Any chances for J2ME

Hi guys!

I was playing with java for a long time, then I’ve started to play with J2ME, now I’m a J2ME developer, working with J2ME games. Today I’ve had a nice dinner with a bottle of nice beer, so I started to think on philosophical questions :). I was thinking that mobile java will die as game dev platform as soon as most of the people will have smartphones or communicators. Of course java is safe, but the price for safity - is limited possibilities (a “sandbox”). Java is crossplatform? No way! If you tried “porting” J2ME project to several dozens of devices you know what I mean.

So in one hand we have:

  1. Users, mostly, don’t care about the safity of game content.
  2. They want good quick games.
  3. More and more of them prefer smatrphones.

In other hand:

  1. You have to port and test your game in a specific device to ensure it run properly.
  2. You want to take all that is possible to take from your mobile device.

Does it mean that Symbian games and WM games will take the place of mobile java?

(Tell me this isn’t right! :slight_smile:

There’s also the problem of phone manufacturers just sticking java in their phones without making sure it works well, just to be able to say “we support java!”

It’s slowly getting better, but the problems I’ve had with it was enough for me to swear off j2me for the rest of my life.

another problem is certification of games, of course not all sellers demend that j2me game must be signed but in future probably it will cheange, certification mean that it prove that your j2me game is safe, BUT you have to make cert FOR EVERY PHONE!!! yes read about it, it costs a couple of dolars but it will be not good for industry, becouse only big companys will have money to pay certifications for all phones, the most funny thing is that f: they treat for example sonyericsson k500 and f500 as two diffrent phones :stuck_out_tongue: but they specifications of them are THE SAME, but you have to pay for them all…

On the other hand i think that in the market is place for both: java and symbian games, as long as people buy them . If they have good game they will just but it no matter in what language it was made, Also symbian devices are not all the same as you could think… and also symbian is not easiest to program as java. Of course testing it is true that the game must be tested on as much devices as possible but it is not the foult of java but the very bad imlpementation of phone vendors ,for example in old simens pones drawing images using standard graphics context was very slow, but on the same device drawing images using specific simens java api was really fast… well no pain no game…

and j2me future , think it will not be so bad, for mass market phones java is the best, becouse:

  • it is industry standard
  • it is well known in the market, phones with j2me sell better
  • some people do not know what symbian is, also they do not need such system, they demand easy to use phone with games (and here j2me has its place)
  • there are probably thousends of shops that sell j2me games, and not all sell symbian games, i know the biggest like handago yes, but that smaller no

remember also that j2me industry goes much faster that you could think, three years ago nobody would imagine that we will have 3d in phones and what now, we have m3g or mascot capsule, and what next http://www.jcp.org/en/jsr/detail?id=239 opengl es implementation for java, accelerated games like w900 (which do not have symbian). So the situation is not very bad, also phone manufactures make better j2me implenetataion becouse buyers demand good quality, so new phones are good for programming j2me games, the problem is with that older ones which are the majority in the market, but slowly people will move to new phones (typical lifecycle of a phone is short generally) which for us is better

symbian games of course will take much more role in future becouse they are faster, look better etc etc and also will be safe becouse of certification (but how much wil it cost i do not know), how phone manufactures see the future of mobile gaming? take a look on this gdc session:
http://www.gdctv.net/basic/index.htm#nokia - on the bootom there is session : “The Next Generation of Mobile Gaming - An overview” this material shows what nokia want to make in next two years, especially watch game trailers, they are very intresting, but also this mean that we have to learn symbian, well industry will demand it and users

It has been demonstrated enough times in the past that the most suitable and technically ‘best’ solution doesn’t always prevail.
Often accessability, reliability, scalability and future-proofing play a more important role.

Once handsets become powerful enough to warrant more capable and efficient VM’s,
(there are already VM’s available with JIT capabilities; Hotspot based technologies will no-doubt not be far off)
I don’t believe the speed differential of the language will be a factor.

However - Now that handset manufacturers appear to have completely hopped onto the J2ME bandwagon,
I think it is imperative Sun introduce a strict verification system for KVM implementors.
It isn’t games that need repeated exhaustive testing, and verification - it is the underlying virtual machine.
Get the VM right, and the cost of developing Java software drops massively.

In the company I work, I would estimate around 70% (possibly more!) of the cost of developing a J2ME title is spent on porting and testing.
It makes me so mad that such inefficiency is caused by idioticly short-sighted decisions made by all the ‘expert groups’ associated with the J2ME platform (including Sun!)
If I were responsible for such failure, I would be utterly ashamed.
It is the miracle of modern business that such bad decisions go without consequence for those people responsible.

@OP:
I do not think j2me will die off because not everyone wants a communicator or like.
And that is our target audience.

Yep, same here.
Had some job offers but if I change, only to something un-j2me related.