So not much work done on Thanksgiving. I did learn how I can render a scene so it'll look cartoony which I thought was neat. I also got around to learning how to produce better videos from Maya. So I went back and redid all the videos I had previously posted. They aren't HD by any means since I still wanted to keep the file size small.
Monday, November 26, 2012
The sunlight I used last time was built in and easy to use. Now I'm going over how to more directly control the lighting. This chapter has a lot of information and less hands on type stuff. There's various types of lights with different effects and functions and of course many ways to tweak them. You often have to use a combination of types to create a realistic look.To make a bare light bulb, I used a downward pointing spot light. Then I created a glowing sphere so it looks like the light is coming from the bulb.
I'm done with animation for the time being and now I'm moving onto lighting and rendering. Each one has their own chapter but they're both very interrelated. On the top left is the wagon I textured awhile ago and how it looks in my workspace. On the right is the same wagon rendered using built in sun light.
Rendering takes a bit of time so what's cool is I can highlight smaller parts and just render those to check changes I make. Here, I toned down the brightness and removed the reflection on the wood panels.
Here's the final render after I adjusted the sunlight to be more towards twilight.
Wednesday, November 21, 2012
I went over some basic facial animation. I duplicated a creepy looking face and made it smile. Then with some magic, I could now make the original face smile a little or a lot via a slider. Next I made another duplicate and raised its eyebrows so now I could do the same with the original face. The main idea is that you can make a number of set poses and then by manipulating all the sliders you can create a variety of facial expressions. This is probably more advanced stuff than I need to know right now. There were more online lessons regarding animation but I figured this would be a good place to stop and go back to the book.
I jumped away from the book again to the online lessons. Those went over making skeletons and how to fix deformations again. Then they had me animate another walking guy. Below I used inverse kinematics instead of forward kinematics. I still have body parts attached to a skeleton. However now I have another chain here linking the foot to the hip. Now I can just move the foot and the rest of the leg moves along naturally. So forward kinematics would be like you yourself moving your arm. You determine how each part moves. Inverse would be like someone holding your hand and moving it around. Your arm follows along with the hand movement by itself.
Next I animated the wheels of a train. Here they were showing me how I can link different objects together so when I move one, everything else moves along with it. All the parts are linked to the middle wheel above so I simply set that wheel to rotate and everything else fell into place.
So the previous guy was just made of floating blocks which was simple to animate. However when you work with more realistic things, problems can arise like how the skin of your model reacts to being bent. Those areas tend to get smashed or distorted or worse. Above is the skeleton I made that goes inside the hand. Then I bent the fingers and my book showed me the various ways I can fix deformities around the joints. None of the methods seemed particularly quick and easy or produced great results though. I was going to animate the hand to flash some gang signs or something but I wasn't happy with the initial skeleton set up they had me do so I just moved on.
So here's an animation of a walking dude. I created a skeleton of movable joints and then attached the body parts to the skeleton. Then I can move the joints around and the body parts will follow suit. I used a technique called forward kinematics to animate the body. It is called this because for example if I rotate the upper arm, everything down the chain ending at the hand will rotate the same amount. Things done at the root will move forward down the chain. It's easy to grasp but can take a lot of work because you have to rotate each part yourself.
Thursday, November 8, 2012
Here's a video of some text flying around. The video goes crazy at the end but you get the gist. Next I went back to the catapult I made awhile ago and animated it. Now I'm done with another chapter. That seemed to go by pretty quick. Next up is another chapter on animation though.
Wednesday, November 7, 2012
Tuesday, November 6, 2012
This next lesson was quite a bit more complex. Another case of me understanding what was going on but probably unable to replicate on my own. The main idea was creating a skeleton with joints to control this mechanical arm. So instead of animating each part, I can just move around the end and the rest of the arm will move accordingly. But they also threw in a bunch of new ideas and tips and that's where it gets overwhelming.There were more lessons on making skeletons and moving characters around but I'm going back to the book to reinforce the basics.
Sunday, November 4, 2012
This video shows how one object's animation can depend on another. Instead of animating both objects individually, I have the door's height depend on the balls position. Now when I move the ball to the right, the door automatically moves up.
And lastly, I have a plane traveling along a track at different speeds and banking around the turn.
Friday, November 2, 2012
A really detailed model usually has a lot of polygons. It looks nice but it's tough on computers to process. This lesson shows another use of textures and mapping. Here the detailed model on the left is used to create a map that tells the program how light should bounce off its surface. This map is then placed onto the low resolution model in the middle and you end up with something like you see on the right. It has the same small amount of polygons but with this extra layer of information, it looks a lot better while being easy for the computer to handle. This lesson was more about showing me what was possible since most of it is over my head at the moment.
Now a soldier guy is a lot more complex than a cracker box and you can see how jumbled up the map on the right is. The trick is to learn how to unravel the mess so you can easily place the 2-D image (or create a 2-D image) and have it go where you want it to go. First off, I separated the soldier into different parts (arms, legs, head, etc). The arm was easy. You can imagine taking a sleeve, cutting along its length and then laying it flat. That won't work too well with a head though. So you can understand how there are multiple ways of unfolding these shapes and learning how to use the right one is important.
So the lesson taught me different ways of unfolding. Once I have the different maps, I can export them and use them as guides to create textures in programs like Photoshop. But here they provided me with textures so it was easy to apply onto the model.
I continue to tread familiar ground with these online lessons. I'm onto applying textures onto models again. This is a very simple introduction involving a box of crackers. I take the three dimensional shape, unfold it to create sort of a map, and then I'm able to place a two dimensional image onto the map which then gets placed onto the model exactly where I want it to go. Now with a box, it's pretty straightforward. Things get a little more tricky in the next lesson.