Psychosynth, and 3D in general...

I noted with interest when Matrix put up a little blurb about Psychosynth the other day, which is an open-source no-table-or-Bjork-required clone of the Reactable stuff. Now, right from the outset let’s get one thing clear: my patience for that whole “breaking down the barriers between the artist and his shit” routine is essentially nonexistent. My interest in these sorts of tools is limited to workflow questions entirely. Personally, I think the audio results of the Psychosynth specifically are underwhelming at minimum, and for the most part downright bad.

Now, with that massive caveat out of the way, here’s why I bring this whole thing up. My job with Audio Damage, aside from providing pithy responses to emailed questions (“why doesn’t Dubstation 1.5 work with Cubase VST in Windows 98?”) is generally related to forest matters. As opposed to tree matters, which are most definitely Adam’s department. So the general look of Audio Damage products, how they fit in to the world at large, and the personality they have are generally my responsibility, while Adam deals with the internal structures, the actual DSP, and turning my flights of fancy in to actual usable products. That’s the general way things work around here, with notable exceptions.

So I’m always on the lookout for new ways to do things that will separate AD from the herd, and one of the things I keep coming back to are true 3D interfaces. I had a rather deep experimental phase going a couple years back, and I kept inflicting a little filter I made with a true 3D pure OpenGL interface on our testing crew. They were nonplussed by the whole thing. I believe one actually pondered whether I still had a firm grasp of my sanity. To be sure, there are several products out there, some well-known, that use OpenGL for drawing. (Expert Sleepers’ Minky Starshine is one such, although it ain’t the prettiest thing out there, and I’m fairly certain that Urs Heckmann uses OpenGL for some of his stuff, although it might just be a combo platter of CoreGraphics and GDI+; hard to tell. Plogue Bidule is done in pure OpenGL, for what it’s worth.)

The reason for using OpenGL is to open up the Z plane; VSTGUI, which we use for everything right now, is firmly bound to X and Y, with nary a Z in sight. To be certain, it is perfectly capable of doing most everything we need to release fine products. But in my head (the afore-mentioned flights of fancy) I see user interfaces for products that evoke Anymails and Fidgt, and for that we need something a bit lower level and more fluid in its execution. I actually have one such product in my head that continues on a theme we started in the early days of AD, but I won’t elaborate on that at this juncture.

Long story short, go watch the Psychosynth videos if you have a minute and think about how a true 3D interface could be used for other things. Personally, as I said, I don’t find the Psychosynth stuff very interesting, but I do see the logical extensions. For the record, Psychosynth was done with the OGRE 3D game engine and CEGUI. He has downloadables there, but only for Linux, naturally. But this discussion I’m envisioning isn’t really about Psychosynth at all, but rather taking advantage of the platform in which we all work, the computer screen, and making things that are completely new, rather than evoking some crusty piece of gear with my Mental Ray chops. How dangerous is this course of action?