This was our first work outside of Alden Hall. The Story of Dr. Faust was being presented at Worcester Foothills Theatre and the director wanted every sort of magic he could get his hands on, including high-tech projections. After many discussions about just what we could and couldn't do (and at this point, that seems to be a major hurdle-- getting directors to realize what we can't do) we agreed to do three cues.
Now going from sixty cues to three cues may seem rather anticlimactic, but this seems as good a time as any to talk about time constraints. It takes a huge amount of time to do these virtual worlds. We're constantly making software and hardware do things that they weren't designed to do, and both learning the equipment and making the necessary 3D models take awhile. So for Medea/media we had a crew of 14 students and myself designing and building 60 cues over a period of about 12 weeks. For Faust we had two students and myself doing three cues over seven weeks, and we just barely made it. We totalled about two and a half minutes of stage time.
The setup:
Certainly much simpler than Medea/media yet it came with its own set of problems. Because the stage and screen were quite a bit smaller than we'd ever worked with, the projector had to be hung over the audience. Up until now, whenever we needed to shutter the projector, we relied on the high-tech solution of a piece of black wrap duct taped over the lense-- the operator flipped it down and we had a shuttered projector. At Worcester Foothills, the projector was a good 60 feet from the operator. Steve Hocurscak worked out a simple shutter operated by a solenoid which was fine for quick shuttering, but we were also in the habit of putting the projector in standby mode to save on bulb life and cut down on fan noise. After blowing a bulb at Foothills and finding replacement bulbs cost $450 each, we decided that conserving bulb life was a priority. The projector comes with a remote, which could put it in standby from the booth, and after some practice contorting themselves to get exactly the right shot, the operators could turn the projector on and off with a good degree of reliability.
We also needed 60 feet of VGA cable. Believe it or not, that's a taller order than it looks like. Since most monitors stay within six feet of their computers, and since the VGA signal deteriorates rapidly over 15 feet, there isn't much call for lengthy VGA cable. We ended up buying 10 foot lengths of cable and attaching them to each other. With all the plugs hooked up in succession, I was sure we'd get signal degradation, but we attached a splitter/booster unit that was good up to 200 feet, and I couldn't detect any picture degradation at all.
Unfortunately, I don't have any production photos, so all I can do is post some screenshots of one of the worlds. We stretched ourselves here because the director asked for a complete reproduction of the medieval town of Whittenberg for Faust to jump on a barrel and fly over. Quake does much better with indoor spaces than outdoor spaces. In fact, it just can't handle allowing the operator to see very long distances. Steve was given the job of making a medieval town in which the streets were all very twisty, allowing the viewer only short distances to see. I was in charge of populating the city, which was the first time we worked with models and skins. I came up with two basic models-- a man and a woman. Each model had five different skins (the texture that wraps around the model) which allowed for 10 different looking men and women to be wandering around Whittenberg. I adapted some randomizing code I found on the net so that each time we started the city the men and women would look different. One time we might have five shirtless workmen wandering about, the next three fops in red and two purple clad pages.
I then
adapted the behavior code for the Quake knight (the only monster in Quake which
doesn't shoot anything). There are three basic behaviors, two of which concerned
us and needed animation frames. The first was normal walking around. A simple
walking for both the man and woman models was pretty easy to generate. The second
was when the monster sees the player. Normally there's a check so see how far
away the player is. If the player is over a certain distance, the monster shoots,
if under the distance, it charges. With the knight's inability to shoot, I only
had to generate a reaction to seeing the player. My idea was to have the men
point and run (after all, they're supposed to be seeing a guy flying over them
on a barrel, so pointing and running after him seemed like a reasonable reaction.
The women would put their hands up to their mouths in a "Home Alone"
kind of reaction. They'd all chase after the player.
Unfortunately,
when building the models, I didn't bother to articulate the men's hands, so
while they did point up at the sky, from close up it looked more like they were
giving a fascist salute.We noticed this during technical rehearsals, which was
way too late to start articulating their fingers. As a solution we walled the
inhabitants of Whittenberg away from the player. Unfortunately (I keep using
that word!) during one performance the operator left the town running instead
of pausing it. Some of the men apparently found a way around the walls and crowded
the operator during the show, affording the audience with a close up view of
our inappropriate fascist salute. Apparently some complaints were registerd,
and we had to make sure pause the town at the start of the show in order for
that not to happen again (which it didn't).
The other
thing the director wanted in both the Quake cues (the third cue was a simple
videotape, which we played every night without a hitch), was to have the player
recreate his movements precisely each night. The actor had to move along with
the motion of the screen, and without precise timing, the actor wouldn't be
able to do it. We used a floating platform and programmed its movement. This
kind of defeats the whole purpose of doing the VR on the fly (we could just
as well have made a movie and played it-- probably would have been much simpler
that way), but without the actor facing upstage, there's no way for the actor
and operator to move together.
This was actually a huge learning experience in terms of what our projection technology was good for and what it wasn't.
Return to Virtual Reality and Project Page.
Return to Theatre and Theatre Technology Homepage.