I am probably going to work on a game project soon which is aimed at Android and iOS. Afaik jME does have this cool iOS Deployment by now. But I wonder how well that is working and if it can be taken into account for a semi professional 2D Sidescroller Jump-N-Run. Or would you say that I should rather go for Unity or another Engine if I just plan to make an App-Game for Android and iOS.
Do you guys have any experience on that? Or even anything you can tell me in advance I have to care about? Any known major cross-platform issues I should take into consideration when making a time schedule? Which leads me to do you think it is realistic to make a 2D Jump-N-Run in 6 months with a programmer, a grafic designer and a project/marketing leader?
I am happy to hear anything useful from you. Thank you.
Well acutally because I am quite familiar with jME but you are right it is not that I don’t consider different engine. (Maybe you also have an idea what would be best for that) It is just I try to check capabilities of engines. And also it might become a pseudo 3D 2D game (if you know what i mean) that has not really been defined yet.
Hmm good point I will take that into account. And actually that is also what I told the project manager that I have to know more about the games graphic style and mechanics before I can make a decision what to use. But he wants kind of an overview what is possible.
And do you think 6 months for a 3 man team is realistic for a mobile jump n run of moderate length for Android and iOS?
Specifically are their huge problems to be expected when aiming at both or is it more or less a common task nowadays with lots of tools who help you with that?
Would you say porting to Android or porting to iOS is more easy? Or would you develop it in parallel?
Yep, totally feasible. Much more so if you’ve previously done a 12h project, a 1-week project, a 1-month project and a 3-month project though
If you haven’t done any of the above, I would add a month to your timeline and spend that first month creating a heavily down-scoped version of your game that you imagine should be doable in no longer than 1 month. By the end of that month you stop development on it no matter what, and from there you can reevaluate whether those 6 months to create the real thing seem realistic.
Android is a pretty obvious “safer bet”, since it’s Java-friendly, and it’s been a deployment target in jME3 for years now. iOS on the other hand is new, but it should work.
Make a prototype and do test-deployments both as early on as you can. Make sure there’s nothing in your core mechanics that blocks either platform. Physics for instance tends to be somewhat unpredictable.
I can speak only from my own experience, and for me the time limiting factor would be the asset quality. But thats maybe because i really suck at blender/photoshop. Still from what i read and hear a asset creation seems to take up a very large potion of the game dev lifecycle.
I would say developing parallel. Develop and test on pc. Retest and make it work on the actual devices. Beside jme i would throw in libGDX.
Regarding the time, that is always feature dependent. Not possible even to estimate without having any infos. Also when calculating time keep in mind that you might required a few specialized tools to be written (level designer and so on) Those can be very big timesinks which are not directly game features…
All in all, i am sticking to my first anwer, setting up the requirements would be the very first step i would take. Maybe having a list of Required, Optional and NiceToHave features. At least thats the way i would start as a project leader…
As a note, my experience comes only from not game dev related projects. Gamedev for me is only a fun time hobby. But i have seen projects failing because of very very bad time managment a lot. (And probably even more because there where no feature plan) So thats why i am kind of biased toward always having a plan!
Jmonkey is an amazing engine and is so easy to pick up that an experienced developer should lap it up. It’s works perfectly with Linux, Windows, Android and OSX and provided you stay away from native methods it is completely cross platform. Fantastic.
However, I must warn you about IOS development. The plug in is in pre-alpha and works fine for core JME (except NiftyGUI) and also has some issues with other GUIs. I’ve tried Nifty, Tonegod and Lemur and they all have their own issues. Note least of which is text rendering which is kind of a problem for GUIs.
I learnt JME from scratch with only minimal JAVA experience (I have many years development experience in other languages) and wrote a complete 3D Chess app with full multi-player support in 2 months.
I haven’t released it yet because I’m struggling with IOS and have spent 2 more months trying to get it working with limited success. The only reason I haven’t thrown in the towel is because I love JME so much and I’m going to carry on until I get it working.
The support on here is great which I have only started to use and they are very good and prompt which is in my opinion is 90% of the decision making process and makes this system a no-brainer.