cheetah3d + microtonic

I stumbled upon this 3d program I really like called cheetah3d. Selling for only $129, this Mac-only 3D modeling, rendering and animation program is clean and mean with a great, easy to use interface. Just clearing out the features you hardly use is a feature in itself. Really nice. It's got me rendering a spaceship taking off at this very moment. Plus, this thing is full of features—IK, F-curves, radiosity, you name it. The one problem for me is that there's no motion blur. I wrote the program's author and he said that's high on the to-do list. For the kind of stuff I do—a couple of 3D animations here and there output with alpha channels, this promises to be a nice package.

The program reminds me a bit of soniccharge microtonic. Like cheetah, sonic charge is essentially a one-man operation. Their one product--microtonic--is an inexpensive, quality software drum synth with a definite electronic vibe.

Bloomin' daisies!

I'm pretty proud of my blooming daisy video for the current theater project. I wanted a stop motion type effect, sort of like the blooming flower in the title for Magnolia. I was actually going to do it stop motion but I couldn't figure out a way to animate the silk daisy I bought. So I figured out another way to do it. With my hands in black socks in front of black paper, I pulled the daisy through a toilet paper tube that I had narrowed down. Played backwards it creates a beautiful blossoming effect.

My attempt to create a burn element wasn't so successful. I lit some black paper in the bathtub hoping to get a slow smouldering burn that I could use for a traveling matte. But the paper burned unevenly into a weird smile shape and then the whole thing went up in flames. The fire alarm went off and the tub was filled with black ash. Bad idea. I ended up using the CC burned film plug in that comes with After Effects.

TV vs. film vfx

Talked to the director again today. He wanted three changes: make the bombs skinnier, make more of them at the end and extend all the animations to 40 seconds. All in all, pretty easy.

I happened to read this tonight in Cinefx 105. It goes well with what I've been writing about for the past couple of days.

KEVIN KUTCHAVER: I think the best thing about TV is its fluidity. On films, we're just taking our marching orders from the visual effects supervisor. They've already spent a year planning everything out; and so, we are just there to carry out a plan that has already been put in place. It is more rigid. In television, often we're working with a cut that isn't locked, so things are looser.

CINEFX: That seems counterintuitive. You'd think TV would be more rigid because of the lack of time.

KEVIN KUTCHAVER: But see, since there isn't time or money for preplanning, shots have to be created on the fly and even though there is pressure in that, it can make for a very creative, high-octane work atmosphere.

SAM NICHOLSON: In contrast, features can become painting by numbers. By the time you even bid a feature the shots have been boarded, they've been prevized, they've been chewed up and spit out 50 times—and then, after you finish it, you do 60 or 70 turnarounds on one shot! So by the time you're done with it, you hate the shot. But in television, you have a meeting with the director to get a creative connection going; and then they say, 'Okay, go out and do it.'

Shim Ch'ong

More information file images. This is from another theater piece I worked on.

Noise "contact sheet"

I'm putting together my info file for school so I made this page of color-corrected Noise images.

The haircut problem & unplayable effects

If vfx are performances, then perhaps they should be directed like performances. Rather than result-directing fx, you'd treat the fx person more like an actor. The idea wouldn't be to get the fx creator to do what you want (result directing). You'd get them to make something that works. A totally different goal. I realize that I've been thinking about this since college. It's the haircut problem. Do you tell the stylist exactly what you want (a little off the sides, not too much off the top) or do you leave it up to the stylist? The best approach is to find someone good and let them do what they want. It all comes down to casting which is something producers and fx supervisors often do take into account.

Next is considering what is an playable vs. unplayable effect. Certain things were just never meant to be seen on screen. That's the conclusion I came to with the bird effect in Noise film. For whatever reason, the scene was unplayable so I changed it. In acting, a good example of an unplayable scene is anything with an emerging split personality. You can play it for comedy. Steven Martin was great in All of Me and as was Jim Carey in The Mask. But for drama? It can't be done. It's the scene that made acting road kill of Glen Close in Maxie and Halle Berry in Catwoman. It's a testament to Alfred Molina's skill that the split personality scene at the end of Spiderman 2 doesn't stand out as being ridiculous.

There's a difference between uncinematic and anticinematic. A third act exposition is uncinematic. You have someone talking for a long time (again, Sleepy Hollow is my favorite example) or in my case, you just find a way to rationalize putting text up on screen that reveals the plot. The exposition needs to be done so you just try to make it as painless as possible.

But certain images are anticinematic. They just shouldn't be put on film. Like the switcheroo in The da Vinci code that I complained about in an earlier post. There's just no way to make that work. That scene looked like bad TV because that's what bad TV is. It's filled with unplayable, anticinematic scenes that could only exist in a writer's imagination. An example of an unplayable effect that was in fact, never played, occurred in a Smooth Crimina-era Michael Jackson music video. If I remember correctly, the story goes that Jackson wanted Rick Baker to do a robot transformation that started with metal coming out of Jackson's pores. Baker told him not to do it saying, "it looks ugly and I don't think Michael Jackson should look ugly."

fx supervisors as designers

It occurred to me that one thing I'm doing for this project is serving as vfx supervisor. From the little I know, it seems that fx supervisors are really like designers in the same way that production designers are designers. They think about looks and approaches but don't necessarily do all the work. I think that's why the bird-into-planes animation was intriguing enough to engage several minds simultaneously. As designers, we're trying to think of the best way to solve the problem. What's interesting about Ben's approach is that it's theoretically possible but not practically possible. Yes, I knew it was possible to find videos of birds and planes. The problem is finding videos that function as plates—with the right bird flying the right way totally isolated and easy to key. Still, before long it will be possible to follow Ben's approach. The information is there in the video. There just needs to be some inexpensive software that extracts and stabilizes the image, interpolates missing areas (or allows you to do so) and creates an animated and textured 3D model from the data. In the end, it's interesting that for this one animation we came up with three different approaches: Ben's (stitch together videos), Craig's (morph-animate still images) and mine (use 2.5 D images and a simple dissolve).

The assignment that launched a thousand kibbitzes

I finished versions 1 of the bird-into-plane sequence and the rain-into-bombs sequence. The bird-into-planes sequence seemed to generate all sorts of comments. Ben had this idea that it should look like a fireworks explosion where the birds come to a point and then the planes accelerate out of that point. Then he had the idea that the way to do it would be to get a video clip of a bird and a video clip of a plane and somehow merge them together. Not only that, after I told him "yes and where am I supposed to get footage" he found me some clips on the web. Then Craig got into the act. On Thursday I asked him if he wanted to work on that sequence. He didn't say anything so I figured he didn't want to do it. But today he sent me a clip. He used a true morph so it was interesting seeing that. Plus, like Ben, he had the planes accelerating out of frame. But the birds were too flat. Plus, I was already done. Oddly enough, the bird video that Ben sent was also way too flat looking. Plus the bird wasn't flapping wings and worse, parts of the bird were going out of frame. So I'd have to track it, rebuild the wings, and then rotoscope to get this bird working and it wasn't very good to begin with. Very cute though Ben, just like when Sean tries to help me make dinner.

It's funny the acceleration idea kept coming up. Ben later looked at my finished video and then decided it would look OK if I did an anticipation and then accelerated out after the transition to the jet. I kep telling him that if I accelerated out, no one would see it. Plus I think the anticipation would make the sequence look too cartoony. These are tiny jets here. Craig's acceleration was way too fast. In the current version I transition the end planes more quickly so we end with the feeling of planes rather than birds. This probably doesn't work that well though. If I have to make changes to this animation, I'll probably make the transition later so the audience will be able to see it better. Seeing the transition is more important than ending with planes I think.