Why it'll take time for good Java games to appear

Business and Project Management Discussions
#1

I believe that it’ll take more time than some believe until we see a mass of good quality Java games on the market.

Why?

Most of the good competent game programmers are already totally into C/C++ and because they have the knowledge and power to create games with it, I don’t see many of them switching to Java.

There will be some train jumpers, but they are the exception, not the rule.

I believe most of the Java games will come from programmers who are not that competent game programmers. They are people who are getting into game programming from square one and choose Java as their language. It’ll take them longer to get the hang of creating quality games than an experienced C/C++ game programmer would.

So, it is not just that it’ll take 12-24 months to create a quality title. It’ll take a couple of years for the Java game programmers to become good enough to start a 12-24 month project resulting a quality game.

This is not Java’s fault. The same applies to any language. If you are not a competent game programmer (in any language), you cannot produce a quality title in your first 12-24 month project with any language. A C/C++ programmer who produces a great game, must have years of programming experience doing less spectacular titles before he/she hits the jackpot. Same applies to Java. The C/C++ community can wait for the new people to learn the art as there are lots of competent programmers already on the market. But the Java community doesn’t have a previous generation at all on the market, we are just building the first generation. That takes time.

#2

If you don’t want to be here then go away, frankly I’m tired of hearing this. Fine if you feel you can do it better in C or C++ then do it and stop coming here. This is a Java site not a C programmers “why you should give up on Java site”.

Troll, troll, troll. Well you got a bite from me take it and run.

#3

Hmm - I am going to bite this one too.
This entire post is BS.

I have met many competent programmers developing for Java - and if you just think about it slightly, there is only ONE reason for why Java is a less than stellar platform for games:
Sun Microsystems Inc.

  • there pure and simple.

The Java language in itself is perfectly applicaple, unfortunately the API and the JVM leaves a lot to be desired wrt top of the line Games.

You can’t do Quake and stuff like that with basic Java, even with j3d installed. It simply isn’t targeted to do that.
lwjgl changes this a lot, but it is hard to say untill we do see a Quake version for Java using lwjgl (now that would be nice!!)

In short - the current problem with doing good quality games, is a question about the tools. And this is changing rapidly. In the meantime, use the Java platform for what it CAN do better than C/C++:
Applets
Java Web Start
MIDP
“Simple” games

#4

[quote]If you don’t want to be here then go away, frankly I’m tired of hearing this. Fine if you feel you can do it better in C or C++ then do it and stop coming here. This is a Java site not a C programmers “why you should give up on Java site”.

Troll, troll, troll. Well you got a bite from me take it and run.
[/quote]
It isn’t a troll, but if you see it as such, be my guest. It happens to be a fact that it takes years to make optimal games on any platform whatsoever, pure and simple. I have played with computers nearly 20 years now. I begun toying with Vic-20 and it took about 4-5 years for it’s games to mature to the best that can be done on that platform. I continued with C-64 and ZX48 and those took 4-6 years each to get games that took full advantage of the system. Same happened with Atari ST and Amiga, first games were nowhere near the stuff people created when they got the hang of it - five years down the road. PS2 was already on the horizon, when PS1 games were showing the best you can do on that platform and so on. Same happened with Windows games in general. Compare the games written for DirectX2 to games written for DirectX6 - it takes years for a technology and it’s users’ skills to mature to their fullest. Java is just beginning it’s road in the game business. Yes there are lots of good Java programmers but very few of these have published quality games for Java. The good ones are yet to come - in a couple of years no less.

Actually, I am sorry if you feel that we are already seeing the best that can be done with Java for that hasn’t seemed like much to me. I believe Java is capable of a whole lot more than is being done with it right now. But if you feel that me believing that is a reason for me to stay away from this forum, then maybe I will. But, if I do post messages here in the future, feel free to ignore them so I won’t mess up any more of your days stating my opinions. Have a nice day.

#5

Pyros is just telling the obvious. It’s true.
And Matzon says there won’t be a JQuake. Also true, also obvious. And I also think there isn’t any need to do Quake in Java. Obviously works perfectly with C++. There won’t be any benefit from doing it in Java.

Nobody doubts there are good Java programmers in place - but I agree there are not so many good Java GAME programmers - yet. I cannot see any reason why a good C+±game programmer shouldn’t be a good Java game programmer as well.

What’s missing is a stereotype for typical Java games. I believe these will be online-, low-resource-, play-everywhere, PC/mobile-hybrid games. Highly dynamic with varying content, much like daily soaps.
Not the current Box/CD games.
That would be the place where it would be technically difficult to live with C++, and so easy to be in Java!
LGWLJWL doesn’t change anything there. There also never will be a LWJGWLGQuake.

Just my 2c,

  • J
#6

Actually I have to agree 100% with Pyros, and I don’t think he’s trolling - it looks like a reasonable take on what you might see when you stand back and look at the scene.

I think in the next year or two we’re probably going to see proper stuff coming out written in Java, and possibly even a so called “AAA” title with millions of pounds wasted on FMV and “30 different weapons” etc. etc. etc.

Hopefully LWJGL will be there making it happen too :slight_smile:

Cas :slight_smile:

#7

What is FMV?

Anybody knows wether Roboforge ever sold something noticable?

#8

Full motion video.

I think RoboForge did reasonably well with respect to its development costs from what I remember. However, I didn’t buy it because the demo didn’t work on my entirely normal system :frowning:

Cas :slight_smile:

#9

It worked for me, but IMHO the game completely sucked (ok, not my kind of game anyway). Nice GUI, but endless transition times and ugly 3D graphics. Hm, do I drive off-topic meanwhile?? Better stop…

#10

I think the genre that Java can really shine in will be the slower-paced games. Strategy games like Age of Empires, Hearts of Iron, and that sort. RPGs like Evercamp and Neverwinter Nights (Magicosm comes to mind). Basically anything where “twitch” reflexes aren’t a big part of gameplay will do well in Java.

#11

Sounds like you’ve got reasons for thinking it can’t shine in all areas… but they’ll probably start a flame war :smiley:

Cas :slight_smile:

#12

[quote]There also never will be a LWJGWLGQuake
[/quote]
heh - I much more prefer JQuake :slight_smile:
I see no technical reasons for not doing it… only limited available time as an issue.

[quote] What’s missing is a stereotype for typical Java games. I believe these will be online-, low-resource-, play-everywhere, PC/mobile-hybrid games. Highly dynamic with varying content, much like daily soaps.
[/quote]
Yes, and the reson for this is that the language itself lends towards this, and also because of the rather large API(though cumbersome at times).
BUT if someone were to create an “easy” 2D wrapper on top of lwjgl, I don’t see a reason why lwjgl cannot be used for creating the above mentioned type of games…Something like Cas has done with Shaven Puppy Game Library, though this is much more “game toolkit” oriented than 2D. btw. When I talk 2D games, I am talking custom UI, not AWT components or SWING.

I do agree, however, that creating 3D games is a lot more work than your joe average 2D game.

#13

[quote]I think the genre that Java can really shine in will be the slower-paced games. Strategy games like Age of Empires, Hearts of Iron, and that sort. RPGs like Evercamp and Neverwinter Nights (Magicosm comes to mind). Basically anything where “twitch” reflexes aren’t a big part of gameplay will do well in Java.
[/quote]
At the moment yes. By the time I’ll have the skills to produce a game like Quake, the standard player machina will be about 8000 Ghz Intel Umptium Pro with 2Gb of RAM. Then you can do your JQuake if you want. But it won’t sell, because it will be lightyears from what people are already doing with C/C++ when theyre writing a sequel to the top seller Ultimate Reality Quake Online. ;D

#14

[quote]I do agree, however, that creating 3D games is a lot more work than your joe average 2D game.
[/quote]
Personally, I will stick to writing 2D games for some time now. The biggest problem I find is how to design a 2D game that is absolutely interesting and absorbing. It is easy to make “yet-another-scroller-2D” but how to make a 2D game that people might want to return to - even pay some money to play it… Now that’s the tough part.

#15

That’s the point! Sure, it can be done. But why???
Is there any reason to leave proven technology? Exclude well-known and established programmer resources? Throw away the code base? Whats WRONG with C++ for Quake?

I you want to change things, you need good reasons! Never change a winning team w/o reason!

We could do SQLQuake as well - again, why should we?

My point is that we should not attack main-stream, C+±games on their field. We going to loose. Create our won field, where they cannot exist and make this area successful!

#16

Hmm, well, maybe.

There are some very good arguments for staying away from proven C++ ground, but I still believe a JQuake is a good idea. :stuck_out_tongue:

No matter if it’s not an area where Java really excels or if there are better places to use Java, if someone produced a smooth, platform independant FPS technology demo, it would be good advertising for the platform. And that encourages people to give it a go!

Please do write software that Java really lends itself to, but don’t forget to pimp the language elsewhere as well. If someone really did write SQLQuake, I guarantee it’d get posted to lots of sites and talked about for years! Anyone here heard of text-mode Quake? ;D

#17

I plan to attack them on their home front, by writing games that only take 2 months to develop instead of 6. That means I’m likely to make more money at it :slight_smile:

Cas :slight_smile:

#18

Ok, meanwhile we do a Java-game. Then they have to fight at 2 frontiers. We enclose them and won’t let escape them out of the trap. And then… world domination!

#19

Anyway - first there should be at least ONE game out.

Nothing has been proofed so far - just dreams and vague assumptions. Currently there is just one forum with only 800 members where there’s some believe…
Nothing actually produced, nothing published, nothing sold.

Thats why it’ll take time for good Java games to appear…

:wink:

#20

Hmmm… you think we only have this? Maybe we should think of this as a source of power. What could we do with this is what counts. Maybe we should start one open source project with the people of this board, as a common practice project for us all?