I'm not interested in how many people have no mobile .

How do you verify if it runs if you dont have a testenviroment to test the claim against?
I can also write a program and put "runs on everything" sticker on.
Does not mean it does it.

Again: dont put any effort into making it run on javaME.
Its wasted time. Especially if you want to make some money.

Whatever the framework can do, only look at what makes it run
on Android and/or iOS, ignore the other limits for other platforms.

And especially: only develop programs for platforms that you
have a testenvironment for.
An emulator can not 100% simulate the device.
Its just a good estimation tool for day-to-day development.

I understand what you say. Spending lots of effort matching your game so it works on JavaME. With what I try to do I hope I only need to go through this trouble once, and then be able to create simple (at least 2d) games to compile for all platforms. So far the trouble I encounter is for all mobile platforms, not just JavaME.

What sort of trouble did you run into specifically?

I have a Nokia running JavaME, and it starts to work; it compiles to,and runs on the device. True, didn't solve all problems yet. I hope performance will improve after making certain parts more efficient. But if other, more heavy apps can run well on it, should be possible for me as well. Currently trying to fake 3d (thus far Mosync only supports 2d for non-android and non-iOS platforms) with some extensive calculations. It draws the performance back too much.

The emulator is also a lot faster than my phone, so I regularly check on the real thing. Mosync uses bluetooth connection to compile towards the device, so I only need to hit "the other" compile button, and it uploads to my phone. Plans are at Mosync to include old mophun (former game sdk for JavaME and other platforms). Still a long shot though.

Some troubles with JavaME:

#1 The biggest: If you want to market the game for JavaME
you have to go get it onto the big portals like Jamba / Jamster,
or Mobile Phone providers.

and this causes #2:

#2 They require the game to be supported for like 80% of the
possible devices, and in different languages.

-> That alone is over your head as an Indy.

#3 Lets say you have 90phones (alls kind of Nokias, Samsungs, SE, Motorolas etc) to test the game on.
-> you will have to make like 4 Category Versions of the game
(very weak, weak, less weak, medium)
Or come to the lowest common denominator engine (leaving you with gametypes like Sudoku or Solitair)

#4Graphics:
there are phones with 128x128 pixel displays,
and phones with 600x400 pixel displays.
->And many many inbetween resolutions

If you dont draw vector graphics, you would have to make different
resolution graphics, and still make some dynamic adaptions
in the rendering.

#5 Memory: some phones have a terribly low Heap-Memory size

#6 Floatingpoint Math:
yes, some phones dont have floating point Math!
You have to do everything using integer calculations.
Maybe simulating floatingpoint Math.

#7quiks:
-> There are many many tiny quiks in the API implementations,
For each you have to make a workaround
Example: Nokia has a big problem using diffent Game Canvases.
While SE does not.

Phones have different Key-Mappings. Some differently than
the API descibes.

Some phones cant use more than a 1-bit alphalayer in pngs...

#8Sound:
You need to support MIDI to get sound on the lowgrade phones.
You cant mix soundFX and MIDI on some devices
Some devices only support SoundFX and not MIDI while running the game.

#9 Some phones can only handle like 10 Classes,
you need to contract a lot of code into a few classes.

#Slow: some phones are just slow, but have (had) a big marketshare.
like the Motorola Razor and many of the Nokias.
(SonyEricsson actually made the way best JavaME phones.)

.....

etc..

You could still sell the game yourself online,
but then: how many people are willing to buy it that way
and are experienced enough to install a jar manually.

Its not going to make much money.

JavaME is somehow "before AppStore tech". Served by only
some remaining big players like Gameloft.

Its still a valid Market though, like in South America.
But whoever want to do gaming is using a smartphone now anyhow.

Thanks Damocles, good list of cons. I do use vector art and my games usually consist of lines and text only. About the other bugs, I'm not sure which ones are not fixed by Mosync. But a good way to know (the indy way) is by receiving feedback from users when the game is released already .

The market is a bit of an issue, I have no idea how many people still walk around with JavaME mobiles. I play games on mine, but I'm probably not representative for the rest of the world .

Still putting this through, if only to confirm .

What about BlackBerrys? In most of the US and EMEA their image is not only limited to business usage only, but also gaming and fun and they have a suitable marketshare in your possible target group aswell as a manageable hardware/software landscape.

http://www.berryreview.com/2012/01/17/ri...oney-from-apps/

http://www.berryreview.com/2012/01/27/bl...rowd-in-the-us/

I have an Android (HTC Desire Z), but I'm considering of getting a.. a... *deep sigh* iPhone due to the nature of my work

I have a stupidphone. one of them old samsungs without even a camera. i dont need a smartphone i work on the computer. when i get out, i am glad to get away from all of that...

My personal phone is a brick as well .

Thanks everyone. Screw JavaME.

My Phone is a LG KP500
It uses a proprietary operating system, a development of LG.
i guess its a linux distribution

it´s equiped with a bad camera and there are no third party software applications ("apps", but i hate this word) available. it has a multitouch screen, built in alarm clock, radio, mp3-player and you can wirte SMS and talk.
-> exactly what i need

but i would like to have a second (android)phone for work.