Monthly Archives: July 2012

First Look At Outlaws

I finally have some still shots from the Outlaws trailer to show you!

I’ve been working for about a month on this project for Bill Weldy’s novel, but I ran into some technical trouble with one scene that shut down production for a while. I knew when I built the set that rendering it would eat up most of my computer’s resources, but I had no idea it would completely crash my system. But it did–three times. Fortunately, the issue was related to physical memory and had nothing to do with my hard drive, so I didn’t lose any data. I was lucky. But I also feared I would never be able to produce a high quality scene that matched Bill Weldy’s vision, and would therefore be limited in all future projects when it came to the amount of detail I could render.

This scene was the culprit:

I’ve admitted that I’m a complete greenhorn when it comes to 3D animation, but every now and then I learn something that might benefit other beginners who stumble across this blog while searching for answers to their own iClone problems. So let me give you the basic specs for this scene. It’s comprised of 226,643 faces, 96 props, 5 avatars, 554 trees, 50 grass groups, 3 lights, 1 motion path, and a steady stream of keyframe animation. It takes 6000 frames for the camera to pan from one side of the valley to the other. With pixel shading turned on, it takes 10 minutes for the project file to open and 15 hours for the scene to render in HD.

In layman’s terms: it’s enough to send black smoke boiling from all but the most extreme gaming computers.

Anyone looking at the 2D snapshot posted here might not understand the complexity of this scene. Keep in mind that in video format, the water of the pond ripples and flows, the grass and trees flutter with the wind, and all the animals move. Well, except for the cows. You can’t see them here, but there are three of them, all extremely high polycount. It would be one heck of a workload for any machine. So at first I tried rendering one section at the time; for example, frames 1-800, then 801-1600, and so on. Unfortunately, when I pulled the .avi files into my video editor, the camera position was inconsistent between renderings and I couldn’t splice them together. Grrr, snarl! This is not supposed to happen with iClone! Each playback of every scene is supposed to be identical to the last! I have no idea what caused this, but I figure it has something to do with the complexity of this set and my poor, overworked graphics card.

Plan B. My friend Gary Thompson told me about filming in layers, where you hide certain elements of the scene, film what’s left, then greenscreen it together in pieces and parts. Well, because of the camera movement, I couldn’t make this work. For instance, if I hid all the trees, filmed the entire 6000 frames without them, then filmed the trees with a greenscreen backdrop, the parts of the trees that should have been hidden by terrain or other props were exposed and ended up blocking the scene as the camera moved around. Aaargh!

Finally it dawned on me. . .the fence. I have more than 60 clones of that prop in the scene, all of which reference both texture files and bump maps. (See my Feb 2012 post for more info on bump mapping.) Texture files are necessary, but bump maps create fine detail that is lost at such a wide camera angle, and they are an incredible drain on computer resources. Aha! I went through the file item by item and deleted all bump maps that I felt were superflous (more than 100 of them, before I was through) and voila! The scene rendered.

Yes, it took 15 hours. But the result seemed to please Bill Weldy very much, and that’s what it’s all about.

So I learned that even though bump maps are great for tight shots where each tiny pixel must be shaded, they are a detriment to larger scenes and can even cause systemwide memory crashes. Hoo, boy. Bet they teach this stuff in Digital Arts 101. Wonder if I should take an online class?

At any rate, I do have characters to show you. Bear in mind that Bill and I might end up making some changes to the Josh character, but in the meanwhile, here are the cyberactors currently cast as Josh Grant and his love interest Jolene:

And here is a promo shot I sent Bill, so that he can use it when marketing his novel:

For any of you wondering about my sweet little baby birds, I believe I can report 100% success across the board. I can never be sure, because I don’t know the ultimate fate of them all. But I do know Shakey went free about two weeks ago and I’ve since seen her traveling with a flock of other starlings and apparently doing quite well. I believe I saw Big Mac last week, too–at least those little toes looked green from where I stood. He, too, seemed happy and well adjusted, but he wouldn’t come near.

Melody, on the other hand, visits daily. Several times daily, in fact. So often that I leave the front door open in the cool parts of the day so she can come and go as she pleases. She lands on the ottoman where I keep a dish of her food, eats a few bites, then flies away again.

Chip was with me until yesterday. I had become convinced he would be a permanent housepet. Yes, he’d been outside once, stayed overnight and came back the next morning demanding to be fed. After that, he avoided open doors at all costs, even refusing to follow Melody outside when she would leave. But last night he took the plunge, and I haven’t seen him since. He’s completely weaned and I believe well-prepared for living in the wild. But is it selfish of me to wish he’d come back? I miss him. He’s quite the little character, and the house seems empty without him.

I Love Being Right

My husband Scott has accused me of wanting to be right no matter what. He might have a point—especially when it comes to a certain feathered friend I predicted would succeed in the wild.

Yesterday I released Shakey. She was the only one of the four that seemed ready to go. Chip and Big Mac are nothing but two big babies with no sense of danger. Melody broke her beak a few days ago (more on that in a minute) and is still wearing a splint. So I set the cage outside, put Shakey on top of it, and boom. Just like that she was gone.

I worried all day that something horrible had happened to her. Every other bird I’ve released has come back at least once or twice for food or comfort. Not Shakey. I watched the windows, set the two boys outside as bait. . .nothing. Ah, well, I thought. Win some, lose some.

Then this morning I saw something curious in my front yard. An adult starling was feeding a juvenile starling, which was fluffed up looking all cute with its beak wide open. And sure enough, the adult popped food into its mouth.

I looked closer. It sure did look like Shakey. Right age, right size. . . .

Then the juvenile flew to the cage where her brothers were taking their morning sunbath. They greeted each other like the long lost siblings they are, with lots of chirps and tweets and screeches. Then Shakey flew away, leaving Chip and Big Mac having fits in the cage where they could not follow.

Soon, boys. Very soon.

But Melody, poor Melody. She flew into a window Monday morning and at first I thought it killed her. It certainly knocked her loopy. Her head bobbed on her little neck and she listed to one side, and her wings slid down to drag the floor. . .oh no! I just knew she was a goner.

Turns out she was only stunned (probably a mild concussion,) because within half an hour all her neurological function had returned to normal. But her beak sat sideways! This would never do. Not only would it affect her ability to eat and drink, but her mouth would dry out and all sorts of problems might result.

I’ve splinted birds before, so I knew that good old masking tape is the best thing to use. Ordinarily I take measures to keep the adhesive off their skin and feathers, otherwise removing the splint might re-injure an unstable fracture. But in this case I just taped her beak in the correct position, much like one would wire a broken jaw shut. She didn’t eat the first day, but I made sure she stayed hydrated with Gatorade drops in the corner of her mouth.

Tuesday morning I removed the splint and her beak was only a little crooked. She ate and drank as best she could, and when she’d had enough, I taped her back up. This has gone on since Monday, and today I was able to leave her out of the splint for about an hour. I think she’ll heal just fine, and in a few more days, she should be completely back to normal.

And, I hope, flying wild with her sister.