[quote]Maybe we should start one open source project with the people of this board, as a common practice project for us all?
[/quote]
Check: http://www.josrts.org/
A lot of ppl there also frequent JGO. It’s still pretty early in the project but there is still plenty of chances to make suggestions on how you think the project should evolve.
I like to think that we’re very much in the “little games and tool-building” stage of development. People are creating little games entirely in Java, Java is being used as a supplimentary technology in various places, and various APIs are being developed to make the job of games creation easier.
As Java3D improves and matures I expect to see a lot more done with it. LWJGL is advancing steadily towards version 1.0. Java2D is improving in leaps and bounds. As these technologies get better, more people will start to use them.
I am, as always, optimistic about the future. :
Oh thanks! I wasn’t aware of that project at all. I’ll have to look into it!
But, that project seems quite serious, too much bureaucrazy and too much complicated high-tech. I was thinking more about something for the less experienced - kind of a practise project where people would get to know the basics of team work and simple game making tech.
You all know what typically happens to opensource game projects?
To create a AAA game today, you need 10 people working 10h/day for 2 years. Java or not.
Does anybody here believe that a josrts or similar things will receive any recognition in todays RTS market? Hands up who believes that!
As cfmdobbie is the optimist, I’m his counterpart.
If josrts.org is serious, I’d hate to see something that’s wacky zany fun! 
I think the impression of bureaucrazy comes from our committment to giving everyone a voice in the major decisions. Many projects I’ve looked at have one or two people who have come up with most of the design document, and consequently I think that there is less of a connection between the other developers who join on because the project really ‘belongs’ to the one or two main coders. Then those coders generally end up doing the bulk of the work as the rest of the team comes and goes over time. Hopefully, because everyone has had a part in our primary decisions, more developers will form a committment to the project.
Once we have a design document and split the work into modules, I think that the rate of progress will be a lot faster.
I think our project is probably not the most likely contender out there, and as Herkules mentioned, it takes a ton of work to produce a working title. On the other hand, 60% of what everyone talks about in this forum is how hesistant commercial games companies are to embrace the new technology. Considering my limited background in Java it may seem presumptious of me to say this, but I think it will take an open-source project like this to get something going.
Even if we can produce an engine that is somewhere along the lines of Warcraft II but was polished and installed/ran easily on Mac, Linux, and Windows, that might be enough of a statement so that Java advocates could link something other than Sun’s rhetoric when defending Java’s potential for game development.
On the other hand, the least anyone who joins in the project can do is learn more about Java game development – even if we crash and burn. Pyros – many of the people who have signed up for the forums at josrts.org are probably in your or my position… perhaps some experience with coding in Java, but no real background in game development or game development with this language. We have some talented, experienced people, which is execellent because some guidance is going to make producing a full game possible, but everyone seems pretty willing to share and learn. The attitude of the group as a whole comes across as really positive to me.
[quote]I think the impression of bureaucrazy comes from our committment to giving everyone a voice in the major decisions.
[/quote]
I never meant it as a negative. I’m a firm believer that JOSRTS will become something of value, if not a full blown game. But it is such a large project that it may be too much for total newbies, either really over their heads, or just too frightening (ya know, “I’m afraid I’ll make a fool of myself” and so on). I was just stating that it would be nice to have a lot smaller and less frightening project open for those who have almost no experience about open-source team work and/or game development. Sort of “lets make a galaxian-clone as a team-work exercise - newbies welcome” kind of thing.
After that, those people could join JOSRTS with something valuable to contribute, and less fear.
As usual a fiery and interesting topic.
By definition Java is different from C/C++, therefore by definition there are different (most applicable) uses for either language.
For now, a JQuake in Java might not be one of them. Given that future hardware performance will improve, the JVM and java libraries become more sophisticated and perform better and also the 2 important features of java being suitable for distributed systems and quick development, I think Java will steadily become the language of choice to develop games in.
I see a future where c/c++ will stilly be used for but less and less so, as java is used to develop and manage the core gaming application.
If you want something like JQuake, how about taking an existing open source project and convert it to Java:
http://wouter.fov120.com/cube/index.php4
From my understanding Cube uses OpenGL so converting it using LWJPL (or whatever you call it) or GL4Java should make it pretty quick work.
[quote]Sounds like you’ve got reasons for thinking it can’t shine in all areas… but they’ll probably start a flame war 
[/quote]
Mainly just because “few” people at this time really know how to tweak java to get the best out of it for games. You can much more easily get “acceptable” performance in the slower paced games, than you can in a really twitchy FPS or simulator for example.
As the skillsets progress, so will the complexity
I’d really love to do a “Homeworld” type game, as I feel it would be pretty darn simple as far as the 3D programming goes (in J3D), and you could spend most of your time on the game design and gameplay… Time as always is the enemy
Pyros: [quote] It happens to be a fact that it takes years to make optimal games on any platform whatsoever,
[/quote]
Well, does a game need to take full advantage of the system to be good (fun)? I don’t think thats the case. Look at the nintendo 8 bit. The level of graphics between super mario and super mario 3 is great. However they are both very good games. I agree that it takes time to figure out how to program all the cool looking tricks and graphics, but this is not the major part of the game. Your game can have all the graphic effects that you want but still be a load of crap. Look at Counter-Strike - there are thousands of servers running that thing and by todays standards - the engine is out-of-date (I know they continually work on it but they are still bottle necked from really improving how things look).
Jumping to another topic …
leknor: [quote] To create a AAA game today, you need 10 people working 10h/day for 2 years. Java or not.
[/quote]
Many people said that about operating systems. There is no way any group of people could write an operating system firstly when the people are not in the same place and secondly not working on it full time. But today we have Linux.
So anything is possible… even this AAA game title if those who are involved want it enough 8)
every post made valid point here - i personally think there is no universal answer to this question - to add my angle on it I would rephrased question and ommit “Java” word from it …then I would ask : why is it we play games at all? … most likely the answer would be we want to be amused … so essence of the game is not about what technology was used to get there - but what story it has to tell … it is lack of general ideas WHAT to do(write) which “slows down” production …among those 800-1000 members here there’s got to be reasonable percentage of highly skilled 3d programmers who can do miracles even with ‘inperfect’ tools(API) …but to get going one has to get motivated - one has to know this is gonna be killer app … and to make killer game is not about technical execution (as I thought for a looong time) - it is about great story(idea) … once you have that one you are in bussiness … role playing - strategy …it doesnt matter - just come out with something which will draw gamers to it every night and that is really hard to come up with … and to end this blah blah …I think Java is already there to be used for that killer game …now just that idea is missing …
later.
I’ve been observing this thread for a while and I’ll throw in my comments now. I think that its absolutely try that in truth there are few game developers who are trying to write commercial quality games in Java. That much is just obvious. The few people here working on games make up the fringe pioneers exploring new territory. Whether or not there ever really is commercial gaming in Java, in reality, depends a lot more on us than it does on Sun. In order for Java gaming to become acceptable there must first be games that are in/use Java - there are a few scattered instances where this is the case, but overall Java gaming is still a dream of a select few.
I will say however, that I don’t necessarily agree that it will be 3-5 years before we see games in Java that match with their C/C++ bretheren. I think it will happen in the next 12-24 months IF the community here at java-gaming.org continues to support each other and share their accomplishments. There are, lurking in the darkness and shadow, businesses who see the potential of gaming with Java technology and who see where they can make money on it. I work with one of these companies on a semi-regular basis. As they become comfortable that they can make money on what the Java community can produce and make money on it, they will push forward the ideal of games in Java and throw money behind it. But at the moment, the business position is still a bit hazy (and that’s probably the most important piece).
The way I see it - Java gaming isn’t really a technical issue… its a business issue. Even taking the Java3D framework a Java developer can compete on equal footing with much of the stuff that’s on the shelves today - and that really should be the realistic goal of developers coding in Java or any other language. Build what you need to hit those demographics and make some money. When you start making money with games made with Java, the rest will follow - and quickly.
Personally I don’t blame Sun for the situation with java gaming - I think they are WAY out of their league. They have yet to demonstrate that they really understand the problem or understand the benefit of providing a solution. And that’s a shame, but since that’s not really a source of revenue for them I can understand why its not higher up on their list of priorities… gotta keep the lights on. I do fault them for absurdly incompetent marketing practices however, but that’s another story.
Herc, you should see some of the submissions at the GDC Independent Game Developers competition. Some of these games are unbelievable and done by a handful (most cases 1 - 3 people) over the course of several months to 1 year. A lot can be done by relative “amatures” in the game development space.
And, quite honestly, I do believe that an opensource
game can receive recognition in the RTS (or any game for that matter) market. There are PLENTY of big name games that had millions of dollars and years of people working on it which bombed and just plain sucked. (Poor poor Soverign. DAMN I wanted that game to come out :()
BTW, we are all focused on the hardcore performance side of Java games, just rember that there are plenty of less demanding games that have become extremely popular. I mean, does anyone here doubt that The Sims could be ported to Java and keep the exact same gameplay/performance?
The opensource thing is great and I am in support of this. In fact, I am going to be running this thread up inside of Sun so the bigwigs can see what is happening down here 
-Chris
You could point them at a couple of other threads as well…
Cas
I easily believe that, no doubt, some amateurs create incredible things. VERY seldom a commercial game gets out of it!
But it hasn’t happen so far? At least I don’t know any.
No, in fact I think that The Sims Online would be the ideal candidate for Java!
But it was made the traditional way and for sure it isn’t an OS project.
If there were a way to form team creating something similar to The Sims, maybe extended to the area of mobile gaming - this could be a breakthrough! I’m afraid this needs to be something with money behind.
Hm, and maybe we should distinguish opensource from non-commercial/volunteer? Maybe a fully commercially produced Java game should be made opensource to attract people with technical background that may help keeping a game alive? Java games are somehow opensource anyway, so it might make sense to turn this into an advantage?
On the surface Herk this sounds like a reasonably sensible plan with Java as it is now - but from a business point of view it’s flawed, considering the PC market.
To be successful in business you need a target market that is large enough to demand enough of your products that it can support the business and allow it to reinvest and develop more products, expand the market, or gain a larger market share.
When you restrict yourself to only one small niche market because of your choice of tools to build your products you will find that a) the market isn’t large enough or b) the market will initially seem large but will not expand or c) someone else has the same idea and your market share shrinks. c) is a big problem if the market is only just large enough.
By using OpenGL and requiring about 500Mhz (it seems) I’m already taking a big risk because at least 50% of all the buying public won’t have a computer that runs my game, having ancient computers as they do. I’m offsetting that risk by natively compiling my code which drastically increases runtime performance and critically reduces download size and installation trouble.
But if I wrote a game which required me to use Java in its unenlightened form I’d be looking at a market so small I would be very unlikely to see any return on investment. Not only would 50% of the computers available be too underpowered to run Java at any decent lick, having 32Mb or less RAM and a 100-200 MHz processor, but of the remaining 50%, how many of them have Java installed? How many of them can be bothered to download it when it’s twice the size of your game’s demo? How many of them will think the decidedly old-tech jerky game is worth parting cash for after the hassle installing it? Every single obstacle is cumulative, not additive.
Cas
Well Cas, you can always include the JRE with your game. In fact - thats the best solution for todays Java 3D games. The user won’t have to go out and download anything. It will be apart of the game. You can use the install anywhere tool to help with the install stuff too.
So I would say your main problem is getting computers powerfull enough to run your game. Every company deals with this issue no matter what the programming language is. Its the company who ways out the performance vs graphics. How many polygons do you want to make your models, how many objects are in one area, ect are just a fraction of the questions that need to be answered when dealing with this.
And, actually, games that use Java in one way or another DO ship the JRE with the game. Look at Law and Order. 100% Java & Java3D, shipps with the JRE on the disk. Shoot, even the message windows display the Java logo in the corner. So distribution of the JRE with “traditional” methods is not a problem.
The downloading of the JRE is not completely insurmountable either. The problem with large downloads is a perception issue. Example:
When someine DL’s flash, they actually only download the loader. To an average computer novice user the DL is quick! Boy that was fast. AH! But then the install takes a bit longer. Well, that’s ok because I’m not doing the download this neat little program is (webstart anyone?). The perception is changed from this big hulking DL (where did I put it? What’s it’s name? which version is it?) to a loader that looks nice that takes care of that yucky compooter stuff.
We don’t care so much about hardcore gamers. We know what to download, run, and why. Change their perception (read: experience), and the big hulking Java installation thing kind of blurs away…
-SG
Unfortunately I’m not targeting hardcore gamers either, nor even traditional retail - I’m writing a game for the new, upcoming massive industry that is the indie game market where demos must be under 5Mb and machine specs are relatively poor. It’s going to be big.
Webstart does not appear to be installed on anyone’s computers yet…
And you should check your research on user patience and experience waiting for things to download:
-
Users typically get bored if they have to wait too long and just go somewhere else.
-
Users get especially irritated when they download in two stages. My target market want a single file that they can stash and play later when they’re not online.
I think you might want to pop down to dexterity.com’s forums and ask some pertinent questions in there. While you’re there take a look at the quality of the games - all easily doable in pure Java2D I reckon. But then take a look at the required machine specs 
Cas
Hmmm…
- WildTangent
- Flash
- Kazaa
- Morpheus
- AOL
- Autopatching of games
- Arnarchy OnLine (ok,ok, that one was TERRIBLE - even if you have DSL)
Cas, All of these services above have really long download /autopatching times, but it has not stopped them from becoming successful. Why? Because the value of the content exceeds the pain of the install. Simple as that.
And remember, it’s not like the user has to wait for a 10M+ download every time they will play a Java game, just the first time. I really believe that this is becoming less and less a problem as users tolerance for bigger downloads increases.
-SG