Finally....

Horses are hard.........zebras are probably one of the WORST and most challenging animals to animate. Since the previous post, I went back and redid the zebra following some live action footage as close as possible before tweaking the legs and back to suit the requirements of the film. It's turned out so much better than any of the previous tests and gallops a lot smoother.

The reason I say zebras are one of the worst animals for animation is the one thing horses don't have, which is stripes. What sucks away time isn't the fact you have to draw on the stripes of ten frames of zebra, that's easy, it's what the movement looks like when you've put all the stripes on and this is the difference between stripes that look like they move with the zebra and stripes that literally look like they have a mind of their own. A lot of R & D for me has been in this section, trying to work out how to make the stripes move with the zebra without making it look like a jittering mess of lines. I tried drawing stripes on one frame then copying and pasting the same stripes onto the other nine frames and just moving them into the places they'd be in relation to the position of the zebra but this looked like the stripes weren't connected to the zebra and didn't reflect muscle movement in the way I wanted to. I also tried keeping the stripes static to minimise any eye confusion but had the same effect. The option I had left was drawing the stripes manually by following the movement of the zebra and disregarding stripe shape, allowing them to change between frames. Although there was some improvement, the irregular shapes caused too much jitter which would put people off looking at the zebra for too long. So for the finished product below, I had to compromise between the two, limiting the stripes movement and keeping them at a similar shape, but allowing them to change. This I did by copying and pasting like I did in trial 1 but tracing over the stripes for each frame when they were in the correct place to make it a bit more natural than mechanical.

Next job is making a herd, and for most studios, it's a simple duplication and offsetting the cycles which I've done below but on playing it back, I may vary stripe shapes on a few zebra just so it looks less like a copy and paste job

Fleshing out

Now that the initial skeletal structure has been mapped out and approved, I've then proceeded to add the muscles, mane, tail and ears. Although the skeleton looked fine I did have to tweak the legs slightly on some frames as I discovered joints were too short in proportion to the figure. Unlike my first animation test, I've animated the ears flopping back and forth in accordance with the gallop motion to put a bit more character into the animal. One note to point out is the head looks slightly too mechanical, as in to say, it doesn't look like the head was re-drawn twelve times for the twelve frames which is what makes a 2D animation look more natural. This is because the head was copied and pasted twelve times and just rotated into the correct position each time. This is deliberate, not accidental because I've found in the past my horse heads increase and decrease in size as it cycles through an animation. It's just a basic guide so when I go to put the final line work over the top, I'm following the same size head each time and won't have to go back and re-size frames later on.
Next up: stripe mechanics.

Roughing it out

After discussions with project leaders, the movement has gone back to the drawing board as they desired a more zebra movement than horse movement. The difficulty presents itself when you wish to understand a zebra's movement in more depth but cannot obtain this information clearly from footage that's from a handheld camera and with no one having really done a proper zebra cycle before, it's hard to know exactly how they gallop. Even looking at The Lion King, the zebras don't really move like zebras but have more of a horse's gallop and even look more like horses.

Zebras from The Lion King

I studied live footage further and have attempted to replicate the same movement in the rough gallop cycle below. I've noticed zebras seem to almost bounce their hind legs instead of each hind leg moving independently. By this I mean once both legs are in suspension behind the back end, they seem to come together and move forward together before one leg takes the lead again.

 

Below is live footage I used to study the gallop.

Refining the gallop cycle

Using Flash this time (instead of Toon Boom) I re-did the gallop cycle by observing Spirit's gallop closer and looking to see what made the animation smooth yet fast. I noticed his head moved less compared to my zebra similarly with his back end which seems to extend and contract less than mine. Legs however, moved in a similar fashion so I can see the problem stems from the head and hind.

In this newer version I took the cycle rate from 12 to 10 pictures. The gallop is slightly faster in the updated version but this is because the original had more of a pause as the zebra is in the moment of suspension (all four feet off the ground). I think there's still more refining to be done, due to the back being too short but because my main focus for this was seeing how to make the head and hind move less, I'm not too worried about anatomical correctness.
I'm hoping the next test will be the final one however unlike a horse where their colour is consistent, a zebra has stripes so I think I'll need to look at what I'm going to call "stripe mechanics." This is because the stripes need to move in accordance with muscle movement and stripes will naturally extend and contract in relation to whether the muscle is in the extension or contraction phase. Again I'll have to deal with fixing jerky movement from the stripes as the first test did present some jitter. For now though the focus is on achieving the correct gallop.

 

Zebra Inspiration clips

 

 

 

Some might think animating an animal like this is easy as it's just a case of putting legs in the right place and the body will follow however I can say, and have the defence from the animation team at DreamWorks that created Spirit that horses are one of the most difficult animals to animate. If one leg or even a shoulder is placed slightly wrong in a frame, the whole movement is compromised. This can be seen in my test above that having the back slightly too short makes the horse look more like a dog. 

 

 

Initial designs- zebra

Because I'm a horse addict and I've been given free reign (or rein, no pun intended), I've opted to make the zebra more horse-like just for my own preference and I understand horse anatomy 100 times better than a zebra. Both aren't much different in terms of anatomical structure and I want the stampede to look more graceful and flowing, like a river which for zebra, who are slightly stockier, would look more like a river of mud.

I've been taking a lot of inspiration from Spirit: Stallion of the Cimarron and a little from the The Lion King (the little screen time zebra's get) in order to understand cycle speed according to frame rate. Obviously horses all gallop/run at different rates, some cycling through a gallop faster than others, so I have to make the decision on how fast the cycle should be.

Spirit from Spirit: Stallion of the Cimarron

 

On observing Spirit further, there are two horses that have differing gaits that gain more screen time than the secondary horses, Spirit and Rain. Spirit's gallop is exceptionally more wild and fast to reflect his wild nature, so he completes a full cycle faster than Rain (below)who has a more graceful, slower gallop and takes a few more frames to complete a cycle.

 

Rain from Spirit: Stallion of the Cimarron

The obvious difference is Spirit is male and Rain is female so a more graceful gallop for a female horse is understandable. What it comes down to is the speed needs to reflect the tone of the scene and the character, being Tilly, who is a wild child. My initial thoughts were towards a more flowing gallop and for that I thought of Rain but if I want to show that wild attitude, a gallop more like Spirit's might be best.

 

Zebra stampede scene

 

What helps in this decision is a similar scene is conveyed in Spirit where he races the eagle, beginning at a slower gallop before quickening his pace to overtake the eagle's shadow.

 

 

 

Below are my initial zebra designs as well as further development of one of the designs followed by an initial run cycle.

 

 

 

 

The cycle has been created on ones and although it depicts all the wildness and high speed attitude, I feel the movement could be more fluid. My initial reaction is there's too many frames so areas that have smaller movement are jerky however taking frames out means I'm making the movements with more impact (i.e. faster) more jerky. (Below is the movement done on twos)

 

 

I began this cycle looking at the Spirit run cycle and then drawing my own movement in relation to this.

 

Spirit gallop cycle

My cycle

 

However this proved to be not enough frames for fluid movement so I had to add more.

 

 

 

More work needs to be done to correct this issue.

 

Zebra R&D for 'The Birds.'

I've joined the third year film 'The Birds' with the task of creating a stampeding herd of zebra charging across the African plains, as they said in their words, 'because I'm quite good at quadrupeds.' I couldn't turn down the opportunity to actually be tasked with drawing a massive herd of horses but before even beginning to look at R&D of zebras I knew that horses and zebras, although from the same family, aren't totally alike. Their build and gait movement are slightly different so although I have a decades worth of experience drawing horses, I have zero experience when it comes to zebra.

Their height difference was important too as this affects build appearance and muscle structure. My initial drawings were just figuring out the specifics: what they look like, stripe configuration, ear shape and height.

 

 

I also did basic comparison drawings of a horse and a zebra so I could see how head shape and joint length varied.

Below is the rough sketch of a horse.

 

 

 

And then the zebra:

 

 

What differs the two is the shorter neck but slightly larger head on the zebra along with the stockier legs with very little difference in abdominal structure to the horse. The significant variation in both is obviously their purpose: horses were evolved by man to be ridden or pull carts etc so they have stronger muscle builds and run faster. A zebra's only purpose is to survive on the African plains so although they can't run as fast as horses, their exceptional stamina means they can run for miles without tiring. And then there's also the stripes for camouflage and confusing the predators as several zebra seem to mould into one.

 

I've played around with style but haven't settled on a definite look yet. Below are my designs which will be finalised soon with the director and producer before I look at movement.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



Blog Update

After the long summer break (and part of the start of the third year whereby getting my head to switch on again felt like trying to start a car with no will to live) I am resuming blog posts!
Before I gear the posts towards third year work and the films I'm working on and animations I'm watching, below is some commissioned work I was given over the summer, which includes painting portraits of pets as well as designing logos for two establishments.

Archie

Muma

Tor

 

 

Now I wasn't just painting commissions for people so as well as this I had been busily painting portraits of horses and ponies at the riding stables I work at so they could be turned into cards and sold as souvenirs.

 

 

And I'm just throwing in my portrait I did of my old rabbit.

 

Smudge

 

 

And then like I mentioned earlier, I was also asked to design a new logo for the stables to use on t-shirts for staff and kids to wear. The client asked for something simple as they didn't want to spend too much money on the embroidery but to just throw ideas their way and see what they thought.

 

I wanted it to have some relation to where the stables was located so close to the sea and because the stables offers beach rides I felt a horse walking amongst the waves would look appealing to the public. The model for the horse even came from one of their own so everything about the logo had some value to the client. The two versions below were turned down however due to budget and the amount of embroidery required for the logo.

Back to square 1 and an idea came from the client about incorporating the name of the stables into the logo like 'Using the legs to form the R.' I said I could try working some ideas out however I did stress that working one letter into the logo may mean losing the letter, turning 'Roylands' into 'Oylands.' However after careful designing and feedback being obtained from family and going back and forth between the client we had a logo which has now been printed on t-shirts and was worn by all members of staff and the children who helped.

 

 

 

This design work then led into me designing a logo for the farrier who visited our stables every week to shoe the horses. The brief was a bit more specific with more focus on the objects in the logo having to look real. In this case he asked for horses and horseshoes with the horse-shoes looking real with the nail holes in the correct places. Originally he asked for a cob type horse for the other part of the logo but because he also shoes race-horses as well as the standard riding horses, he was persuaded into having a more universal horse shape.

I played about with the horse design on a small scale then drew up one logo with several colour variations which I showed him and he accepted the logo on the spot alongside the colour scheme. One very happy customer!