Let there be llama…

Posted by Jason Steele | Sunday 22 February 2009 11:53 pm

Here is an example of a specific e-mail question I get about once a day:

“hey wutup ferrts was funy how u make ur moveis rite back xthkx”

Seeing as there appears to be some sort of interest in how I actually put together my movies, I decided it might be a good idea to have a handy blog post around that I can point people to, instead of just ignoring those e-mails outright. Which is what I have been doing.

Here was my production process, from start to finish, on “Llamas with Hats.”

Step 1: Make myself an iced mocha latte.

 

This is very "yum," as they say.

This is very "yum," as they say.

This is a very important and often overlooked step. If you aren’t a fan of mochas or iced drinks in particular, a regular latte or even a straight double-shot of espresso will do just fine.

 

 

Step 2: Kill some sort of leprechaun and steal his gold.

This will assist in purchasing the required software.

 

 

Step 3: Write the script.

I write my scripts using standard screenplay format, and then number the lines by hand like this:

script

The numbering comes into play later.

 

 

Step 4: Record dialogue.

I use a program called Amadeus Pro to record my dialogue. This is what one read-through looks like:

recording

I use Amadeus’ noise-removal feature to get rid of any background noise, and then get to work splitting up the file into lots of little files. This is where the numbering from the script comes in. Every line is saved as its line number, like so:

takesThere are usually many takes of every line.

 

 

Step 5: Create an audio track.

I use Final Cut Pro for this. I import all of the voice files as well as any sound effects I have recorded for the short. In this instance, the only sound effect being used is for the background ambiance, and I’m using one that I made long ago for another cartoon.

Every character gets his / her own track. This makes it easier to do lip syncing later on. I find which take of each line works best, and then place them all into my timeline:

tracks

For Llamas, I then exported this into 3 separate sound files. One with just the brown llama’s voice track enabled, one with just the gray llama’s voice track enabled, and one with everything enabled. This comes into play later.

 

 

Step 6: Character artwork.

I usually do character art before background art. The programs I use vary depending on the cartoon, but for Llamas I used Adobe Flash.

The first thing I will draw when making a character is a basic line-version using the pen tool.

llama1

Then I use the Paint Bucket tool to add color.

llama2

And then to shade I use the Pencil tool and draw “shade areas” on top of my image. This allows me to then use the paint bucket tool with a darker shade and fill in the area between the lines and my new pencil marks. The lines are deleted after this is done.

llama3

It’s important to note that everything I plan on animating is drawn as a separate layer. Eyebrows, pupils, etc.

Finally, I draw all of the mouths that I’ll need for lip syncing, and then export everything as PNG files for importing into Photoshop.

llama4

Step 7: Background artwork.

I use Photoshop for my background art. The best way I can describe my process is this: crap loads of layers.

layers

That’s just a small sampling. There are over 50 layers in the background art for Llamas with Hats. Everything is its own layer. The door, for instance, is five layers: the door itself, the door frame, two hinges and then a layer for the small shadow it casts onto the wall.

I begin the whole process with an incredibly simple sketch of where the main objects in the room need to be. This is why I draw the character art first – because now I have the character art as a reference for how I need to design the room.

background

 

Then, using various custom brushes, I do the textures for the largest elements in the room – such as walls, floors, etc.

background2

To cut an incredibly long story short, I draw each object individually in their own document, and then load it into this document and place it where it needs to be. This allows me to re-use certain bits of art in other backgrounds if it’s a larger project (windows, doors, etc.)

The last step is lighting – I use the pen tool and very large brushes to draw on the places I want more light, then I’ll blur that layer and set it on “Soft Light” mode, which is what I did to give the window area a little glow.

background3

 

Step 8: Lip syncing.

This is where those separate audio files and mouth drawings come in. I use a program called Magpie Pro 2 for lip syncing.

This is Carl’s project file. I have imported his audio file and mouth artwork, and now I can go frame by frame and match every sound he makes to the appropriate mouth position.

magpie

Magpie then lets me export this data as a keyframe information file that I can copy / paste into Adobe After effects during animation.

 

Step 9: Animation

All of the artwork and the main audio file is imported into After Effects, and set up in a new document for animation. Because I exported all of the art as separate layers (eyebrows, pupils, etc), I can animate all of these layers individually in After Effects.

This is what part of Carl looks like, keyframe-wise:

aelayers

Each of those gray box-dots to the right are “hold keyframes.” A hold keyframe is a keyframe that only moves on that point, and not in-between the points. I don’t use “tweening”, which means when you see something moving I’ve actually gone and moved it every frame, the computer didn’t fill in any of the movements for me.

Once all of the animation is done, I load the whole scene into a new composition to animate the “camera.” To do this, I simply enlarge / shrink and move the scene file when I want to zoom or pan the “camera.” The process is different for more complicated scenes, but for something as simple as Llamas it works very well.

I render this, and I’m done.

Well, that’s it.  I am obviously leaving out a great number of details, but to include them all would take way too much time. I will however be doing more blog posts in the future discussing specific aspects of cartoon production in greater detail.  I hope this has bored you all greatly.


  • Digg
  • del.icio.us
  • Facebook
  • Fark
  • NewsVine
  • StumbleUpon
  • Reddit

7 Comments

  1. Comment by Brownie — February 23, 2009 @ 6:06 am

    I honestly can’t think of an instance in this particular movie where you could have once used tweening. Other than that, guess all I have to say is that your background process is just rediculous, even if I can’t think of another way for you to do it.

  2. Comment by Devils Dog — February 23, 2009 @ 12:01 pm

    i thought it was cool, if i ever catch a leprecaun (which is unlikely, as for some reason they never be in Ireland where they should be) and get all the stuff you have, i probably could make something pretty cool…

    also i thought this post was sexy

  3. Comment by Reizza — February 23, 2009 @ 5:12 pm

    Comeplex Llamas.

  4. Comment by Christopher J Kowalski — February 23, 2009 @ 7:19 pm

    Oh my god this is great. I think I’ve emailed you five times asking about this. Thank you!

  5. Comment by dillydally — February 23, 2009 @ 7:48 pm

    Nah, I wasn’t bored at all. Now that I’m done reading it, though, I’m bored now.
    Anyway, this post cleared up quite a few things. Gracias, Jason.

  6. Comment by Captain E — March 1, 2009 @ 1:48 pm

    WAIT. Why is one of the llamas names blurred out in the script…?

  7. Comment by Me — March 5, 2009 @ 2:49 pm

    What is the whole script? x

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.