CoGe 0.85b - Quick Start Guide from luma beamerz on Vimeo.

Apple’s visual goodies and modular patching tool Quartz Composer provide some building blocks for live visuals. But to actually connect these into something you can use live requires some work. Quartonian by Roger Bolton was an early effort to do just that, but while it’s a fascinating application, the interface and underlying patches are more than a little idiosyncratic.

CoGe, an open-source, semi-modular visual app, takes a more conventional approach. The interface focuses on clip triggering and effects and mixing modules. It now even uses the playback module from CDMo contributor vade’s own v002 app. I’m not personally crazy about the interface, but I do like that it’s open source, and if you’re a fan of Quartz Composer it certainly seems worth a look. It’s yet another tool Mac users can add to their visual arsenal.

Quick start video above; after the jump, a video of the new features in the 0.93b build released at the end of last month.

CoGe Site (lots of info inside in the forums)

Full disclosure: I’m enjoying my shiny, new, absurdly cheap PC laptop, so I’ll be busy rocking out on vvvv and some cross-platform / OpenGL-based tools and not getting so into the Mac-only stuff. But then, I also upgraded the MacBook to Leopard… hmmm…

shift v.2, audiovisual installation at Museet for Samtidskunst from hc gilje on Vimeo.

HC Gilje sends along Shift, an “audiovisual landscape that combines multichannel sequencing, audio generated by video, and mapping/masking projection onto physical objects.” In short, big blocks become a sonic, visual sequencer through digital audio and projection. It’s really evocative to me, and part of what we’re talking about as we talk about the potential of mapped projections. (I hope that, for you as for me, it starts to make you think of other possibilities with these kind of media.)

HC’s research is “conversation with spaces,” and that’s fitting — after being caught doing visuals without real sound, or stuck in a “flatland” of our own making that’s in two-dimensional projection, visualists can now enter space.

From his research blog (which has lots of other interesting philosophical reflections, as well):

I decided to give my current series of relief projections a name, shift: moving from one place to another, changing the emphasis, direction or focus of something. It also has a loose relation to the idea of shapeshifting. As mentioned in my previous posts about my relief projection projects, shift combines multichannel sequencing, audio generated from video, with masking/mapping a projection to fit physical objects. This creates a dynamic audiovisual landscape, a spatial light painting. The software to create the installation has developed over almost two years and some workshops, and I have shown documentation of the development, but never exhibited it as a final work. It is only this autumn that I have found the right opportunity to show it in an exhibition. I was invited to participate in the Total Aktion exhibition at Museet for Samtidskunst in Roskilde, Denmark. I had the opportunity to exhibit there in 2005 as part of Get Real, a exhibition with real-time art as the focus (which was also shown at Kiasma in Helsinki, Finland). It also resulted in the book where I wrote the essay “Within the space of a moment”.

Shift became a sort of drone installation, with slow light/colour changes of volume, sometimes cut off by sharp white planes. The video documentation is a cut version showing some of the different scenes. Here is a slide show of still images.

shift v2: relief projection installation

Keep sending this stuff in — your own work or others’ — as we hope to have a round-up soon.

The Projection Tool

Conventional projection hasn’t come very far since the magic lantern days of a century or so ago. Projector + flat, rectangular surface = image. But naturally, with computers, it’s possible to do far more.

The idea is to contextualize a projection in its surroundings, and give the illusion that instead of being simply a rectangular surface (not that there’s anything wrong with that), the image interacts with the reality of a space, objects, and surroundings the way we’d expect. Our own Jaymis Loveday asked last week about the possibilities of mapping — check out the discussion that ensues. I know he’s working away on some projects, but I have to point to the magical, evocative video above of some experiments. ggml writes:

hello. here is a clip with some mapping scenes i have done in recent months using vvvv. they are contextual approaces to improvisation sets, made on the fly, rather than pre-mesured setups. lines are drawn with a 2d drawing-patch, using a mouse pointer, observing the projected image in real space rather than the screen image. other objects are put into proper perspective using the homography node (something like PSP’s distort).

And several of you pointed out that vvvv, the Windows-only, 3D and visualist-savvy generative modular tool (free for non-commercial use) has an edge in this stuff. The reason: the ever-vigilant vvvv community was nice enough to put together an extensive tutorial.

How To Project On Complex Geometry [vvvv wiki]

It’s listed as a work in progress, but like other corners of the rich and wondrous vvvv wiki, there’s quite useful stuff there, made friendly even if you’re new to the topic. Now, the actual topics covered so far are just the basic first steps, but they should get you going. For fancier techniques, I hope this is an area we’ll revisit over the coming months. If you’ve got more resources, send them our way and perhaps we can put together a wiki page of our own.

And yes, I’m now back from my European Road Trip which means you can again look forward to daily posts on the CDMs.

Something’s happening in software. Generative techniques have been around about as long as computers, but from Spore’s game design, soundtrack and creature editors to new music software like Nodal and Noatikl, in 2008 we’re seeing those techniques more accessible than ever. Good news for fans of the demoscene (an underground movement melding coding and art): it’s back with a vengeance, now interconnected with the larger Web and friendlier software-making tools.

It’s only a public alpha, but Toolbox, bargain-priced at 20 Euros (EUR50 when released), suggests what graphics apps might look like with an entirely different metaphor, built on generative lines. The creator describes the tool as a “node-based editor for making digital art,” or a “visual programming language” — the latter something we usually associate exclusively with patching tools like Max/MSP/Jitter and Quartz Composer. The difference here is, whereas those are open-ended software sketchpads, Toolbox is a single-window editor and integrated environment for making visuals, more along the lines of a Photoshop or Illustrator. I’m not suggesting you’ll toss your Creative Suite 3 license out the window, but what this does mean is you could generate an asset from start to finish in this tool — and, perhaps, take it out to another program.

Toolbox App Product Page, Download
Video Album on Vimeo
Flickr Set

The whole project is the work of one developer, Simon Strandgaard. (Remember, too, Quartz Composer began as the project of one Pierre Oliver-Latour.)

What does all this mean? It means you can make UI elements quickly, or destroy existing graphics, and play with vectors in a fluid, magically generative way. The alpha state can make it slow and unpredictable to work with, but it’s already capable of some fun stuff. Here’s a look at vector filtering:

Random vector filter experiments from Simon Strandgaard on Vimeo.

Visual patchers love the fact that they can create sophisticated stuff just using on-screen patch cords. But when you want to go beyond the capabilities of the default set of objects — and for solving certain problems more elegantly — code is the way to go. Code your own objects, and you get the best of both worlds. So the news that vvvv is adding a facility for easily creating your own objects (or "nodes" as they’re called in vvvv-speak) is good news, indeed. (Jitter and Pd have similar extensibility, though having looked at what’s necessary to write a Jitter object, you may or may not want to … go there. This does look quite a bit easier, for simple tasks, at least.)

Phl shares some additional resources in comments on our previous story, so I want to make sure you don’t miss these.

sourceforge site with the first plugins (go to code -> svn browse)

VVVV Plugin HowTo

Plugin Forum for more questions

Note that you can use any language you like. I do like the open source SharpDevelop they mention here. For any of you who think Windows doesn’t have a fledgling open source community of its own, think again — perhaps because of the glut of working Windows-based programmers, there’s actually quite a lot going, arguably more than on Mac. And C# is a nice language for this kind of task.

Seriously, let us know if you build anything; we’d love to see it!