My street

On the 1st of March 2012 I begun the daunting task of creating the sets and models for all the scenes for the film.

The first scene was set in Sam's street.  At first I'd envisioned this to be quite a pretty street with your typical cartoon suburban  look.  Perfect houses with large open drives.  When I thought about it a bit more it seemed less and less logical to do this.
Sam is meant to be in his early twenties, so probably still at Uni, or if not he's just started working and I'd imagine doesn't have a lot of money. After all he is using a bag that he must have had for quite some time.
So it made more sense to have the street more run down with smaller houses.  Luckily I don't have a lot of money and so my street fits this image perfectly. 

I grabbed my camera and went outside to take lots of pictures of the houses around this area.  I did actually have a woman come running out of her house at one point asking why I was taking pictures of her drive.  I think she was worried I was some sort of debt collector.  Once I explained that I was making an animation and needed reference she seemed to calm down.

Like I mentioned in an earlier post, it's always worth having reference.  No matter how well you think you know something, taking reference will always give you those extra pieces of information that you'd never think of.

Below is some photos I took of the area.  The final film has very similar houses, although I made a few things slightly oversized so that the environment would suit the characters better.  So things like the windows and doors were bigger then what they're really like in reality.

It's ALIVE!!!

I completed the bag model at the end of February 2012.  Although in terms of modeling and texturing it was complete, there was still one more process and that was to give it the controls to allow me to animate.

This process is technically called "rigging" although in plain English it basically means to add controls that allow you to move you characters arms and feet for example.
You build a skeleton for your character, very similar to how skeletons in real humans or animals work.
Once the skeleton is built you have to tell each bone what part of the model it should control. So for example you've added an upper arm bone (Humerus) to your character.  You'd want this bone to just move the upper arm section of that character.
This part of the process is called skinning.  You're attaching the polygons,  (see previous blog for explanation of what a polygon is)  to the bone.

Finally you need to add some user friendly controls that the animator can select to move all the bones and skin (polygons) around.  I guess you could compare the controls to that of a string puppet's wooden handle. And the strings and hooks that attach it to the puppet as the underlying bones and code that allow the software to know what moves what.
So after that lengthy explanation I hope you have a slightly clearer idea of what rigging is and why it's needed.

Having said all that, I'm actually not very good at rigging and I knew to have an easy time controlling the bag I'd need to give it to someone that knew how to rig well.
This was actually a worrying time because I didn't know any good riggers that were free. It wasn't in till mid April that I managed to get hold of a friend of mine, Jim Ovnik.  He's one of those people that just seems to be good at everything. Not only is he a great animator but he really knows his technical side.
He was pretty busy with lots of other things but he really pulled it out of the bag (sorry bad pun) to get me a working rig.
While I was waiting I got on with other animation that I knew didn't involve the bag.
Things like close ups of Sams face, or the end sequence with the evil new bag.

It wasn't till around June the 15th that I finally had a finished animate-able bag.  This only gave me around 20 days to get all the shots with the bag  finished, so as you can probably imagine this was quite a panicky time, especially considering the straps on the bag had to all be animated by hand!

Below is a video demonstrating some of the bags functions and a look at how the rig works.   Sorry, for some reason I decided to commentate this with my most boring voice. I promise I'll try to make a more upbeat one next time :)

Building the bag,

Right from the beginning I had a rough idea of what I wanted  the bag to look like. I knew it had to look ragged and used. But I also wanted it to have cute appearance.
First thing I did was to look at lots of reference of real suitcases.  I was trying to find shapes or parts of a case like zips or pockets that could be used to make up a face. Similar to how cars look like they have faces with their grills and headlights.
It was actually quite difficult to do. Not only did all the cases I found not have much similarity to real faces but I also realised that the pockets and straps had to be positioned just right to allow a lot of the scenes to work.
For example the main case's flap lid on the front is completely made up. I didn't come across one piece of luggage that actually has that kind of design, but I knew I needed something like that, to not only allow all the clothes to spill out as the bag flew through the air, but also the bag needed something that could pass as eyes.
Having big buttons for eyes seemed like the perfect solution but obviously buttons have to have a function, thus the reason for the flap like lid.
The buttons went through a small iteration.   Originally I made them based on real sizes of buttons.  It wasn't long before I realised they just weren't cute enough.
It's a well known fact that big eyes make cuter characters. Apparently babies respond better and seem happier around people with big eyes. I guess sub-consciously this stays with us through to  adulthood?
With this in mind I decided to make the buttons over sized.

Here is a few basic sketches of some of the facial expressions that I wanted the bag be able to do.
I really liked the design of the sorting hat in Harry Potter.  The way the crinkles in the material made up the face.  I decided that I'd use similar techniques for the bag and the way it would frown or look surprised. 

Below you can see a video of the process of modelling the bag.
It started as a simple box, using a package called Zbrush I sculpted it into the final shape that I wanted.
You'll notice around half way through the video that it appears with lots of lines on it.  I used this technique to help with a process called retopology.
Basically all 3D objects are made up of lots of planes called Polygons.  Because these are generally square in shape they don't always bend too well. When scuplting these polygons can end up in all shapes and sizes and aren't always best laid out for when it comes to animating.
By laying them out in such a way that they form nice loops you can get them to bend without causing unwanted distortion.  Drawing out how I wanted these polygons to lay, before actually laying them down allowed me a quick way to see what would work and what wouldn't.
Once I was satisfied with the way the lines flowed I then created a new set of polygons that followed the lined pattern, thus creating a new version of the model with the polygons flowing nicely.

Zbrush is a great package for sculpting.  Although it is getting better at creating more hard surface models like vehicles and buildings I still find it easier to use a more traditional 3D package to complete areas like the wheels and handle.   I used a package called Maya, I also used Maya for all the animation that you see within the film

Basic pitch

I should of mentioned this before the storyboards and pre-vis but actually forgot that I did this process.
The basic pitch is a good way of getting your story idea across.  It usually consists of your main story points laid out on a board similar to a comic book. It doesn't have to be works of art...as you can tell as long as it can give people an idea of what you're saying.   Unlike Storyboards you don't need to worry about every camera angle, only the ones that move the story forward.

I started this with the intent of showing it to friends and family, so that they'd have a better idea of what I was on about when I kept telling them that I wanted to tell a story about an old bag.

I didn't get very far,  to be honest it got to the point where I decided it was quicker for me to just storyboard everything then show it to people. If there were things that didn't work I could then change them in the storyboard.
But if you are trying to get funding or trying to pitch your idea to colleagues, then this is a good way to go.

Another area I skipped simply because I was running out of time having only four months left to complete the whole project, was mood boards and concept art.  These are great to get an idea of what the final image could look like.  It gives the artists a good idea of what to aim for when creating the worlds and colour palettes that are within the film.
I had a pretty good idea of what I wanted in my head and seeing as there was no one else involved I decided to skip this process.

Having said that I always use reference. It's never a good idea to go into any part of the process blind, be it modelling an airport or animating someone lifting something. Always have reference at hand, even if it is just one photo of one small video clip.

Pre-visulization

Something that in the last ten or so years has become common place amongst major movies, is something called Pre visualization. In the old days, before computers or any of that fancy technical stuff,story boards were really the only way to get an idea of the camera angles and field of view/lenses that should be used for the movie. 
These are still used today as it's still a lot faster to draw a sketch then it is to model something in the computer. Even simple boxes still take longer to put together then a stick man.

However one thing that storyboards lack, is motion.  Arrows or written descriptions have to be used to get across things like the motion of the camera and characters. Terms were invented to quickly describe these actions.  For example POV (point of view) or CU (close up).  Although these help the director and anyone else involved, it doesn't tell you much about how long a shot should be,  or if the camera starts off locked but then starts to Dollie, how long is it locked off for?  And when it comes to characters positions in the scene, it becomes even more vague.

So this is where the Pre Visualization comes into play.  Most of the time these are created on the computer.  But this isn't always the case.  
Way back when they were creating Star Wars Return of the Jedi they didn't have computers powerful enough to quickly create scenes that could be iterated, so they had to use models. 
 One famous scene that used this technique was the Speeder bike sequence. They simply used toys and cardboard trees to create the scene. By sticking small camera on the bikes and moving them around by hand, they could get a pretty good idea of what the final scene would look like.
Computers have now become so fast that studios can quickly produce pre-vis and change it on the fly. 

Below is the pre-vis for baggage. Most of these shots remain pretty similar in the final version.  One of the biggest differences was the beginning. I  removed this, as I decided it worked better not knowing what was causing him so much grief at the beginning.

Live action animatic

I filmed this back at the end of January 2012.  After finishing the storyboards and before producing an animated pre-vis I wanted a quick way of knowing roughly how long the film would be in relation to what I'd created in the storyboard.

The time limit was 2:20 so a very short time.  Creating this live action video made me realise how far off I was in cutting my film down to the 2:20 mark.

I'd already decided to cut out the scene with the granny and the escalators. But as you can tell by this videos length, it was still well over 2:20 

I did about 4-5 different edits of this to try and cut it down to 2:20. The first of which took everything that I felt was good and left a really basic story. This was a good way of really working out which parts moved the story forward and which parts were just fluff.

Generally I find that comical animation is usually faster then live action so I knew that if I could get this to the 2:20 mark then I could get the animation to fit.