Would you donate your time and talent to create a commercial grade game?

General Discussions
#1

On another thread here on java-gaming.org, it was said that someone or some group of game developers would need to develop a commercial grade game to get a company like Sony to consider Java games for the PS3. But developing such a game is a major financial undertaking. So my question to you guys is that, would you be willing to donate your time and talent to such a game if it could improved Java’s presence in the gaming industry?

I think I would if the direction is right and a good group of artists were also involved. Also it had to be a cool game.

#2

This is how I see it - sorry for being gloomy.
The amount of people here actually capable of doing a commercial quality game is very few - probably around 1% of the userbase in here. They are probably way busy doing their own games or real life and so probably cannot partake in the project. Those that actually have time are probably smiling a bit about the whole project, since they have seen it soo many time - a grand idea, a flurry of people initially working on it - and 1-2 months into the project 2-3 people left and the general mood crappy.
On the other hand you have a lot of other people that really want to participate coz it sounds like fun, however they dont have the skills nor the dedication to keep on the project.

I really don’t know how to solve this common issue - but to start with a core dedicated group - and then enlist the skills needed slowly.

#3

Or to put a brighter perspective on it: there are already many, many people working on the pieces of a game. The LWJGL has a dedicated group behind it which enables games in the first place. Then there’s jME and Xith which make doing 3D stuff pretty easy. Or there’s SPGL for masochists.

As usual what is missing is graphics artists and someone to put it all together and this is where 90% of the effort goes… and almost no-one has time for this. Nearly all such projects are doomed to failure with the odd rare exception (Oddlabs being the only bunch of guys I know who have stuck at a huge project for over 5 years and actually produced the goods)

Cas :slight_smile:

#4

You need a java virtual machine for the ps3 first. Java games would only work well with a virtual machine that could optimize the code to match the performance of a native ps3 game. Without matching the performance criteria this is a waste of time. If it is already difficult to find people that believe in java gaming then finding someone with the tech knowledge required to do a competing java vm for the ps3 would be impossible unless if it was sun doing it.

I sugest you try something easier first. There is very few people that believes enough in java gaming or themselves to participate in an ambitious project like this. So instead you could join one of the already existing projects and try to help. I don’t know what are the top java games being made at the moment just search that game repository that blah^3 made. I supose he keeps it updated with latest news and the status of each project and stuff. But don’t expect too much from just one person working on it. You can also pick a project from the java gaming projects and see if there was any code updates in the last weeks. That usualy helps detecting the dead projects from the non dead ones.

I think this is the best strategy. With very few people and less shiny things to atract new java game coders the situation will only get worse unless Sun actualy moves their asses and spend more money suporting java gaming. Since this is very unlikely, the best strategy is not to further spread the few interested people in, unlikely to succed pojects and focus atention on those people and projects with a bigger chance of succes or that resisted for longer than all the others. Natural selection is a great thing in games in this case i belive it points us the right direction.

#5

I wasn’t trying to recruit anyone nor create a game for the PS3. I’m just interested in the kind of replies that Matzon and princec gave.

This is very true.

This is also very true.

#6

A lot of news is auto-generated now (new games, new releases, new articles, etc), but other than that I add none myself - I haven’t time, because I’m too busy adding features and fixing bugs in the back-end.

I would love for someone to take over submitting and editing news items on the site - everything’s setup for you to do that.

That aside, there are various other things you could do that would help a lot, the most obivous of which is simply to go out and find java games and persuade their authors to mirror them on JGF.

#7

Oh, and writing beginner-level articles would get you a LOT of praise and respect from an awful lot of people. Web stats prove that a good article written not for experts but for beginners typically gets hits more easily than anything else (e.g. think how much effort it takes to do a good game :frowning: )

#8

Im sorry if my answer doesn’t interest you. I was just replying to the question you made about Sony considering java games. To make a short and hopefully less confusing reply, my answer is no. Even if someone would make a top-selling java game, the lack of interest of Sun in building a competitive java vm for PS3 is something that kills the argument completely. It’s not that Sony needs to see a good java game to know that java is good for games. They have enough smart people to run some tests and confirm that. But without Sun anouncing a ps3 java vm and show people some real nice benchmarks we are just trying to find a way how to make omoletes without eggs.

#9

I think that besides good tutorials people need simple and minimalist apis specialy made for indie games that will let them learn from pratice. Apis with around 20-40 classes that don’t use complicated interfaces with excessive customization options and yet complete.

I have been working on a remake of Starflight for the last months in C++. Unfortunatly i could not convince anyone to use java. We are building a gui framework on top of OpenGL and an AI framework on top of the GUI framework. The game engine is being rebuilt to be completely data-driven with all the data in .csv databases and structured text files similar to Half-Life2 or Doom3 configuration files anyone can edit with a text editor.

We don’t use any xml api like apache (however we have considered tinyxml) or a database library or any third party api like ode or cegui because any of these apis are huge. We decided to drop anything except the most minimalist apis that can do the job well.

What we have concluded from this, after Starflight remake having more than three different engines in the past is that it’s completely impossible to do a comercial quality game engine without geting a team of pro. programmers working on it full-time during several months. It’s just impossible and there is no magic trick to overcome this limitation.

We are doing this just for fun. We have had douzens of people coming and going and making many contributions but it’s still an independent non-profit game that cannot be managed like a comercial game. Still is something very fun to do and for people interested in getting a contract as a games programmer this is the way to learn.

I hope people consider doing Java games in their free time not for doing a great games to compete with commercial games but for doing great indie games that are fun to work and can teach a lot about how games work.

#10

No … I’m very much interested in what you have to say. I just wanted you to understand what I was asking. I understand what you are saying about Sun’s lack of interest. I’m just wondering would most people donate their talent and knowledge and time to build a commercial grade game in Java.

“Z” you have to admit yourself, if a team of java developers wrote a commercial level game for the PC, so popular that it had gamers camping in front of Game Stop three days before it came out … Sony and M$ would have to take notice. ;D

#11

well it’s hard to compete with game like Doom,
but if the game is just shareware perhaps we have a chance to compete (e.g simple game like othello or other game that can be downloaded in download.com)
although the main language for game industry is C/C++,
doesn’t stop me using java to create a game with java, because it simply fun :smiley:

btw it need a lot of stamina to create a full game. :slight_smile:

#12

It’s very easy to compete with a game like Doom3 technologically. Doom3 brought very little to the table apart from some slightly flashier internal scene graphics and seeing as that’s nearly completely fillrate limited there’s absolutely nothing stopping someone from doing a comparable game in Java; what it had in abundance though was millions of marketing dollars.

Cas :slight_smile:

#13

More important Sun would notice that and hopefully do something. They are the ones who own jvm tech so even if Sony is interested in having java games it must be Sun to move first. Perhaps a very good PC game would make Sun think about it.

On the Doom3 subject Carmack has spent a lot of time working on an optimized algorithm for rendering shadows or wathever that was. It’s true that his tech appears to be a very simplistic design. I mean linear action games, very simple gameplay with almost no text to read (doom3 changed that for the worst by adding more text ;D ), only a couple of scripted events for using key cards and pulling levers, basic pathfinding and ai that just throws enemies rushing at the player. His games are very well optimized and no one can optimize a game like he does to pull more frames than the competition. But they are still very simple games with very good graphics and a simple and very effective gameplay. That didn’t stop Doom3 from sell.

When Doom3 source becomes open, this woukd be a good oportunity to adapt it to java and see the benchmark difference. If a java version of Doom3 could compare to the c++ version this would sure raise a lot of noise. This would be a bit hard however. I noticed when looking at the Doom3 source there was a lot of optimized assembler code mixed with c++ in some places. The jvm may have an hard time competing at this optimization level.

#14

I don’t think so, and why should it? So you made a commercial level game in Java. That don’t make Java any better than C / C++ or any other language. Sun will not care either unless they can make some money out of it.

Why wait for Doom3. Q3 was just opensourced. But it’s a 6 year old game. People will argue that it’s out of date, and they might be right. If you wait for Doom3, then it will be out of date. Your only option is to steal the source for Q4 and make it faster than the real thing. That might convince some people that Java is good enough.

#15

[quote]Perhaps a very good PC game would make Sun think about it.
[/quote]
About a very good PC game, take a look at java.com in the desktop games section. You will see the following top quality games:

  • Xpand Rally
  • SpecForce

There are other very good PC games there such as Tribal Trouble, Puzzle Pirate, Cas’s games,…

The question is how java has been used for the big tittles Xpand Rally and SpecForce. Were they using Java only in a scripting engine or/and was it use intensively for graphics and sound?

#16

Have these games been release already ?

If they have then theres something very wrong with the way they were advertised. Just look at the gamepsy info about them:

Specforce

http://pc.gamespy.com/pc/chrome-specforce/#news

The game looks and probably plays very good, similar to unreal:

http://media.pc.gamespy.com/media/708/708450/img_2877932.html
http://media.pc.gamespy.com/media/708/708450/img_2877937.html
http://media.pc.gamespy.com/media/708/708450/img_2877939.html

Xpand-rally

http://pc.gamespy.com/pc/xpand-rally/

Also very good

http://media.pc.gamespy.com/media/478/478948/img_2264514.html
http://media.pc.gamespy.com/media/478/478948/img_2264515.html

But do we see the probably here ? It looks like there was a huge problem advertising and selling those games. Check out what gamespy has to say about. If I wanted to play those games most likely i would have to order them because they don’t appear on the shelfs in my country or in other countries of europe and probably the us too. If it was a crappy game from Ms you bet those games would be everywhere with the Ms logo on it and selling tons of copies.

#17

Interesting isn’t it, both of those games run on the “Chrome Engine” - which was the game engine used for the game of the same name (although its been updated plenty). This does of course mean that that Java wasn’t use for the core game, but rather as a plugin game logic language.

Nothing wrong with that, just an observation:

http://www.gamershell.com/news/17104.html (XPand Rally uses the Chrome Engine)
http://www.gameseek.co.uk/productdetail/PChs20zz2jp7ac0f3v/ (The game is actually called Chrome: SpecForce)

Kev

#18

[quote]When Doom3 source becomes open, this woukd be a good oportunity to adapt it to java and see the benchmark difference. If a java version of Doom3 could compare to the c++ version this would sure raise a lot of noise.
[/quote]
Again, why port to java just for the sake of a benchmark?
It has been done with Quake2, and that didn’t raise a lot of noise in the industry… I don’t believe that today java’s performance is the reason it is not used that much for games.
I think what would put java on the map, is when a hugely successful game comes out on the shelves, written in java, made with less man hours, and immediately available for windows, mac and linux.

#19

Why doesn’t sun add a bytecode assembler language feature to java, doesn’t C++ have something like this with

asm {
mov 0, 1 // I don’t know assembly btw
}

?

#20

And if you’d look at C++ dissasembly you’d see an abuse of asm inline by compiler. Not to mention that specialized IDE for assembly is MUCH better than interleaving ASM code with your other code. Actually for majority of work JNI_SomeMethod is better, and safer than could be such assembly

BTW how would you change >MAD r11, r22, r8 > to something that could be used by Intel procesor? (IA32 has just 6 usable general purpose registers)