[quote]I still like the theme idea, maybe a 3rd contest…
[/quote]
I thought the theme was to make a game in 16k? 
That’s right. The theme is “16k”.
Cas 
That’s the whole point. Not to write easy to read code, but to see how you can squeeze an extra byte here or there.
16k is a good theme.
‘Set it up and let it run’ was a Ludumdare gameplay require ment one time. If you’ve not participated in any other game programming contest then I strongly encourage you too. They all have innane ‘themes’, rules, and requirements such as these. IMHO they don’t encourage creativity, they squash it.
[quote]16K > 4K but it is still a mem limit and to improve my game within the rules, I would have to do some hacking and ugglyfying my code.
[/quote]
I really don’t see a problem here. If you don’t like squashing code then go code a game for the next idev games contest or ludumdare or speedhack or whatever.
Looking at the 4K games, I’d say that with 16K it should be possible to even write some half-decent looking code without resorting to nasty hackery and single-letter filenames.
Cas
Na, nothing would get finished that way
Kev
[quote]resorting to nasty hackery and single-letter filenames.
[/quote]
Whats the fun in that?
[quote]Looking at the 4K games, I’d say that with 16K it should be possible to even write some half-decent looking code without resorting to nasty hackery and single-letter filenames.
[/quote]
Yeah, right! I’m already planning on how to represent vertex data in 1.5 bytes :o
Any sign of a start date?
Kev
nope - busy getting 0.96 out the door first. Once thats done, we’ll do some internal organizing, and find out how, where, who and why
if the contest can wait til at least mid-April, I can have the web-based entry submission ready (well, it’s actually done now, just haven’t tested it much) so people won’t have to rely on me checking my emails… heheh :
Mid-april or later is fine with me…
Could someone working on LWJGL get in touch with me please? Emailed Cas some time ago but no response (about 5 days ago?).
adam at grexengine.com
Patience patience! I was just waiting for 0.96a to be released, last night, before I replied with any ideas
Cas
@all
If you want prizes for this competition … PM me or email me with details of where you’re going to promote it :). We’ve got at least one sponsor lined up but - understandably - they’d like to be sure the competition is going to get decent coverage before they commit more than a token prize.
I need a list of news sites, forums, etc where we’re going to promote this. (some blessing from Sun + a note on the front page of JGO would be handy, hint, hint ;))
Thanks
Having never used LWJGL, I wasn’t sure about this comp, but a quick look at lwjgl.org is tempting me; it would be good to learn some openGL. There doesn’t seem to be much in the way of documentation on LWJGL though. Anyone care to comment on whether it’s possible to get by with an OpenGL book & try to find the corresponding calls in LWJGL?
If it were LWJGL150k, the engine from my recent 3D game demo would port over no problem (Still need better gameplay thought ). However 16k is a tenth of that and makes me wonder what would be left after some ruthless pruning. I’d believe it impossible if not for looking at the 4k games and seeing what has been achieved there. Maybe go for 4 or 16 colour images rather than 256 colour. That would save quite a bit. Cut down on the configurable options, remove support for sloping floors & ceilings. Maybe that’d get me down to around 80k. Hum, no where near 16k yet. Can it be done… and more to the point, do I have enough ‘spare’ time?
Alan
You have to forget all your old techniques to make a game in such little space. For 16k its best to think about procedural textures and gameplay that doesn’t require elaborate meshes to be fun. I’m wondering if anyone will use animated meshes in the 16k contest. :o
[quote] author=jbanes When one considers that early video games often had less than 4K of space to work with and were often more fun than today’s games, one has to wonder where the idea that 4K won’t work came from. Have we lost that much of our creativity and “push the envelope” mentality? Does anyone else remember the days when only the best of the best could even hope to code a video game? Not because others were incompetent, but more because you had to know what every byte was doing and how the system used it? There were some serious hardware limitations in those days, and it was up to the software people to make up for it.
[/quote]
Some good comments, here are my thoughts. I just finished rewriting my Double Deck Pinochle 8086 DOS game into Java which took it from an 11k .exe to 110k of classes. Not bad at all, in my opinion.
I originally wrote in Z-80 as tight as possible to run on a 16k Radio Shack TRS-80, and the rewrite in 8086 didn’t add that much size. I didn’t have any idea how it would size out in Java, and speed is my main concern in any event, but it didn’t balloon out that much to my surprise. Both are text interfaces so are directly comparable.
I am especially pleased with the speed. In initial testing I didn’t bother putting pauses between plays of the players thinking it would hit a programming error and crash but instead it played and displayed all four players cards instantaneously. I couldn’t have asked more from assembler.
I wrote to an abstract interface API so I am now replacing the JTextArea with a GUI implementation of the interface. This is my first Java project and I plan on staying in it now for a career. I’ve got a lot more to write yet before looking for work, though.
rd
[quote]Having never used LWJGL, I wasn’t sure about this comp, but a quick look at lwjgl.org is tempting me; it would be good to learn some openGL. There doesn’t seem to be much in the way of documentation on LWJGL though. Anyone care to comment on whether it’s possible to get by with an OpenGL book & try to find the corresponding calls in LWJGL?
If it were LWJGL150k, the engine from my recent 3D game demo would port over no problem (Still need better gameplay thought ). However 16k is a tenth of that and makes me wonder what would be left after some ruthless pruning. I’d believe it impossible if not for looking at the 4k games and seeing what has been achieved there. Maybe go for 4 or 16 colour images rather than 256 colour. That would save quite a bit. Cut down on the configurable options, remove support for sloping floors & ceilings. Maybe that’d get me down to around 80k. Hum, no where near 16k yet. Can it be done… and more to the point, do I have enough ‘spare’ time?
Alan
[/quote]
Don’t forget, you can compress it all into a .jar file. Also jbanes provides a SuperPackMe tool for compressing images.
Well, I downloaded LWJGL and gave it a blast. The keyboard & mouse polling is ludicrously easy to use. It just works. The openGL bit was easy too once I’d figured out those poxy NIO ByteBuffers. The vector library is quite handy but works considerably differently from mine. However I’ve been kicking my heels the last week waiting for the comp to start.
I’m planning a 3D game based on numerically calculated terrain with a model system refactored from my Storm the Castle demo. Probably first person. Ideas range from a Robin Hood type game (but rather a lot of trees to draw = lousy framerate), to something set in the american west. Probably a shoot-em-up, but possibly a buffalo herding type game. People mostly like shooters though, so that’ll probably be it. Maybe a storyline to do with outlaws or bandits (to keep it out of the towns) but I quite fancy the us civil war as there’s lots of good historical stuff on the web (I’m really enjoying just reading it). Anyway, wheels spinning, poised for the off, assuming this happens…it all looks a bit quiet.
Not sure what to do about sound though. Procedural tones and noises are easy enough, but what about music. LWJGL supports mod format, but from recollection those are a combo of sampled sounds and midi-like sequencing & probably too big. I’d like to use MIDI, but it depends on whether the rules state that the judges machines will have soundbanks installed. If not a possibility to to generate procedural tones & write a software sequencer. Would sound pretty awful though; worse than Java MIDI and that has the sound quality of a strangled cat.
Alan