As you may have noticed, I quickly abandoned my plans to make a daily record of my progress after I spent the better part of Tuesday thinking and writing about what I'd done on Monday.
From there, having realised that I needed to return to the basics, I went over rigging a character for animation using Maya's own "getting started" tutorial. (http://download.autodesk.com/us/maya/Maya2012_Getting_Started/index.html) I'd already gone through the modelling, texturing and animating parts of the tutorial, and I find that they're a good introduction to the program, although they really do only cover the basics. Unfortunately they're not enough to let you successfully create, rig and animate your own model, as I found with some interesting results...
After much perseverence, I managed to get something that looked a bit like a bear trying to jump, so I went with that, and ended up with this:
Shocking animation and terrible compositing, I know. I found animating in Maya frustrating, because I got a much poorer result than I would have done in 2D, despite spending a lot more time on the work. Setting up the model for animation was particularly time-consuming, and even then it didn't deform the way I wanted it to, and was difficult and unrewarding to animate.
I do intend to continue trying to learn 3D software for animation, but I've realised that I severely underestimated the steepness of the learning curve. Although the basics can be picked up relatively easily, making anything which actually animates well and looks good is much harder. While looking for uselful resources, I came across Digital Tutors (http://www.digitaltutors.com/11/training.php?cid=5) which offers some free video tutorials, and then the rest on subscribtion only. I didn't use any of the tutorials, but they seem to have a lot which go into depth on rigging and animating in Maya.
For this project our instructions are to choose a filmaking technique or approach that we would like to learn more about and to do some individual research into that. I've wanted to learn how to use 3D software for animation, specifically Maya, since it's one of the few major packages available on mac and, on top of that, Autodesk offers a free educational lisence to students.
I'm not so much interested in making whole films in 3D as being able to use 3D elements in combination with 2D. To research what aspects of 3D are most important for this technique, I decided to look at films which use 3D elements in a way I like, and find common factors between them. These are the films (I thoroughly reccommend all of them, if just for the visuals):
While looking into how some films were made, I found that several films which I had thought used 3D where actually made entirely in 2D (and some which I'm still not 100% sure about).
This goes to show that good animation and clever compositing can convincingly imitate three dimensions, but for the sake of broadening my skills I'm going to make the effort to learn the basics of 3D animation anyway, and hopefully some tricks to help blend 3D and 2D elements.
Looking at the films above (including the 2D ones, since I like them, too), the design choices that I feel all of them have in common are:
Flat or cell-style shading (this is a little ironic, since the 2D ones are cell shaded)
Restricted colour schemes
Grain, noise, or brushstroke effects to introduce analog-like elements
Sharp geometric shapes contrasted with curves
High levels of stylisation in character design
Abstract elements
Obviously, five days isn't enough to thoroughly reseach all of these aspects, especially when I first need to learn basic modelling, rigging, animation and rendering in Maya. I've therefore decided to try to model and rig a (very) simple character, animate a small movement, and learn to render and export the result in a way which lends itself to compositing with 2D elements (i.e. cell-shaded).
With complete disregard for logic, I started at the end by checking two tutorials on cell-shading in Maya.
http://shikijiyu.deviantart.com/art/Maya-Toon-Shading-Tutorial-156746043 - easy to follow instructions
While both are worth a look, neither of them provides a complete route to the result I wanted. The first has clear, easy to follow instructions which give a reasonable result, though only after some tinkering with the settings (see the 1st and 3rd images below). The second completely skims over how to render a cell shaded image in Maya, and instead focuses on compositing seperate outline and fill renders in photoshop. This wasn't particulary helpful to me, but it did indicate a possible direction to take, though my efforts to get an outline that matched the shape of the model ran into some difficulties (see the 5th image below).
Oh, I modelled the bear based on this design, but it could use a bit more work:
Also just finished (well, nearly finished) the Memory Project, a short film about
my first pet. I'd still like to go back to it and add some colour,
but for now this is how it stands. In retrospect, I think I tried to fit too much action into the film while trying to keep up with the narration, making it a little bit confusing to watch.
Here are some screenshots to make up for the low-quality video:
Just finished work on the animated short for a project with the Apex hotel chain. The video will be officially launched online early next year, but for now here are some screenshots. Sorcha unfortunately left the project, so I ended up completing it on my own. The shooting was done in seven days in the studio, and I made the film with as little post-production as possible, since our original concept was to focus on the fact that we were interacting with the same physical object that hotel customers would find in their bathrooms. The animation process actually ended up being more of an exercise in special effects than anything else - I spent more than half the studio time on the snowing section. Lighting the scene was also a challenge, as I wanted the light to be bright enough to erase the gap between the platform and the backdrop, but not so bright as to erase detail on the duck or cause multiple strong shadows. The solution I came up with wasn't perfect, but time restraints meant that it had to do.
The other project we've got this semester is the Apex Project, a semi-commercial (i.e. for a client, but unpaid) piece of work. We had the choice of two briefs. One was to create two 30-second pieces of animation to be played in the hotel reception on a special three-screen set up. This is to be related to some new branding for the hotel chain, and was intended to be a more abstract piece.
I played around with a couple of ideas involving maps, then star charts and RGB lighting, but this image was as far as I got with that.
Although I'd like to work further on something like this in the future, I didn't feel comfortable enough with this sort of abstract motion graphics subject to pitch it to a client, so I chose to go with the other brief instead. This was to create a 1-minute animation using the Apex hotel chain mascot, a brightly coloured rubber duck, using the theme of one or more of the four seasons.
My initial idea for this brief (after playing around with
different ways to draw the duck) was to use the rubber duck itself as a
canvas for the animation, rather than a character within it. From there
it was a short step to come up with some seasonal imagery than could be
painted and then animated on the duck in stop-motion.
Because this technique will be particularly time-consuming, I arranged to work on the project with one of my classmates, Sorcha. The second part of the idea came when we were discussing ideas for the Winter/Spring section and we realised that we could substitute the rubber duck for parts of the film with different models, giving us a lot more flexibility and the ability to physically transform it into different shapes. For example, a snow duck, sand duck and topiary duck:
This helped us create a much more interesting storyboard to pitch to the client, combining painting on the duck with stop-motion animation.
The pitch went well, and we got approval to go ahead with the project. Next we'll work on a finalised, and hopefully improved, version of the storyboards, and do some animation tests to work out how much time in the studio we'll need to realise some of the more complex ideas.
