That stupid headache is back. Yesterday it felt like it was coming from the base of my skull/top of my neck, but today it’s localized near the top of my head. Kind of like if you drew a four-inch segment from the front of the eyeball towards the back of the head, then drew a three-inch segment going straight up from that — right there. Ibuprofen isn’t helping, but running really cold water over my scalp seems to provide a little relief.

Harry Potter and the Order of the Phoenix comes out tomorrow. I’m planning to catch the 5:30 showing over at the Village Centre theater.

Playing Cooking Mama on the DS, I’ve come to the realization that the average Japanese homebody must have a really warped perception of western foods. Like, when making spaghetti they use ketchup as the sauce. And they press their steaks when they pan-fry them. So gross.

I started playing with Apophysis this afternoon and produced this:

The program is kind of nifty to play around with for a few minutes, but the controls are pretty trial-and-error. You can’t give a whole lot of direction to the forms you generate. But thinking about those limitations reminded me of ContextFree, which I hadn’t messed with in at least a year. They’ve added some cool features — a new primitive shape (triangle), Z-ordering, alpha transparency, tiling, include directives…

So, of course, I DL’d the new version and started writing “design grammars.” I began with a simple grammar for making curvy branches (something I’d done before) and played with the color adjustments a little, which eventually produced this image (not very impressive by CF standards):

Then I got a little more adventuresome and wrote this CFDG code, which produces a sloppy jagged line that changes hue every time it zigs. (See the first pic below.) I thought it was pretty neat, so I made it longer by decreasing the “stop” rule weight and added some recursion so it’d draw roughly a thousand of these “graffiti lines” (well, for each cycle through the loop, there’s about a one in one thousand chance the program will end; I think that means the average number of loops will be a thousand, right?). The two pics on the right are the results of those changes (they look different because they use different random seeds). It takes a couple minutes to render usually, but I think it looks pretty cool.

You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.