Developer Gamzee recently scored $1 million in seed funding to develop HTML5-based games for mobile devices and social networks. Apple, Facebook and other companies are all pushing to improve the markup language, in order to enable applications to run across computers and other devices more easily. However, the standard itself is still being developed, as is Facebook’s role in helping to promote social-mobile apps.

With HTML5 still coming into its own as a programming language for social and mobile games, few developers have released games that actually do run seamlessly across networks.

We’ve seen larger social game developers take steps toward realizing this goal, with Zynga acquiring an HTML5 engine just last fall and EA building an entire game and mobile component with it in 2011. The question is what Facebook decides to launch in addition to what it already provides for mobile devices, such as single sign-on for iOS. For now, app stores on native applications for iOS and Android have been the main way that people discover games and other third-party titles.

Josh Levitan, VP of Games at Gamzee, walks us through the challenges of developing an game in HTML5 that works on all platforms. Levitan joined Activision and Acclain co-founder Howard Marks at Gamzee after launching a social games division at Nexon (which is just now getting its first social game out in the form of Maple Story Adventures) and previously working at Playdom on a game called Kogamu, which was apparently shelved by Playdom following a beta test period. Gamzee’s first game is designed for mobile, but will work on Facebook as well.

Inside Social Games: Why HTML5?

Josh Levitan: Two reasons, the first is Facebook announced back in February that they were redoing their platform in HTML5 for mobile. Basically we saw that as a great opportunity because when Facebook released their web platform, they had all these great things where you could put games out as apps that would run inside Facebook — with all the Facebook frame, with promotion, with discovery… and they’ve added all sorts of things like Credits to enable people to purchase items in game more easily. They’re [apparantly] going to be doing the same thing on mobile.

The second reason [for our company] is that, having operated a bunch of games at Acclaim that are download games for PC and then making Facebook games that were Flash games at Playdom, we wanted the opportunity to make [multiplatform] games because as a gamer, I want to be able to play my game on whatever device I happen to be on. You can’t really do that right now. There are some games that allow you to have the same sort of integration across servers, but it’s usually a different client, you have to download an app for your iPhone… and that just gets really messy. With HTML5, you can make the same game on all platforms.

ISG: What are some of the challenges to being truly multiplatform? We’ve heard some developers say it’s just too much of a hassle to deal with phones’ variable WiFi connections…

Levitan: So phone speed is certainly one challenge. That seems to be going away with the fact that a lot of phones — pretty much all smartphones now — have WiFi connections and 4G is becoming more prevalent in a lot of cases.

The other big challenge is just differences among different devices and having to support a lot of different devices like the screen resolution on an iPhone 3GS is different than the screen resolution on an iPhone 4. And the way that those guys handle javascript and certain elements of HTML5 are different on those platforms as well.

ISG: Do you find yourself focusing on specific mobile devices just to cut down on all the demands?

Levitan: Our goal is to have it work seamlessly on all platforms, but we’re going to be focusing on iOS devices and Android, because those are the two biggest chunks of the market. Then we’re going to expand out to other things like Windows Phone 7 and Palm.

ISG: You said the game would also work on Facebook. What about on other social networks?

Levitan: Right now, we’re just looking at Facebook, but the way we’re architecting our code, we’re sort of platform agnostic. So if in the future we want to integrate our game with a social network in China or Japan, we’re able to do that. Or if Google gets their social network working and that turns out to be great, we can integrate fairly easily with that platform as well. Our focus is definitely on Facebook right now because it’s obviously the 800 pound gorilla pretty much over the entire world. There’s just so many users and so many people that play games and so many people that have proven they like to spend money on good games.

ISG: Do you find that being platform agnostic puts limits on your game design? Social games and mobile gamers have very different play habits, for what we’ve seen.

Levitan: There are certain differences. The first game that we’re doing is going to be consistent across the platforms, so we picked something that lends itself well to short bursts or longer play. We don’t want to give away what we’re working on yet. We’ll just say it’s a traditional Facebook gaming genre that’s done very well.

But for example, city-building games have proven to work pretty well on Facebook and then there are similar games on mobile — [like] Tap Zoo or Smurf Village or Zombie Farm or something like that — where you could call them the same type of game even though they’re on mobile.

ISG: And now we’ve got CityVille coming to mobile, too.

Levitan: Exactly. But that’s a good example where it’s not the same game. I can’t play my city on my phone. It’s a separate app that has some light [Facebook] integration and we think that gamers want to be able to play their game anywhere. Even on their TV if it has a web connection.

ISG: What are some of the limitations you find with developing a game in HTML5 for mobile? We’ve heard that first-person shooters are definitely out for now…

Levitan: I certainly wouldn’t try and do that right now. There are definitely some limitations for doing things on mobile. Just because of the fact that there are different implementations of hardware acceleration and the way that graphics get processed on mobile than for desktop.

When you’re looking at HTML, if you’re using canvas for example, canvas performance on desktop is almost identical to the performance you can get out of Flash on a desktop. But if you’re running canvas on an iPhone 3GS, it’s pretty bad at the moment. So the kinds of games that we’re focusing on are not 3D. They’re 2.5D isometric games. Similar to ‘Ville games or Ravenwood Fair, because that’s something that javascript and HTML are able to handle fairly well even on a mobile phone.

ISG: How much longer do you think it will take before HTML5 could handle something along the lines of a 3D first-person shooter?

Levitan: We’ve played around a little bit with iOS 5 — they have the dev version out — and that runs canvas pretty well. So you might be able to do [shooters]. We haven’t played around with doing 3D graphics or trying to do something that simulates 3D. WebGL is also coming for phones in the next probably year or two and once that’s in place, it should be able to handle 3D just fine.

[HTML5] is not a mature standard yet and there’s still some holes in it. For example, audio. There isn’t a great way to do the same kind of audio that you would in a regular application and so, for example, a lot of times when people are making web apps in HTML5, you’ll have a fallback audio player that loads a separate Flash player just to play audio files. Which is obviously kind of a kludge-y solution and it wouldn’t work at all for mobile. And canvas is slow, and there are a few other things that aren’t locked down yet.

But overall, people have been able to make some pretty good apps with it. People have been able to make some games — you can find a bunch of demos of all sorts of different kinds of games, like arcade games, 2D shooters, platformers and things like that. Facebook actually got Doom 2 running on HTML5 and fairly well. They’re pretty heavily invested in it — on their developer site, they will put up little benchmarking tests that they’re doing just to show the difference in like sprite performance or texturing amongst different mobile platforms. For whatever reason, they decided to do a java port of Doom 2 in HTML5 and canvas and it was pretty neat.

ISG: If you integrate Facebook Credits for the product, that implies Credits will work on mobile. But, how will that work on mobile? Do we have to pay Apple for a download from the App Store and then pay Facebook for in-game purchases?

Levitan: There’s still a bit of mystery as to what the [Facebook] mobile platform will support. We expect that it will support Credits, because otherwise they’d be giving up some money. If for some reason it doesn’t, we still plan on using Credits because we can just integrate with the regular Credits API and treat it as a regular Facebook app because our game is basically a web app that just happens to run on any platform.

We may at some point have an app that you can download through the App Store, or we may not. Our game is like a regular Facebook app where through the mobile platform, you’ll just authorize the app and because it’s a web app, you won’t actually have to get it through the Apple App Store or the Chrome app store or something like that. We may put out an app for you to do that for additional discovery, but basically that’s just a shortcut that will run the web app.

ISG: What about getting your game onto Android? Does that platform have any differences that will impact your game?

Levitan: It seems like you make, for native [Android] apps, about half the money that you make on the iOS ones even those there’s way more [Android] phones. I think Apple just had a much better experience overall for downloading apps. Purchasing is one click. Apps just tend to work. Whereas Google’s been a little bit slower, they had that weird thing where you could buy apps and return them for a long period of time… I think all of that just made it a little bit clunky.