It's nearly time again, the Java 4K 2009 Competition is just round the corner (it starts in Decemeber). In this epic battle the residents of the Java game development community come together to test the mettle by creating gaming masterpieces in JARs that come in at or less than 4,096 bytes!

The contest runs for three months and is judged by a distinguished set of judges (normally also from the community). In past years there have been awards for different facets of the games (technical, visual etc). It's not quite clear whats happening this year yet but I'm sure it'll be the normal crop of varying quality with some incredible gems.

And yes... I'm hoping to actually enter the year. Woo Hoo!

I recently got asked why I always ask for a test case when I'm presented with a bug on the Slick forums - is it a diversionary tactic or is there an actual reason? Well, first, yes, diversion is key. If someone really can't be bothered to create a test case it's not that important to them - I don't get much time so need to prioritize :)

More importantly, it's how I work when trying to solve bugs. I tend to try and boil the faulty use case down into a single class that demonstrates the issue. Then start tweaking the implementation and adding debug (or debugging) to determine whats going on.

So.. yes, it really is useful to have a test case. However, the more complicated the test case the less useful it becomes.

And.. so far... it's working out pretty well. Most people are happy to give a test case and it nearly always makes it pretty quick to solve the problem. Thanks all! :)

Got this at work from a friend, looks pretty interesting. If you're in the South Wales area and interested in technology at all I'd pop along:

If you had five minutes on stage what would you say? What if you only had 20 slides and they rotated automatically after 15 seconds? Around the world, web enthusiasts, creatives and entrepreneurs have been putting together Ignite nights in answer to this question and now Cardiff Web Scene and Nocci are bringing Ignite to the UK. Join us at Sodabar at 6.30pm on Wednesday 3 December. Entry is FREE and you don't need a ticket. Thanks to Skillset Cymru, there will also be some free nibbles so why not come along? If you want to take part simply send an email to hello@cardiffwebscene.com letting us know what you want to talk about and we'll get back to you if your submission is successful. For more information about the event please visit www.cardiffwebscene.com or www.noccinet.com. The Ignite community is sposored by O'Reilly.

This article (and it's subsequent parts) make interesting reading no matter on which side of the fence you stand. It fits in with some of the stuff I'm experimenting with at the moment - flash/flex development.

It's all pretty exciting to me at least. The last time I posted about Flash vs Java my attitude was pretty much if you want to write simple web games with loads of accessibility use the right tool (i.e. flash) rather than spending a lot of extra time trying to make things work.

That was, of course, based on just thought. I had a very limited experience of flash development. I've got a little bit more (still not much) but I have at least released stuff to the community and tried to sell stuff on. So far, it's much easier experience.

I posted this over at JGO and it's stimulated some pretty interesting conversation. The most poignant comment to me was this:

Pick your goals and market, then pick a technology that fits.

Right now, it's all just still an interesting experiment.

Classes that are called AbstractExampleThing that arn't actually abstract.

Uck.

Couple of new flash developments, check them out:

Patch Match - Collapse style game

Tricolore - Pattern finding game.

Instructions for both in game, one evening development for each. This isn't too bad going even if the games are essentially dross. :)

This was an idea that sounded ok in my head, looked alright in paper, but ends up just a bit wierd. Take the old classic columns, make it possible to shoot in columns for each of the four sides and then... um... when search for a way to make it interesting ...

Try it here.

EDIT: Fixed the link.

One of the projects I'm still interested in is the 2D game library, Slick. It's been going a few years now and I think has started to become a really useful and quick way of getting moving with 2D games. Distribution issues aside the use of GL stops you worrying so much about performance and the API is pretty powerful making it fun to work with.

That said, when I read reviews and comments on the library, one of the most common things said is a lack of documentation. Now, I'm confused, what exactly are people looking for? Slick's documentation is pretty much together though sometimes runs behind the development curve a little. First theres the JavaDoc it's complete and rich with information. Next there's the Wiki which serves as the user guide to Slick. It's under constant update by the wonderful community members and while it's never going to be convering the features that were added yeseterday it's for the most part written really well.

Then we move on to code examples. First there's a wealth of test cases covering each feature in a very simple manner. As you can imagine these server a double purpose in that a run a subset before releasing a version. Next, theres examples that get posted to the forums every few months. There's the "scroller" and more recently the platformer. These get posted to the forums, to the Slick Blog and often here on C&C too.

Finally, the forums provide a great resource of information about different approaches and their results.

It seems to me there's a fair bit of documentation/help to go with Slick - much more than Phys2D or Thingle etc. It seems like people want tutorials on how to use Webstart, or how to use Java, or how to use Eclipse to be part of the documentation of Slick. That seems totally out of context to me?

So.. if you've commented before that theres not enough documentation and it's slowed you down, or worse made you give up - what exactly was it you were looking for and where have you seen examples of that type of documentation before. It'd really help because I'm totally at a loss how to make the situtation (if there is one) better?

I've spent the last few weeks experimenting with Flash, or rather Flex and AS3 for 2D game development. I've done quite a bit of 2D development in C, C++ and Java in the past and so it's interesting to see how AS3 stacked up for me.

In the other languages I've never had to pay for tools, so in this case I stuck to the free tools for Flex and AS3. This means I used FlashDevelop and the free Flex SDK.

I've knocked out two games so far, Star Dodger and BubblePop. Productivity wise it's taken me 2 weeks evenings and weekends to get comfortable and up to speed with the tools. There's alot of tutorial help out there, unfortunately unlike my previous experience, quite a lot of the tutorial code available shows either things that don't work or really bad coding practice - which is shame because the noise hides the really good ones.

Anyway, to the quick run down.

Pros

• Quick to pick up and get productive in
• Free tools available
• Massive community
• Easy Distribution
• Large SDK (see below though)

• Counter intutive rendering approach if you've used other langauges
• AS3 is butt ugly - seems to be trying to be Java but failing
• It's slow, very slow, not for rendering (if you're using it the right way) but for data manipulation
• Lots of details that you just have to know
• The SDK contains alot of duplication and misdirections. The API isn't well thought out in most cases and leaves you hanging for that last critical feature.

I quite like Flash in the same way that you find a lame puppy more cute that fully grown Alsation. However, the SDK and performance make it hard to work with - almost like fighting with Java 1.1 to squeeze that last little bit out of it.

Still, it's wonderful to have another tool to utilize in the varied world of software development :)

I've ported my old bubble popping game over to flash. It's got a luke warm response from the flash communities, 50/50 - some love some hate. Maybe thats a good thing, but it's still fantastic to recieve feedback from so many random people so quickly. You can try the flash version here:

Play it in Flash

There are some issues with scaling images in the approach to bitmap rendering I'm using in AS3 at the moment. Scaling is pretty slow and poor quality. A bit of research provided me with a bunch of workarounds, none of which actually worked.

Next one is a new game idea, should be demoable in the next couple of days. The idea doesn't seem to be panning out though, 4 way columns should work right? :)