Interview with iPhone developer Josh Rosen

Josh Rosen is unlike most other developers we cover here at Know Your Mobile. He's only 17 years old - but take a look at the video preview of this upcoming game SciFly 2 and you'll see that his work doesn't look like something produced by a minor.

SciFly 2 isn't his first game either, it's his third. He made Chalkboard Calculator and SciFly while still at school before deciding that iPhone game development was worth pursuing full time.

He took some precious minutes out of his busy schedule to talk to us about his experiences with game development, what he thinks of the mobile market and how aspiring game developers can hone their skills from home.

First though, we had to find out exactly how he made a pair of games from scratch while still having to deal with homework.

Know Your Mobile: How did you manage to find the time to develop two iPhone apps while still being at school? Don't you sleep??

Josh Rosen: Sleep, what's that??

While I was in school, skipping sleep was definitely a big part of it. I also took a Computer Science class and a 2D art class so I could work on SciFly during school as much as possible.

However, since then I've switched into homeschooling, which has lead to a very different schedule. Yesterday, for example, I literally worked on SciFly 2 for 30 hours straight. Of course, after that I have to crash for a long time. I think it's efficient to do things that way because there is a lot of overhead associated with starting and stopping a project every few hours.

Anyway, that's getting off-topic...


KYM: When did you start programming?

JR: I am very lucky to have a dad who is a programmer. I remember him teaching me the basics, like how to define variables, when I was very young; I'm guessing age 9 or 10. The funny thing about that is, now I'm teaching him how to program 2D games for the iPhone.


KYM: Do you have any advice for bedroom game designers looking to make their own iPhone games?

JR: Absolutely!

If you are looking to create 3D games and are new to game development, I highly recommend "modding" (modifying) an existing game like Unreal Tournament 2004.

The great thing about this is that you can choose to focus on artwork or programming. I recommend grabbing Maya PLE from Autodesk and learning how to make basic 3D models. With just a few lines of code, you can import those models into Unreal and make your own tanks and hovercrafts and such. It's pretty awesome.

As for the iPhone specifically, I use my own engine, but it seems like Unity 3D is a great tool. It gives you similar benefits of modding in that it emphasises artwork and leaves programming to tweaking the details.

Links
Maya Trial
Unity 3D
Unreal Tournament 2004


KYM: SciFly II is looking to be a lot flashier than its predecessor. What did you learn from making the original SciFly?

JR: The original SciFly was basically an experiment to test if the gameplay fundamentals of 3D dogfighting games are fun on the iPhone. Although by today's standards the original SciFly is very basic, I think it's pretty fun.

So, from a gameplay perspective I definitely learned a lot about dogfighting games. From a technical perspective, porting over the graphics engine I wrote for the Mac to the iPhone was an important thing to get done before I could start working on larger-scale games.

KYM: One-man-band games tend to have trouble keeping the quality of their visuals and gameplay consistent. Do you spend a lot of time perfecting the graphics in your games?

JR: The original SciFly's graphics engine was very basic because I had to spend a lot of time just porting and removing features from the Mac version. I didn't have very much time to perfect it for the iPhone hardware. However, with SciFly 2 I started with the fully functional engine and spent several months maturing its core rendering code to work much better on the iPhone. A lot of this work went into enabling the engine to handle the massive open-world environment of SciFly 2. Instead of using a menu, the player can actually fly from level to the next.

KYM: SciFly was a fairly pure aerial dogfighting game. Does SciFly II stick to this formula, or is it a more complicated beast?

JR: I suppose you could say that the goal of SciFly 2 is to put the basic gameplay of the original game into a much more full-featured environment. While I have certainly tweaked the gameplay, I've mostly been working on creating a richer experience for the player that's built around the fundamental gameplay of the original game.

With SciFly 2, I'm adding aircraft customization, open-world environments, better AI, better controls and all that good stuff. But, when the player engages an enemy aircraft, it's not too far off from the original.

KYM: What are your favourite iPhone games?

JR: Changes all the time, but right now I'm playing Zen Bound, F.A.S.T., Eliss, Glyder, and Rolando 2.


KYM: If you could add one element of functionality to the iPhone to make it even better as a gaming device, what would it be?

JR: The last time I was asked this I mentioned that the orientation detection of the accelerometer lacked the ability to measure all 3 rotation axes as the Wii controllers can. It's possible that the 3G S's magnetometer ("compass") solves this problem, but haven't had the chance to test this myself yet, though.


KYM: How do you think iPhone gaming, and its audience, will develop over the next 12 months?

JR: Looking back, it's clear to see how rapidly the iPhone games industry has changed. Only months ago, the Top Paid Apps list was notorious for being flooded with ridiculous $.99 fart and flashlight apps. With games like Rolando, Sims 3 and even DOOM (just to name a few), this is definitely not the case anymore. Now that developers, and more importantly consumers, are realising what the hardware is capable of, the demand for high-quality games will continue to rise.


KYM: The future of the iPhone games market is a frequently discussed topic. Some say most smaller developers will be squeezed out before long as the quality watermark rises and game development budgets rise. Do you think this is going to happen?

JR: While I don't think the AppStore will ever resemble a market similar to the big consoles, I do think the hardware is capable of large-scale games that will be difficult for one or two man development teams to produce. While there are still a few small-sale games on the current Top 25 games, the list is pretty well dominated by large-scale games.

Just looking at a game like the original SciFly, you can see the trend. SciFly was released about 8 months ago and did fairly well. It also able to get coverage from many of the major review sites.

If I was to release a game like that today, it would just get laughed at. I've had to seriously step it up to get SciFly 2 to the point where it can compete with games like DOOM and Hero of Sparta. There's only so far an indie developer can keep up with larger operations like ngmoco and Gameloft.

Of course, there will always be those timeless simple 2D games that will continue thrive regardless of the hardware capability. I doubt games like Pocket God, Eliss or Peggle should feel threatened by DOOM.

Also, it's definitely worth mentioning a few things about the Top Paid Apps ranking structure. Since Apple has decided to rank by units sold and not revenue (units sold X unit price), there has been a huge advantage to small-scale games from developers who are very willing to sell their games at $0.99. Recently, however, large publishers are beginning to eat up those last few top ranking spots by dropping the prices of previously released games which have outlived their original $5+ shelf life (namely, Hero of Sparta from Gameloft).


KYM: How do you think the increased power of the 3G S is going to affect iPhone games released in the next six months or so?

JR: There will definitely be a development lag for the 3G S because right now you just can't ignore the 40+ million previous generation devices. However, I suspect we will start seeing games that have an option to enable OpenGL ES 2.0's hardware shaders for post processing effects like motion blur, bloom, and depth of field very soon.

Personally, I can't wait to start implementing these features because I love working with shaders. But, I don't plan on having shaders until after the initial release. In fact, I doubt there will be many games which are 3G S exclusive until 50% of iDevices are OpenGL ES 2.0 capable. I imagine this wont happen until either the 3G contracts start running out, or Apple leaves the AT&T exclusivity contract.


KYM: You've obviously got talent on the game development front, but you seem to have business acumen too. Would you be tempted by an offer to join the staff of one of the big game publishers, or is it all about going it alone?

JR: I definitely hope to either create or manage a game development studio one day. Right now things are great because I have 100% creative control over my projects. I would definitely miss that benefit if I was to get hired by a large publisher.

The thing is though, I would love to work on games that are on an even larger scale than SciFly 2. To do that, I would either have to get hired by a larger developer, or get the funding to form a studio of my own. Since the latter is not really practical, I am definitely interested in the former. It would be great to work for a developer who clearly has some creative spark left in them. To be specific, I am definitely interested in developers like ngmoco, Firemint and Glu.

...Neil, if you're reading this, call me :)


KYM: What happens for Josh Rosen after SciFly II?

Assuming I don't get hired, I have a few big ideas for games that are heavily multiplayer based. If SciFly 2 is somewhat successful, I will be traveling to attend any developers conferences I can get myself into. So far I have been to GDC, E3 and WWDC, which have all been absolutely amazing networking opportunities.

That's all I really know at the moment.