Motion Capture in a Nut Shell

Motion Capture is the new shiny object for the animation industry, using the word new loosely. However, many people still don't understand what motion capture really is or how it works yet. So we are going to take this post to explain the different versions along with the pros and cons of motion capture.^[1]

First up is "Mocap."  This is the generic version of motion capture that everyone seems to be the most familiar with.  In this version an actor is given a spandex suit with glowing orbs around it and is put inside an empty set with dozens of cameras tracking the position of each orb as the actor moves.^[1]

I-Mocap also known as inertial motion capture uses smaller sensors than the generic mocap.  This makes use of tracking bright colors on a suit. The cameras track simular to the previous version, but they have the advantage of letting the actor move more normal by not having to make use of the orbs that can get in the way.<sup>[1]

[2]</sup>


There are also now some companies making fully thermal motion capture cameras such as the Kinect from Microsoft. These cameras could be considered a cheap route but they also come with more flaws like not being able to track as much data on exact position of body parts, and can not track any facial animations.

Advantages:

• fast results ( also using the term fast loosely)
• director can correct acting mistakes before they are applied to an object to be animated
• more data is taken in a shorter amount of time than traditional animation processes
• There are some free ware softwares that reduce costs ( just remember you get what you pay for )

Disadvantages:

• Specific hardware and software are needed to make use of this process
• Expensive, most smaller studios can't effectively use motion capture
• Actors still can't break the laws of physics so animation will sometimes still be done by hand
• Models or objects that are not in the same proportion are much more complicated to adjust to fit than to just start traditionally.^[1]

References:

1. motion capture explained Uknown Author.  August 14 2011. Computerstories.net

2. Match Move Reel vimeo video active on 11-18-2012

The Animation Process in Comparison

The previous posts have talked about the fundamentals of animation, the start of the digital era, and how technology has effected the industry thus far. So now we begin to get back to animation as a process and move away from the idea that it is a product created by one person.

The original animation process with pen and paper required up to one hundred animators or more to complete the Disney feature films due to the shear amount of drawings and coloring that needed to be done. For the film "Snow White and the Seven Dwarfs" there were roughly 1,500,000 drawings and water color paintings to complete the film.^[1]

Now art teams are now typically around 30-40 people strong for a feature film and less for a television series. The animation team could be broken down into 3-7 animators for an entire film.  Along with the technology of motion capture now animators don't always have to find references to what movements look like. Actors are now able to fill a decent portion of the animation process and the technical animators will go back and clean up the movements.^[1]

Before digital animation because popular the art team used lead animators to make the pose by pose pictures and the other animators would make the filled in frames of the animation. This took a lot of time to complete due to having to wait on a few artist to finish their work so that the others can start working on their part.^[1]

Much like before digital animators still use animatics and blocking to get the animation ready before they spend time making final versions. Animatics are drawings of the animation with basic movement showing the direction of camera and characters. Blocking is small increase from an animatic where the 3D animation is happening with rough shapes of characters or primitive shapes to replace an object in 3D space so that the animators can get a feel for the performances before adding the final finesse to the project.  This keeps animators from wasting time working on shots that the director does not like for production.^[2]

This video is a good example of a blocked out animation. Going from pose to pose just getting the major parts down.^[3]

References:
1.Making of Snow White
2.About Animatics
3.Animation Blocking "WIP Lever"

Technology and How it Effects Digital Animations

Technology advances have increased the amount of work that can be applied to an animation within the same amount of time.

The average computer in 1995 had 8 megabytes of RAM and also had 420 megabytes of  Hard Disk Space. 

In comparison the average computer for today's standards has 8 to 16 gigabytes of RAM and has 1 terabyte of Hard Disk Space.

RAM vs. Hard Disk Space: 

RAM is the memory that your computer uses to compute processes and is not permanently stored. 

Hard Disk Space is the memory used to store information to be recalled whenever the user wants and    is stored until told to delete the data.^[1]

For those who don't understand the differences in megabytes and gigabytes:

1024 megabytes = 1 gigabyte^[2]

1024 gigabytes = 1 terabyte  ^[2]

Just to give some scope to the amount of memory.  Toy Story used around 3 terabytes of Hard Disk Space to hold all the files for the movie at completion.  This was held over multiple servers of memory, now days that could be held on a couple desktops. 

This information is important due to memory and processing power has been the "wall" that animation has been held back by.

The increases in the memory along with processing speeds that have increased at an equal rate to the memory allows the computer to do more functions at once but have not caught up with the pure power behind the software right now such as Autodesk's Maya or Pixologic's Zbrush.  

One of the best ways to show the changes due to technology is to just give examples from industry references such as, "Toy Story was originally rendered back in 1995, and to our best estimate probably had frame render times which averaged in the range of 4 hours or so." according to Craig L Good a pixar artist. Later on Craig Good estimated that with the performance differences in computers by todays standards, they could render each frame in only 15 seconds. However, just because they can redo a movie from 1995 in a flash does not mean that every render time is that low.  An example of this is in Transformers 3 each frame of a Transformer to render took around 72 hours according the commentary on the DVD.^[3]  

Computers have become faster but that has only increased the effort to push the material farther into realism.

References:
1.Computer Comparison via Relatively Interesting

2.Computer Memory Chart

3.Craig Good via Quora