When do you decide to go multi-platform?

[unexpect3rd]

once again, thank you @Frand, very insightful. I will have to read it couple more times to digest it. You should consider writing a blog post on Gamasutra regarding this matter, it could raise some worthy discussions.

[mobile1up]

the best time to go multi-platform is to make a decision before you even start writing apps.

choose a language that can be used on multiple platforms (objective C - is not one of these languages), the better are C or C++. you can bridge C/C++ with objective-C, java et al.. so, this is the natural choice.

try to separate your game elements from platform specific code. this will aide big time for you to increase your chances of porting. then all you need to focus on is the platform specific elements, link against your "common" business logic code (game logic et al) and you are done.

we have done this approach, our games have zero platform specific code inside them, when we add a new platform, we just recompile the game against the new platform layer. it took us 1-2 days to support webos, blackberry 10 and the same will happen for android if we ever bother.

[Frand]

Originally Posted by mobile1up

we have done this approach, our games have zero platform specific code inside them, when we add a new platform, we just recompile the game against the new platform layer. it took us 1-2 days to support webos, blackberry 10 and the same will happen for android if we ever bother.

Writing clean platform-agnostic code is of course absolutely essential for porting (our games take the same approach as well), but the bulk of the work is not getting a build running on a single device. That indeed is usually something that takes a few days.

The problem, particularly on Android, is those thousands of different handsets with different hardware, different memory amounts, different audio implementations, different graphics drivers, different versions of the operating system itself, different screen resolutions etc.

On Android you will need to deal with that sea of complexity, either at porting and QA phase (by acquiring devices to test with), publishing phase (marking only a limited subset of devices as capable of running the game, therefore limiting your market), or in support phase (dealing with all the emails from handset owners experiencing compatibility issues).

[mobile1up]

Frand, you and i have been doing this stuff for 10+ years..

you hit the nail direct on the head why i have not even bothered with android fragmentation is a killer on that platform. i may go for some specific markets like the amazon kindle but to focus on the main group of android devices? it will just be too much overhead to support than to actually make it worth my while.

other alternatives to ios still exist, blackberry 10 (current: blackberry playbook) is doing very well as a platform for us. webos still makes sales, but it would be risky to put any investment into it right now - we need to see how the open webos environment will spin out.

(work plug)
if any developers are interested in blackberry 10, contact me. i actually work as a senior technical evangelist for RIM - i ported my own games very quickly and i have a tonne of references to porting guides from iOS to blackberry 10. games are even easier to port. not to mention the market place is not saturated yet, like iOS back in 2009.
(/work plug)

[Ovogame]

Originally Posted by mobile1up

if any developers are interested in blackberry 10, contact me.

What's the market there? My cross platform engine in C++, OpenAL, (OpenGL1.1/DirectX 8). How easy would it be to port, knowing that the platform specif stuff is tiny and already extremely encapsulated in few files. Is there any help we can get if we do the work (free device, promotion)... Check my games in my sig.


JC

[mobile1up]

Originally Posted by Ovogame

What's the market there? My cross platform engine in C++, OpenAL, (OpenGL1.1/DirectX 8). How easy would it be to port, knowing that the platform specif stuff is tiny and already extremely encapsulated in few files. Is there any help we can get if we do the work (free device, promotion)... Check my games in my sig.

contact me on aardiri@rim.com - it would not take you very long to get your applications ported over - especially since you are using C++, open AL and open GL. these are supported out of the box. since your profile says you are in france, i can even send you a blackberry playbook. i am looking specifically for developers in europe

there is a bb10jam soon to be done in stockholm and malmo, where i will be speaking - you can qualify for a blackberry 10 dev alpha there as well. http://www.blackberryjamworldtour.com/

as i said, follow up with me offline and we can chat about the market. if you get all your apps converted, i can ensure that you are featured in the app world market place as well.

[R3v]

There is really no one working multiplatform with Adobe AIR? Flash games on desktop, native apps on iOS + Android.

[Ovogame]

Originally Posted by R3v

There is really no one working multiplatform with Adobe AIR? Flash games on desktop, native apps on iOS + Android.

Yes, there are people working like that, but they are coming from the web dev scene.

JC

[ChaosInteractive]

Hello mobil1up. I am interesting to port my games on blackberry. I have send to you an email.

I have also create my own C++ opengl cross-platform game engine.

Edit: And to answer to the topic: I wrote a cross platform game engine because I started working on Nintendo DS. I developed on PC and compiled for NDS. It is just easy to port now my game engine.
When Androïd out, I decided to port my game engine on it because I believe in it.
Later, I added Bada OS for a samsung contest.
Now I come on this great forum for have advices on how to port my game engine to iOs. And why not on others platforms.
Indeed, it is to earn more money. And it is very cool to develop on different device. It's always a little challenge even if now I am accustomed.

Thanks

[mobile1up]

i want to share with you the story of the first developer who contacted me

http://crackberry.com/smileys-pop-blackberry-playbook

some of you are in touch with me; and i expect the opinions to be the same (yeah, this is crap - porting isn't so easy et al). but, as you can see, Jean Claude has been impressed by the tools and the commitment that RIM has a company personally; i like reaching out to developers directly - working one-on-one with them.

so, who else is interested?