Cloud gaming is rife with naysayers and glaring backfires. Nintendo president Satoru Iwata downplayed cloud gaming last year to investors citing latency as a major drawback. Cloud gaming company OnLive went bust back in 2012, forcing it to restructure. But cloud gaming is not going away any time soon, having recently seen Sony’s purchase of Gaikai and its cloud gaming technology for its PS4 and PS3.
While replicating the intense graphic experiences of games like Halo 4 or Battlefield 4 via the Cloud remains a difficult feat, we know that what makes up a great game is more than just dazzling 3D graphics. Otherwise, why would Angry Birds, MineCraft, or the incredibly addictive 2048 become such huge hits? Nielsen estimates there are more than 115 million TV homes in the United States with nearly seven million people who don’t own a console. Talk about a large untapped opportunity.
Leave it then to a gutsy Israeli, Guy De Beer, CEO and founder of PlayCast Media, who seeks to take on the U.S. market later this year. PlayCast Media is the only game streaming service that is currently delivering games and making money at it. PlayStation Now and others are still in their infancy. Guy was at this year’s GDC promoting PlayCast Media’s cloud gaming platform and I sat down with him to learn more about why his company has done so well selling cloud gaming abroad and why other cloud gaming companies have fared so poorly.
Q. PlayCast Media has done very well in European countries like France and Portugal. What is the key to your success?
A. We have a different definition of what cloud gaming should mean for consumers. The user experience, latency, video, the cost structure, and overall quality of the experience, including packaging, are the types of issues we go after when it comes to attracting users. We provide a level of service unlike any other cloud gaming platform currently out there today.
From our perspective, we’re going after the casual users who are looking for the console experience but in more of a Netflix type of model. You can’t assume these users will buy your hardware. They already have lots of hardware at their disposal. We’re offering more of a platform, which is on the server side, purely for software, but still provides you with the performance of an embedded system that you control.
Q. You’re going to be expanding into the U.S. marketplace soon. What challenges await you?
A. We went through the labor of getting the quality to be where it is now, and over the past seven to eight months we’ve been getting to know the US markets. We’ve taken our international experience and are applying it here although networks in the US are so different than in Europe. It’s not as much about bandwidth availability (although that issue comes up in different parts of the United States); it’s more about QoS packet loss and network phenomenon. Today, Netflix generates 30% of Internet traffic and the networks are loaded. Even in places like the San Francisco Bay Area, everything works great until 5:30pm in the afternoon when everyone, including kids, jump online and can overload the networks, whether it’s a 2 megabyte per second stream or an 8 megabyte per second stream.
Q. How did you handle a problem like overloaded networks?
A. We had to pretty much develop new types of algorithms to handle this overcapacity. The way you deal with network issues is through buffering, TCP, etc…Our threshold is to do a single frame on the single server side and a single frame on the device. It doesn’t leave you with a lot of latitude. To work out any network issues, you have to do it on the fly. Also, the subframe speeds bring in a new plethora of issues. We’re now at the point, however, even using say, hotel wifi, in terms of robustness, the level of service we can provide is very good. We also brand our service over DSL, cable, legacy cable, on demand, and fiber, too.
Q. At the end of the day, what makes or breaks a cloud gaming user experience?
A. Our focus is exclusively on the user experience because that is the only thing that matters. The hardware side is not an issue. On the content side, we’ve been serving products from the major game publishers for three years now. We know which titles should kick in and the publishers know what to expect from cloud gaming. It’s a small footprint, Europe and Asia. By now we have a good amount of content. We have approximately 15 publishers, including AAA (from Sega to Warner Brothers), and offer 80 + games now.
We’re creating a package that has enough value to retain users, give them perceived value with the newer titles and give them a lot of titles to choose from.
Q. Talk to me about cloud gaming consumption patterns. What do they look like?
A. Consumption pattern is different with cloud gaming. It’s more like channel surfing. For the first person shooter, average session would be 50 minutes. A gamer on a typical gaming day will play up to 45 games, play for 20 minutes one game then move on to play another game, which is pretty standard behavior for kids and/or teenagers. These patterns are not all that dissimilar from usage patterns with television, or even to some extent, mobile. But it’s all happening within the same transaction of users. They subscribe to a package and then play, in and out, and jump between games.
Q. What about your packages? What will you offer in the U.S. and how similar is it to what you’re already offering in markets like Europe?
A. In France, for example, we have three packages, one that sells for 7$ all the way to17$ monthly, respectively. In the United States, we will also offer a series of similar monthly package deals, some will be for all you can play or an a la carte offering for those who only want to play select games. We will be rolling out a free to play or a free pack model in one of the markets very soon as well, where users can have 5 to 6 games for free. With this kind of cost structure you can bundle it in many different ways but always making sure you are providing enough value to the user.
Q. When you are launching in the United States?
A. We’re starting with a public trial with one of our platforms shortly and I expect by around June we’ll be offering this service for trial. We’re now opening it up for a limited number of users but once we see it ready for larger acceptance, most likely we’ll fully launch towards the end of 2014.