Join Kim Lee for an in-depth discussion in this video Digital vs. physical fabrication, part of Kim Lee: Digital and Physical Production and Design.
- My brother came to me with a project that had landed in his lap. He came to me and he showed me this drawing, he actually sent it to me on the phone, and it was a pencil drawing that the director had given him of this big machine that he was tasking my brother with making, and I sat down and I was like so what is this? It's for the Toughpad. They want to make this testing ground machine that's kind of Rube Goldbergish with all this crazy stuff, a torture test kind of device, and that it goes through and it can get wet, it gets hot, it gets cold, it gets dusty, get all these things and pops out at the end.
I'm like, "Oh that's cool, that's really cool, "but let's sit down with them and go over it." So we got the job award and proceeded to go to their office, and I spent a few days sitting with them doing previs. And we would just sit there and so we had the drawings up on the wall that the director had done. So all right, let's plan this out. We're not going to build each machine but we're going to kind of block out what the action needs to be. So we did a previs of with very primitive geometry, more for timing, more for pacing, more for getting a sense of how this is going to feel and how much time it was going to take to get through the machine versus how much time they had for a limit for the duration of the piece, which would also inform us for the next step of designing each piece.
How big does this particular part of the machine need to be? Well, if we need to get through it really fast, it can't be this long 12 foot piece of machinery. It's got to be shorter so it takes up less screen time. So we went through that process. We did a previs. I gave the files to the director. He would tweak them in his 3D program and give them back to me and I would tweak it a little bit more and clean it up, and then we'd make a previs so they could show their client. That got signed off. And that's all typical stuff that you would do even if it was a full CG commercial, even if it was just a regular live action commercial without all crazy mechanics or anything like that.
So pretty standard stuff. Once we got through the previs phase, then we were tasked with sitting down, like all right, what are these machines actually doing, and how are we getting this device from this tablet, from one machine to the other? So it was really kind of a big engineering thing. It would fall under engineering more. I have no degree in engineering, so it was fun. (laughs) So we basically just dreamed up all these crazy things of what we could do and what we couldn't do.
The director gave us a lot of input like saying, "Well, you know, we have right now in the previs "we have an awful lot of conveyor belts." Because it's a very easy way to move something linearly from one point to the other. So it's like I think we're relying on too many, too many conveyor belts in this. So, okay, great. Now we got to dream up, we got to reinvent the conveyor belt now. So it was like wow, we could have it on a zip line. Okay, that's good. Or we can have it slide down and then something lifts it back up.
So we were thinking along those lines. So we spent a couple weeks, maybe a week or two, to plan out mechanisms, like how are these going to really be built, and that I did in 3D. My tool of choice is 3ds Max. So I would sit down with Max and just bang out this model and actually use some of the Mass Effects physics from it to just try to test things, not trying to be super accurate with the physics, but just say is this plausible, will this kind of work, and we'd always tweak it in the real thing.
So we did that for a bunch of mechanisms for this thing so the machine that we were making ultimately was this little car on a track that had the tablet on it and the talent in the piece would push a button or release it and it would go down this track and they would trigger a point that would cause a torch, one of those butane torches, to fire, burning a string, releasing something else, and these pool balls would fall down.
By now our tablet is rested on this platform and the pool balls would hit the screen of the tablet. And when the third one hits, it would trigger the tablet to get dumped into a ramp that would go into a boat that was on a canal that would get blown by fans and then up an elevator that was triggered by a Jacob's Ladder that lit a fuse that released a can of paint to pull the elevator up and it would dump it onto like kind of a strange slide onto a platform of strange mechanical arms that would move it through essentially a conveyor belt, but our cheek for not doing a conveyor belt, and it would move it through that and get blasted with all this dust, colored dust, to show that it could get dirty, and then it would get dumped onto a zip line that would go through shower heads that were getting it wet and these umbrellas were closing as it was getting to them so it would make sure it would get wet, dumping it into a tank of live piranha to show that it could be submersed, and then picked up out of the tank of piranhas by a drone that had a basket underneath it that would go through another part of the course and dump it into our freezer machine, which was a conveyor belt that would bring it through this cold area and you'd see the temperature and all this vapor from the chilly air.
And then it would go to another machine which was the oven showing that it could get hot, come out of the oven onto another cart, down tracks in front of fireball machines that were shooting real fireballs, onto an RC car with saw blades as wheels, going up a track through a car. This was the noisy section and we dropped a like a 300 pound boulder onto the hood of this car which would trigger its alarm system to go off so you'd hear all this noise.
The idea was that you could hear in a loud environment, you could still hear this thing. And then it would get catapulted at the end of that track. A catapult would throw it onto the ground to the finished piece where they would pick it up and say, "Look, it still works." So we built that. (laughs) We had to figure that out. And we couldn't have done it without melding the two worlds, like doing a lot of digital planning, working things out, make sure it was going kind of fit, and then that would inform how we build it.
We'd have dimensions then. We'd have tested it out. The client would have seen the idea like, "Yeah, that idea is fine." And we could go forward. There was already a tough schedule. But if we hadn't done that, then we would have wasted a lot of time building something that didn't work, that we thought might work but didn't. Yeah, that was a really good mix of the two worlds, I thought. The hope is that we don't lose the traditional. Like for example, the show I worked on, The Nick, when I was doing all the specialty CNC cutting I remember asking them at first, "All right, can you send me the CAD files?" Because I had to set up cutting files for all these pieces so everything's going to fit and the carpenters don't have to do a lot of work when they put it together.
And they said, "Well, we'll send you a scan." Like send me a scan. So they send me this scan, it's a hand drawn scan. I was like all right. When I finally visited the stage and saw the art department that was working and the guys who were working there making all this stuff, it was all like the best union draftsmen sitting there. Brilliant work but they had got them, and none of them were on a computer. They all had drafting tables. They were doing it old school. They had the slide rulers, and everything about that set was drawn on the big paper like old school.
And I was like really? Nobody's using AutoCAD. Nobody's using, like wow. Okay, hey the stuff looks great. My point being there are certain sensibilities. There's a lot of knowledge, a lot of experience, on the non-digital side, that I hope doesn't get lost in the transition. I hope it's not, all right these people know this tool, and now they're doing it, as opposed to a mix of that and people who know and have the eye and have all this other knowledge that's tool agnostic, them coming over and learning some of the digital tools.
Maybe not trying to be as good as the kids who grew up with it, but I'm hoping enough of that happens to preserve a lot of the craftsmanship and the artistic stuff that's in all these artists who came from the traditional. Digital and physical. For a 3D app, my app of choice is 3ds Max, as far as the main package. But that doesn't mean I won't use other stuff.
I've used Sofimage, may it rest in peace. I've used it, it's great. I've animated in it when I was freelancing for SYOP way back in the day. It's a great tool. I played around a little bit with ZBrush but not enough to say I know ZBrush, but it's another great tool to use. I've worked with RealFLO quite a bit. A lot of the plugins in Max, the fume effects, things like that, editing package, I mean because I'm interested in the whole process and I've been around long enough to do all of this, so I've composited of Photoshop work, all that stuff.
So all the usual tools, all the Adobe stuff. I like Premiere a lot these days. I used to be all Final Cut and I'd looked at Premiere recently again, I was like, "Oh, this works pretty nice. "This is good, it's not like Premiere 10 years ago", when probably was the last time I looked at it. So it's not really a fair comparison. So I'm using Premiere Pro for editing these days. Compositing, I'm actually for myself I'm really interested in playing around more with Fusion.
I know it's not the popular choice right now, but it looks very interesting to me, so I want to play with it some more. But I've worked with Nuke before. i don't really do much compositing anymore since we have a department for that. So for CNC cutting, especially when I'm doing dimensional cuts, anything 3D and contoured like a face or something like that or a spaceship or a robot like we did for Autodesk, I tend to like DeskProto.
It's a really great package, very affordable, that gives you a huge amount of toolpath options. Some of the other software is a little bit more limited with that. And then, once it's made and I'm in the physical realm, it may sound embarrassing, but at home I use Ryobi because they work fine and I've actually used them for years on set doing stuff and I get laughed at because I don't have a Dewalt, but it puts the screws in, so it works.
I like Ridgid tools for some of the saws. We have a bunch of Ridgid saws at the shop. But Dewalt is also there. We have Dewalt nail guns and drivers and things like that. I've been building a lot of tools lately. So like I'm in the midst of building a big CNC wire cutting machine so I can cut big four by four by eight blocks of foam in like one piece and just do contour stuff with that. So it's not really a brand, it's like Lee Tools.
That's funny, I'm not used to being asked about like what are your favorite tools when it comes to the physical side. So it sounds silly to me say like, "Well, I'm a Ridgid man", or, "I'm a Dewalt man." And it's making me think how silly we sound on the digital side. Like, "I'm a Maya guy." Or, "I'm a Houdini guy, I'm a Max guy." Like you're a 3D artist, it doesn't matter. (laughs)