m aware, there were very few of those deals, certainly not enough if the recent issues are any indication. I’m not suggesting that full FTTH is necessary to enjoy a service such as this one enough that you’d want to use it frequently enough to subscribe. But it’s quite clear that the regular ADSL or cable offering that most of the western world has access to today, with its 1M upload rate (tops, if you’re lucky) is insufficient.

The sad thing is that when the tipping point for good quality broadband is reached, it will seem obvious looking back that this was a meaningful service concept, one users would be willing to pay for if they have good enough broadband (and are at least casual gamers, obviously). Just like there were many Youtube lookalikes before Youtube who all died for lack of decent broadband.

Edit: Marc Canter points me to a much longer article (Onlive Lost) that delves into some behind the scenes. I think it should be read with caution, but it points to a number of very bad management decisions as the main cause for the downfall. However, the article also says:

If you ask employees, they can point to any number of reasons why the paying users didn’t come. Broadband in the United States is fickle, and even in households with a good connection, if one family member started up a YouTube video, that could disrupt the other’s game. Within those bandwidth constraints, the service streamed games at a compressed 720p resolution, on par with consoles, but which didn’t look as good as titles rendered by a proper gaming PC. That could have alienated early adopters, whom the company relied on for positive word-of-mouth. 

 

13 Responses to “The Sad Lesson of Online's (Near) Demise”

  1. Martinalderson September 3, 2012 at 11:31 am #

    Definitely doesn’t need more than 1mbit up. All the upstream is used for is sending controller commands. The downstream, on the other hand, needed to be about 3-8mbit/sec for the picture to be decent quality.

    What went wrong was the huge amount of equipment and servers they bought. The had >8000 servers with (I’d imagine) multiple very high-end GPUs. They generally couldn’t get more than 1,000 concurrent players at any one time.

  2. marccanter September 3, 2012 at 12:10 pm #

    Hey Ben

    You want details – here are the details:  http://www.theverge.com/2012/8/28/3274739/onlive-report

    This is a classic Silicon Valley story.  Steve Perlman has made a career of failure.  Good news is that he was able to get OnLive as far as it did.

    But declaring exclusive deals, treating employees like shit and….

    Well lets just say “it’s all about execution.”  Good ideas are a dime a dozen, it’s ALL about execution!

  3. Franca Condo September 6, 2012 at 9:04 am #

    Can you believe David Perry got away with selling Gaikai to Sony for a whopping $400 million?  Its hilarious.  

    The truth is that the problem with these services is not with BANDWIDTH.

    The problem is that it takes a whole PC to run a game like Call of Duty and thus Gaikai and Onlive would need millions of servers to really offer these games to the masses.

    Gaikai sold a demo to Sony…. a Fantasy…

    Its nice that they can claim to be a future technology.

    Maybe Facebook should do the same.  Pretending they will make money in the future but not yet today.

    Its bullcrap.

    Truth is that Sony will be out of business before Gaikai type technology ever happens?

    Do we really need less local CPUs?  No.  Our Phones now have strong CPUs.

    Cloud gaming should be defined as Multi-Player gaming.  Not that a company sends the entire screen to your device 20 times a second.  That is just stupid.

    If Gaikai really had something they wouldnt have sold out.  They had just raised $50 million.  

    Truth is they had nothing.

    And Sony is so stupid for buying them.

Trackbacks and Pingbacks

  1. Marc's Voice » Beginning of Sept. 2012 blogging - September 4, 2012

    [...] Many hopes and dreams were dashed with the failure of OnLive.  Its a real Silicon Valley story – and one of execution.  Steve Perlman’s career has been based on great ideas and poor execution. [...]

  2. When it comes to startups timing really is everything — Tech News and Analysis - September 5, 2012

    [...] not forget about the role of timing. I was reminded of Andy’s comments when I read Benoit Felten’s post about the demise of Onlive. While there are  a multitude of reasons, including architectural problems, it seems that the lack [...]

  3. When it comes to startups timing really is everything | Apple Related - September 5, 2012

    [...] not forget about the role of timing. I was reminded of Andy’s comments when I read Benoit Felten’s post about the demise of Onlive. While there are  a multitude of reasons, including architectural problems, it seems that the lack [...]

  4. The Wright Computer Services » When it comes to startups timing really is everything - September 5, 2012

    [...] not forget about the role of timing. I was reminded of Andy’s comments when I read Benoit Felten’s post about the demise of Onlive. While there are  a multitude of reasons, including architectural problems, it seems that the lack [...]

  5. When it comes to startups timing really is everything | Turning Social Media Views in to Customers - September 5, 2012

    [...] but let’s not forget about the role of timing. I was reminded of Andy’s comments when I read Benoit Felten’s post about the demise of Onlive. While there are  a multitude of reasons, including architectural problems, it seems that the lack [...]

  6. Global Tech Review | When it comes to startups timing really is everything - September 5, 2012

    [...] not forget about the role of timing. I was reminded of Andy’s comments when I read Benoit Felten’s post about the demise of Onlive. While there are  a multitude of reasons, including architectural problems, it seems that the lack [...]

  7. When it comes to startups timing really is everything |Trax Asia™ - September 5, 2012

    [...] not forget about the role of timing. I was reminded of Andy’s comments when I read Benoit Felten’s post about the demise of Onlive. While there are  a multitude of reasons, including architectural problems, it seems that the lack [...]

  8. GIASTAR – Storie di ordinaria tecnologia » Blog Archive » When it comes to startups timing really is everything - September 5, 2012

    [...] but let’s not forget about the role of timing. I was reminded of Andy’s comments when I read Benoit Felten’s post about the demise of Onlive. While there are  a multitude of reasons, including architectural problems, it seems that the lack [...]

  9. When it comes to startups timing really is everything | Electronic Staff - September 5, 2012

    [...] let’s not forget about a purpose of timing. I was reminded of Andy’s comments when we review Benoit Felten’s post about a passing of Onlive. While there are  a crowd of reasons, including architectural problems, it seems that a miss of [...]

  10. When it comes to startups timing really is everything | Xcluesiv Cloud Technology Blog - January 7, 2013

    [...] but let’s not forget about the role of timing. I was reminded of Andy’s comments when I read Benoit Felten’s post about the demise of Onlive. While there are  a multitude of reasons, including architectural problems, it seems that the lack [...]

Leave a Reply