It never ceases to amaze me how much functionality you can pack into a tiny file with Flash. I started out writing a test for generating random asteroids, and it blossomed into a full-fledged game. Doom III it is not, but considering it’s less than 1kb in size (1020 bytes [UPDATE 04/24] now down to 1007 bytes), I think it’s pretty impressive, having time, points, levels, bonuses and penalties.

Game play is simple: Shoot (click) all of the asteroids before time runs out to progress to the next level. You’ll want to be fast and accurate – you get time bonuses at the end of each level, and are penalized for missed shots. Have fun, and post your best score / level in the comments below.

Continue reading →

About a year ago I released this demo of grid based proximity management, which was loosely based on some work I had done in Flash 5 (at that time I was calling it zone interaction). Briefly, it allows you to determine which sprites on a 2d plane are near to one another. It is very efficient, and can deal with literally hundreds of sprites (the main limitation is usually the sprites motion and graphics, not the proximity management). For instance, with 350 sprites grid based proximity uses about 700 operations whereas normal iterative proximity testing would use about 61000 operations. For a description of the concept, see my original post on the subject.

Continue reading →

As promised, here is the AS2 source code for the Varicose-G experiment that I posted a couple of weeks ago. I stripped out some of the ancillary functionality like the grid and labels to make it easier to understand.

If this helps you out, or you use it somewhere, please let me know in the comments, or by email. It makes releasing this kind of stuff worthwhile.

Click here to download the Flash MX 2004 source files.

While I’m in Sydney, Australlia for MXDU (MX Down Under), I will be running a one day workshop based on my very well-received (if I do say so myself) 3 day Flash MX 2004 Enterprise Development workshop. It is being organized by the same amazing folk who organize the conference, and will be running on February 16th (day “0” of the conference).

If you are looking to boost your development skills to the next level, or polish the OO skills you already have, this is a great start. The original workshop received rave reviews (see some of them below) – the most recent run of it (in London, UK) received an average rating of 9.5 out of 10 from the attendees. It has a strong focus on real-world applications, and on delivering a maximum of retainable, applicable knowledge in a minimum of time (I pride myself on pushing attendees to just under the “brain exploding” point).

The full workshop description and reviews are below. For more info, or to sign up, go to http://mxdu.com/go/workshops/enterprise-development.

Continue reading →

Blog comments are back on. Sorry for the outage. If there is much trouble with blog spam, we’re going to switch to an approval system.

gModeler has been updated to v0.6.1. This version includes minor fixes including:

• AS2 stub code generation now includes the return type of methods properly.
• Online help works again.
• Feedback system works again (or will shortly).
• Version number is now display on the right side of the toolbar

gModeler 2 is in progress as well, but with the amount of work we have on the go right now, it’s hard to say how long it will be before it’s complete.

I was recently browsing through the SWFs and FLAs I’ve collected over the last few years because I thought they were interesting or unique. During my reminiscing, I stumbled on this vein-1 experiment by Mike Johnson from May 2002.

It caught my eye, and I decided to rebuild it in OOP, and spice it up a bit. I wanted to make it look like an old anatomy textbook come to life. My fiance thinks it’s creepy, so I must have done ok – judge for yourself though, the result is below (warning: it’s fairly CPU intensive, and requires FP7).

I’ll be releasing the code to this once I’ve cleaned it up a bit. I’ll also be talking a bit about how it came together at my upcoming conference sessions.

Also, to give credit where due, the original experiment by Mike was part of a 25 lines experiment. I’ve nearly doubled that for the Vein itself, and then nearly doubled it again to add camera motion, grids, and labels.

UPDATE: The source code for this experiment is available here.