LWJGL16k

Contests 16K LWJGL Competition - 2005
#21

The reason I’m going to enter the 4k comp is that I feel I have the same chance of winning as everyone else. My entry, like all the others, doesn’t rely on fancy graphics or sound, or huge levels full of monsters it concentrates on what everyone wants and thats gameplay.

A 16k LWJGL comp would also be interesting, I agree with previous posts that maybe the memory limit isn’t the interesting part of the comp (I know for 4k I’m spending 90% of my time trying to get it into 4k).

A short timeframe, e.g. 1 month and a theme that the games must include (Kangaroos etc) would be more appealing to me.

As an aside, here in Sydney each year we have an independant film festival, any budget/size is allowed, but it has to be 7 mins in length AND contain a signature item so that the producers of the festival know that it was written from the date of the announcement (this year it was ‘umbrella’, previous years have included ‘rock’ and ‘match’).

I don’t think you should limit the APIs that can be used, since if you release the game to the public you’ll be running on a standard JVM anyway, but I think you can make it a mandatory requirement that you do use LWJGL (obviously!)

Andy.

PS: Do earlier versions of Java produce smaller class files for the same peice of code (I’m not using any 1.5 features)? I’m really looking for ways to shave bytes off my entry.

#22

You know, I was just reading the Readme.txt file I did for Defender 4000, and I realized it sums up my reasons for entering the 4k contest pretty well:

#23

Why have some size limit on the game. Why not just best game. What are we testing, coder skills or game development skills.

Would also bring some attention to lwjgl

#24

I think the 16Kb limit is just right. Think about it like this.

To draw a rectangle in java2d, its a simple one liner:

g.drawRect(0, 0, 10, 10);

to draw a rectangle in OpenGL:

glBegin(GL.GL_QUADS);
glVertex3f(-5, 5, 0);
glVertex3f(5, 5, 0);
glVertex3f(5, -5, 0);
glVertex3f(-5, -5, 0);
glEnd();

So really, it should be 24Kb as opposed to 16 with that logic… But 16 is more of a challenge :slight_smile:

#25

[quote]Why have some size limit on the game. Why not just best game. What are we testing, coder skills or game development skills.
[/quote]
Because without limits in place to level the playing field, no individual has a hope in hell. Anyway, there’s already a category to pick the best game out there - it’s called “the game store shelf”. ;D

#26

This is a very good contest.

There is something similar going on here:

http://theprodukkt.com/

They usualy go with 4k demos for 2d and 64k demos for 3d games and its mostly procedural textures, music, sound and models. And these i believe are for assembler programs that can become quite small.

For Java, however, its probably better multiply that value to take in account the amount of junk the javac compiler adds to class files.

This is a very good idea. Probably the best thing that can be done to mobilize java game programming is to have a competing with a size limit.

#27

cough glRecti()

I think 16kb is just right and hopefully it’ll stop smartarses from doing all that procedural stuff and concentrate on games instead :wink:

Cas :slight_smile:

#28

[quote]cough glRecti()

I think 16kb is just right and hopefully it’ll stop smartarses from doing all that procedural stuff and concentrate on games instead :wink:
[/quote]
Hmm… no wonder you haven’t done any 4K games. glRect() adds another method reference, which spells big losses if you need glVertex(). The solution is to use glVertex() and let the compression program turn it into nothing more than a bit or two of data. :slight_smile:

#29

No need for glVertex. Put all the vertices in a single vertex array.

.kkrieger is an excelent game. Liked it much better than doom3.

16k seems like fair limit. I think there should be a judging bonus if the game is actually 3d. Needs to be an insetive to make it different from the Java2d 4k games.

…and extra bonus if you do the whole game in the main() method 8)

#30

[quote]…and extra bonus if you do the whole game in the main() method 8)
[/quote]
I quite agree. That is, if by “bonus” you mean “slap”…?

Ach, I suppose a 16KB 3D game with readable source code would be a bit much to ask for? :wink:

#31

[quote]I’m very annoyed that 1.5 isn’t on Mac yet, as I don’t have a Java high resolution timer and no LWJGL either :’(.
[/quote]
$500 gets you Tiger (OS X 10.4) Betas + final version, ADC membership for 1 year, developer tutorial DVDs, and Java 5 betas for Mac – right now.

While I admit I haven’t given them my $500, it does seem like a reasonable offer, considering the OS upgrade alone will be about $130.

LWJGL is working fine on the Mac today with Java 1.4. I’ve played the latest from Puppy Games and more and there are no major issues with them.

As for the LWJGL competition, I think it is a great idea… but what I would like to see more of is Java entries in other game programming contests that are NOT directed a Java.

The Mac community has an annual contest for indie gamers and I’m sure that Puppy Games and Odd Labs could score big there.

The idea is to show the larger game programming community that Java really is ready for mainstream games. We aren’t about cute little applets anymore. Once we show off a few star titles in places where they had to compete against the traditional gaming languages the Java gaming position will be that much stronger.

#32

ah yes, but my users won’t be shelling out $500 bucks just for Java alphas to play my game :slight_smile:

When Java 1.5 is released for Mac, will it be available on 10.3 as well? I hope so.

LWJGL works great on Mac, I am using it as my renderer of choice now for Xith3D. What I meant is that one can either use Java 1.5 or LWJGL to get a standard high-res timer. Alas, for the 4k compo I can’t use either. In the end it actually didn’t matter, since I am using asynchronous draw updates anyway (which is cheaper bytewise).

Will.

#33

[quote]When Java 1.5 is released for Mac, will it be available on 10.3 as well? I hope so.
[/quote]
Apple hasn’t said. It may happen, but I wouldn’t bet on it. My vote is that OS X 10.4 is required. Apple makes Java an integrated component of the OS. Sometimes Java problems are solved with OS updates. E.g. I wouldn’t be surprised if they added hooks to the window manager to handle openGL acceleration in AWT better or something like that.

[quote]LWJGL works great on Mac, I am using it as my renderer of choice now for Xith3D. What I meant is that one can either use Java 1.5 or LWJGL to get a standard high-res timer. Alas, for the 4k compo I can’t use either.
[/quote]
Ah, I see.

#34

The 4k compo is over and it was a great success. Is it time to start a 16k compo?

What will be the rules/deadline?

It’s important that to get as meny entries as possible. How about allowing any major java game lib to be used?(Jme, xith, jogl, jaol, lwjgl, java2d, javax.vecmath, ode, etc)

#35

[quote]The 4k compo is over and it was a great success. Is it time to start a 16k compo?

What will be the rules/deadline?

It’s important that to get as meny entries as possible. How about allowing any major java game lib to be used?(Jme, xith, jogl, jaol, lwjgl, java2d, javax.vecmath, ode, etc)
[/quote]
First off… the 4K compo is still being judged, so hold your horses :wink:

I’m planning on hosting the 16k much like I did for the 4k… only not nearly as “last minute”. I’m working up a site right now as we speak, I think I can have it up by the end of March or even earlier than that. I say go ahead and start coding… and the contest should begin in April… and give the coders 3months to do their stuff (which means it would end at the end of June). Any objections to this? How about some rules? I say the 16K should be freestyle like the 4K… so lets not limit it to LWJGL! Someone correct me if I’m wrong… but if the end-user has LWJGL installed on their machine, the game does not need to package its own LWJGL, correct? This can mean a full game in 16K without having to worry about the bytes taken by packaging LWJGL with the game. We need to settle on the target versions of the common libraries (LWJGL, JOGL, GAGETimer, etc.). I think that should be the first order of business. :smiley:

#36

[quote]The 4k compo is over and it was a great success. Is it time to start a 16k compo?

What will be the rules/deadline?
[/quote]
We had a poll that went fishing for the various sensible rules, and I started work on a 64k + libs comp with a 2-week time-limit. Although, given the success of J4k this year, perhaps a 2-month time-limit would be better?

I’ve secured some commercial assets for use in the comp (sound effects etc), but didn’t want to start anything until the dust from j4k had settled. The main thing is to be careful we don’t have lots on the go at once, because that’s likely to spread the attention too thinly :(.

Given that GDC starts next week, I’d rather wait till it’s finished (and see whether GTG announces a new comp) before starting anything else.

#37

The 4k comp this year has been a great competition. It even got me to post an entry ::slight_smile:

Looking at recent posts it seems that a number of peope are wanting to to go a head with a larger scale competition, which is great, I might even take part again.

I’m just a bit concerned that in a month or two we’ll have a number of different competitions that are being run by different people, with slightly differnt constraints at the same time.

Are you guy’s talking to each other about whats to do next? It’s just that otherwise we’ll end up spreading entries between the comps too thinly.

The good thing about he 4k one was that most people in this community gathered round and focused on it (some more than others :wink: ) If we have a number of differnt competitions then it will be to the detriment of all of them.

I hope this makes sense and that I havn’t read the situataion wrong.

Regards,
Dan.

#38

If we could have the LWJGL, DevIL, JOGL, JOAL, JInput, and SDL libs specified as extensions and hosted, then we could mandate that the entry is launched via webstart and references the specified extensions. The size restriction is on any non-library jar files, excluding the JNLP file. Sound ok?

Cas :slight_smile:

#39

:wink:

#40

Sound ok?

Ye.

P200/gzip allowed? Because that would make a big difference especially if you have a buch of classes. Effectively 25-35% more room.