STARNUT.com

A few weeks ago, I stumbled upon Thibault Imbert’s Banana Slice Component and since I’m always on the lookout for new techniques to improve the performance of my upcoming games (somewhen at the end of march) it immediately caught my attention.

Thibault is currently working on a book on ActionScript development, so, he didn’t have time to publish neither the component nor the class it is based on, so far. So, I started pondering on a concept of my own, experimented a little and finally got totally caught up in the challenge of creating a component that highly improves vector based animations by rasterizing them. So far, I was able to make some tests that affirm my basic concepts but there still is a lot to be done. Plus, I have several approaches in mind that can be derived from this concept. I want to evaluate which of them make most sense in the different use cases before publishing any details.

Thibault, thanks for this fruity inspiration! I’m so excited about this :)

While everybody else seems occupied digesting the fun they had attending Flash in a Can in Amsterdam - why, WHY did I miss it? - I’m trying to mentally ready myself for tomorrow’s Air Camp in Zurich.

Since the workshop is being followed by a 24h contest in building the coolest Air application, we’ve been racking our brains to find ideas for suiting programs we could enter with. While my buddy Sandro still seems unsatisfied with his options, I’ve given up thinking and decided for a small desktop game thing. Don’t wanna go into detail too much, right now, but it seems I’m gonna be using Milgra to implement multiuser interaction.

Milgra is a small lightweight open source java server that I recently stumbled upon. So far it has only reached version 0.5 but that one already comes with a good bunch of needfull features and handy examples on how to use it right away. It seems quite easy to setup and use for as little as I’m gonna be needing. Red5, on the contrary, seems way to heavy for my needs.

Can’t wait to dig into this and develop my first multiplayer game *shiny eyes*

Sometimes it seems unbelievable how time flies. It’s been a week already since we had to call off our second devjam, since Wolli caught the flews. So we’ll have to delay the work on our new game.

But my own game, that I decided to call “BubbleMath”, advanced pretty well during the last weeks and I put some extra attention to the engine that’s behind the game: I added a ScreenController and AbstractScreens, which allow building static menus easily. I hope to be able to post a first screenshot soon. Right now, I’ll have to work on a quick slideshow player for my parents big birthday this weekend. Pew!

In other news… I got an invitation to the Adobe AIRcamp, Switzerland, which takes place in Zurich on 28th and 29th of February, and signed up right away. Besides a six hour AIR workshop the free event includes a 24 hour contest in developing an AIR application. The best program does not only win a Macbook but will also be promoted at this year’s best of swiss web awards in March. Only one thing left to say: Wheeee! :)

So, time to practice my Flex skills, which reminds me: my Liip colleague Sandro and I will be giving speeches about two ActionScript 3 workflow approaches at the Swiss Flash user group meeting in Zurich next Monday. One will consider the basic workflow setup with FDT and one will be about our recently evolved Flex template approach. So if you’re around and interested, sign up today! :)

Initiated by our Christmas Devjam I am currently developing a small game engine in ActionScript 3 in my spare time. The aim is to create a setup that allows quick and easy development of casual games in Flash. The animation system, that I just successfully debugged, is based on the idea of checking the current frame label of the animated MovieClip as described by Jens Franke in this post (make sure to read the comments, too).

This technique completely separates code from design, allowing the designer to delimit animations without writing a single line of code, but by simply placing one label (e.g. “jump”) at the beginning of the animation and one at the end of it ( e.g. “done_jump” ).

So on each frame that starts with “done_”, the engine triggers a method that decides what to do next: play another animation? Loop this animation a certain amount of times? Jump to another animation? Destroy the object? … So, all the programmer has to do is override the function that makes these decisions and implement the coherent animation functions. Easy, isn’t it?

Right now, I’m planning to somewhat opensource the engine, once it’s done - and it doesn’t turn out to be a heap of useless code. As soon as I make progress in this, I’ll let you know.

As announced, Wolli and I sat together on Thursday for our personal Christmas devjam 2007:

At 10am the first coffee was brewed on my tiny stove and we exchanged ideas on the concept and the implementation of our little Christmas game. Then Wolli picked up his artpad pen, I started up my most beloved Eclipse and soon the room was filled with scratching sounds, the clicking and clacking of keys on the keyboard and the smell of RECK’s Moka Yrgacheffé coffee beans (anyone can pronounce that?).

We worked about 11 hours spread over the day interrupted by several coffee breaks and a major present shopping spree. After a lot of drawing, coding and even a broken Eclipse, we finally stopped the session at one o’clock in the morning stuck somewhere about half way through. So, sadly, all we have to show at this point is this nice little screenshot giving an idea of the game graphics:

Although we did not achieve the high aim to develop a complete game that day, we enjoyed the motivation and fun that came from sitting together, tossing ideas around and looking at the first results shortly after. So, we figured that this will have to be repeated!

But what about the game itself? Well, I might find some more time to code, but, still, we probably wouldn’t make it on time until Christmas eve :( But since we want to see this game on screen as much as you do, you needn’t worry: we will finish it sooner or later. Whatever we achieve, I’ll post it right here. Stay tuned!

Christmas time is game dev time - at least that’s going to be this year’s motto since I’m already on vacation. The plan is as follows:

Wolli and I will sit together, think up and implement a christmassy flash game within a day. The aim is to create something challenging and entertaining to shorten the unbearable time till Santa rumbles down your chimney.

Actually, we’ll have to cheat a little: Since this will be our first AS3 game, we’ll reserve some additional time to take care of the very basics such as evaluating a workflow, coding a bunch of classes that consider collision detection and animation treatment, buying a proper amount of coffee and stuff like that.

Now, the dev jam will take place on Thursday. That will be the day when we actually put our heads together and create most of the game code and graphics. However, things are already in motion: a basic concept has been sketched out and today I will start to pull up some classes so we’re ready to go when dev-day comes. Thus, we’ll be able to focus all our Thursday-energy on the game itself and hopefully we’ll have something presentable by the end of the week - watch out for news and updates!

But enough of the talking - let the coding begin! 8)