The everyday blog of Richard Bartle.

RSS feeds: v0.91; v1.0 (RDF); v2.0; Atom.

Previous entry. Next entry.

11:14am on Friday, 24th December, 2010:

Non-Motion Capture


If I were to build a 3D virtual world, I'd want to do it in Lego.

Not the low-level details and the texturing — that's for artists. Not even the general terrain — I'd generate the first cut for that procedurally. For structures, though (buildings, caverns, cityscapes), I'd want to do it in Lego.

Lego is digital and 3D and really quick to use. Yes, I know there's wondrous software that enables you to build 3D structures in computer memory, zooming in and out, rotating shapes, hollowing out tunnels, painting surfaces with walls, and that's great; however, I don't care how good your 3D mouse is or your 3D screen is, you won't get a consistent, coherent prototype structure made as quickly as you would if you used Lego. Maybe — just maybe — if you could use some system that allowed you use two hands, such as the Kinect, and if you had a big enough screen, you might be able to get close or even exceed what you can do with Lego. The same could be said of animation software, though, and yet people still use motion capture. I want non-motion capture.

I want to be able to take a large baseboard and place blocks on it and then read those blocks into some standard format that can then be tuned up using regular systems. Those three blocks are a pagoda: some artist can make them look like a pagoda on the screen, but if I'm building a world I don't actually need to see it as a pagoda, I can do that in my imagination; I just want to see where the pagoda is relative to everything else, and whether it should be higher or not. On a different scale, those three blocks are a chimney. On a different scale, they're a hatstand, or a chair leg, or a pencil. They're all yellow because when the same colour blocks touch each other they mean it's the same object. It won't be a yellow pagoda, that's just the bounding box the pagoda will approximate. I just want to be able to put things in places really, really quickly, among a bunch of other things I'm also putting in places.

The reverse is possible: there are systems to make Lego blueprints from CAD software. There are also AI computer vision systems that can capture a scene in 3D very quickly and load it real-time into models usable by robots. They only do visible surfaces, though. They don't do interior surfaces, like the inside of a shop or the cellar of a castle. I couldn't make a staircase spiralling down into the depths even if I made it from negative space (so the Lego represents where there isn't rock, rather than where there is rock). That's what I want to be able to do.

As for how to do it, well it can't be all that difficult. You'd only need to stick something inside a Lego brick — an RFID chip or a piece of aluminium foil or something — that can be picked up in a single scan and then converted into a 3D model inside a computer for the art team to descend on. It could even be a product in itself: it would be pretty cool if kids could make a town out of Lego and then load it into their computer and walk around it as if they were the same scale. Lego lets you build a 3D map.

Besides, if nothing else, a rapid spatial prototyping system would mean that whatever inn, cave system, castle or shop you went into in an MMO, it wouldn't have an eerily similar architecture to almost all the other inns, cave systems, castles and shops you've been into in that MMO...

Latest entries.

Archived entries.

About this blog.

Copyright © 2010 Richard Bartle (richard@mud.co.uk).