For more information on Spacetime Fusion, check out the project’s website at University of Washington’s CS Departments page here. I’d say something witty, but I’m too stunned.

Shaders, snippets of code for processing pixels and 3D points on your 3D card’s GPU, are cool — that much you may know. You may even know that you can use shaders — designed for 3D applications — to perform powerful video-processing tricks, as well, at high speeds, even on a relatively lowly laptop. How to actually build your own — that may be elusive. So, at long last, Cycling ‘74 has published a great, beginner-friendly (even for non-programmers) tutorial on building your own shaders:

Your First Shader

The author is Andrew Benson, who is my hero as far as coming up with great Jitter examples. (Every time I’m looking for some model for a technique I have in mind, I keep stumbling on his sample patches in the Jitter folder and still more in his By the second page, you’re already building your own custom, glitchy visual filters. Great stuff.

Now, of course, this tutorial isn’t limited to users of Max/MSP/Jitter, but Max is really an ideal environment for testing shaders. (It can even work well as a prototype environment before going elsewhere.) This code will work in Processing, though, and I could see using a combination of those two tools (still working on my own workflow there).

We’ll be practicing our shader chops, because there’s definitely a need for more information like this. This sentence says it all: “If you want to learn more, I highly recommend poking around the “jitter-shaders” folder and grabbing the official GLSL specification(PDF) or the GLSL Orange Book.” Good advice, and the example Jitter shaders included with the program will already do a lot of what you’d like need. But, as fair warning, the Orange Book and other official OpenGL documentation can make your head hurt, fast. (There’s a reason “… for Dummies” isn’t on the end of the title.) I still recommend picking up a copy, but there’s definitely a need for at least intermediate documentation. Being something of a dummy myself, I may be able to help.

Animator and musician Brent Green creates truly unusual animated films and performs live musical soundtracks to them, as he did for fellow Makers at Maker Day during last week’s Maker Faire. Hand-drawn frame by frame, his work seems barely to contain its own movement, described as “cobbled together” yet strangely beautiful. Evocative, simple soundtracks accompany the work, and the result feels like a performance even when recorded. Beginning as a short fiction author, he selects some mysterious subjects, as with his first film, Susa’s red ears:

A little girl with a firetruck in her head on the day the sun explodes saves herself, and very little else.

Watch the films at Nervousfilms.com (check out the Califone collaboration, music fans)
Podcast interview at Directors’ Notes

Want to wear your video on your sleeve? Wearaware Design hacked the cradle for the Archos line of video devices in ways that make the device both more practical and more … wearable. How? By eliminating the cradle, an always-worthy goal:

The cradle itself is not necessary for accessing any of the A/V i/o ports at all: the signals pass right on through. The cradle is bulky, wasted space; its octopus of fixed cabling is long, messy, and weighs more than the cradle itself. All the cradle really offers is IR and Serial protocols and a power tap for the meager circuitry. Plus, with a serial-USB dongle, and perhaps a microcontroller you can still likely add them without the cradle.

With less space, it’s possible to wear around your Archos video player around. And since the Archos has video outs, this could quickly become a fabulous VJ/visualist tool. Media artist/designer (or, we’d say, visualist) Ross Bochnek presented his hacks at a Maker workbench over the weekend at Maker Faire:

Archos PMA/AV Series Cradle Connector Cloning (Detailed instructions at at pointlisse.com)
Wearable Computing Hacks (workbench details at makerfaire.com)

Now, I need to see if I can learn from this lesson to eliminate every single unreliable Sony connector I have on my DV cam.

Mike Creighton has published a preview and overview of his VJ performance application VISP, visual performer. What is interesting about VISP is that it leverages Adobes Apollo framework to allow cross platform applications built in Flash. I wont pretend to have enough Flash-fu to explain it in detail, but fortunately Mike does. Jump over to Today Create to read his thoughts on Apollo, VJ application development and VISP features. Unfortunately VISP is not released to the public just yet, but it looks to be an interesting addition to the ecosystem with a clean and simple interface, cross platform compatibility thanks to Apollo, and 3rd party module development. As a Max/MSP and Jitter .. ahem.. afficiando, I am curious how frameworks like Apollo will stand up to Max and PD for realtime performance.

Jaymis and I are out in Meatspace being visualists, leaving little time to, you know, blog and such. Here’s how to find us:

If you’re in Australia, find a big rockstar with big hair (listen for screaming girls and boys), and Jaymis will surely be nearby.

If you’re in the Bay Area of California, find a convergence of technology gurus, crafters, and DIY counter-cultural geniuses, or small downtown venues cramped with people, and you’ll probably find me. Actually, after thinking I might get away with just doing sound and not image, I’ve managed to find lots of projection possibilities so I’m going to go nuts and do both.

There are some promising-looking visualist goodies here in San Francisco, so I hope to get some images up as soon as I can and will have a full report next week.

Digital process has often been the enemy of craft. Italy, once the land of highly skilled typographers, calligraphers, and music engravers, has given way to static, boxed-up, boring fonts like the rest of the world. It’s only fitting, then, that some lovers of type and digital media would fight back.

Andren writes in to share his Master Degree Thesis in visual communication design at Politecnico di Milano, an open source investigation of type and process called “A Digital Remake.” The results are simply stunning: drawing upon the 1906 work of Edward Johnston on type, he reconstructs in digital form the process of constructing illuminated type. The translation from traditional media to digital manages to be loving without being slavish: this is truly a digital analog to the original process. Generating random particles and connecting them with beautiful curves, the new type evolves organically, unmistakably digital but rooted in the past.

A Digital Remake Project Page, with open-source code, poster guides to the project in PDFs in English and Italian, and Italian research
EXP Research Team blog

I can’t look at these without getting ideas for animated text. It’s surprising to me, in fact, that text hasn’t inspired more visualist exploration and live VJ sets. Previous font coverage here on CDM:

Free, Open Source, Remixable Fonts, and Embedding Fonts in Flash 9 / AS3

The project was coded in Python using NodeBox, a Mac OS X 2D design tool. The syntax and design concepts are actually not unlike Processing, and I could imagine Processing coders being inspired by the results to think of some analog in that tool, with the added option of 3D. Here’s a brief, oversimplified but hopefully vaguely accurate comparison of the two (I’ve only used Processing, not NodeBox):