Five Seconds of This Evening.

A (very) short segment of timelapse animation as the sun sets in Vancouver. A diptych of two angles shot from the same vantage point.

AS3 SharedObject

Looking into whether the flash player can set and retrieve cookies, I stumbled upon the ‘SharedObject’ – it’s like a cookie, but the flash player stores the information locally (and not in with the browser cookies). If you’re looking to set real cookies, try here. The shared object seems handy for a lot of things though, and it’s really simple to use:

package{

	import flash.net.SharedObject;

	public class shrObj extends Sprite{

		var nom:SharedObject; 

		public function shrObj():void{
			nom = SharedObject.getLocal("datas");
			nom.data.test = "hello world";

			getCookie();
		}

		public function getCookie(){
			trace("sharedObject is: "+nom.data.test);
		}
	}
}

I put together a test to see how it worked:

Enter text, click set, then refresh the page.  The content set into the shared object will persist.  Which could be handy for a bunch of things, no? Seems like storing the current state of a web app is a must in case the user accidently closes the tab or navigates away.

Source files for the example here: sharedObject.

Fractal Explorer

Having just started my first forays into Adobe’s Pixel Bender, I’ve been blown away by the scripts which are currently available for it. As a way to temporarily avoid actually learning the language, I spent quite a bit of time playing around with Subblue’s Fractal Exporter. Results below.

It’s particularly nice to see these shapes morph as the parameters change. Will get them into After Effects soon to render out these shifts.

Jeremy, the Pan Britannica Bunny

Simple character design for a rabbit to be built in 3D and rendered as a shadeless, one-color animation.  It’d be fun to composite this style over video of cityscapes.  Might try out Blender’s new spline IK system to get the joints movements looking really fluid and curvy.

The pan britannica bunny

vector file: rabbit (pdf, 204KB)

Game of life

I made a flash version of Conway’s ‘Game of Life’ a while ago.

In an early version, I failed to clear out the array of cells which needed to change state that round (from dead to alive, or alive to dead).  The result was far more mathematically arbitrary, but made some interesting patterns. In this version, as the dots flicker across the screen I see a robot, an alien, a fat man, a mouse wearing a crown, and a smug tiger.  It’s way more space invaders than the proper version.

Source (including proper, fixed,  game of life) available here: Connways_gameOfLife

Dots.

Just been exploring the excellent toxi color libraries in Processing.

Dots for a slow day.

Dots for a short night.

Catch.

I’ve been working through a few of the examples in Keith Peters’ Advanced Actionscript Animation book recently, which is as well thought through as its prequel.  The section on steering behaviors with 2D vectors is incredibly informative, and got me working on a game of tag between the little arrows.  Using the behaviors I’d just worked through in the book examples, this was surprisingly straightforward.  The players all avoid whoever is ‘it’, while he-who-is-it looks for the nearest player, and moves towards him (or, rather, his destination based on his speed and direction).

Catches are pretty frequent when the arrows bounce off the walls, as the predator tends to be behind the prey as they turn.  Getting rid of the walls shows a more interesting pattern as the prey swarm together:

Things have a tendency to get stuck on the edges still.

May have a go at skinning it with some frantic chickens running aboot after each other.  Then onto the chapters on isometric rendering.  Whoop.

Code available here: catch.

Drops of the world.

Expanding on yesterdays idea, I’ve just put together a processing sketch which takes a ‘panoid’ (extracted from streetview links), and creates a printable net of the location.

To create the net, I modeled the drop in blender, and used its handy ‘unfold’ python script (scripts > mesh > unfold).  This creates an svg of the flattened net, which I imported to illustrator to change line weights and add tabs for gluing everything together.

Streetview tiles

Blender model

Rendered.

Unfolded net.

Added tabs.

The processing sketch takes a ‘panoid’, which can be extracted from the link to a google street view.  It then extracts the image tiles, and squashes them to fit the net, before saving it out as a jpg.  At the moment, it’s making pretty small files (water droplets of the world), which may be modified in future to create higher-res, larger nets.

It’d also be nice to have this accept an address, and do the geocoding to find the nearest streetview point.  Looks like this should be possible with the services google are now offering (getNearestPanoramaLatLng()).  With this in place, it could all be wrapped nicely in a user-friendly website.

There are definitely still some problems with the mapping of images to the net that will be resolved when making a higher resolution version, but it’ll be interesting to see what these look like once printed and assembled.

The processing files can be downloaded here: dotw.

Streetview Baubles / a drop of the world

Just started a processing sketch which pulls images from Google streetview, and maps them to the outside of a Christmas bauble or rain drop.  It seems like there are a few things which could be done with it.  Besides the obvious digital spinning / pin on a map / morphing objects, it might be nice to have it create printable nets of a place which could be printed and folded up into little drops of the world.

New Year’s Photos

New flickr set made up of images from a North West Coast Adventure in America.