Upon starting this project, I changed my mind many many times about the story I wanted to tell. There was only one thing I was sure about: that I wanted to use what I had learned in 3D during this program and more.
My first love has always been drawing, and I cannot quantify how much I learned from Shaun and Steve in this area since I started, however, they also told me to take risks and experiment. I cannot see a better time than now. I have a great support system in terms of tutors and mentors and institutions.
Secondly I know I wanted something very personal. I have always been a person that likes to laugh and share the joy around me, I wanted my film to transmit that. In such a world that we live today, it is so beautiful to just forget about it and laugh, even for a second. Animation has always done that for me, that is what attracted me to it.
After agreeing on that, I wondered which character could best represent this trait in my personality. As long as I can remember, we had had cats at home. Dogs too, however, cats have always been the ones related to the funniest stories, the ones that cheer you up in dark times. I cannot remember how many times I said out loud how I wish I were a cat, with no trouble and worries but just eating and sleeping all day. Because that was essentially what all my cats were doing. I always had lazy cats whom would just move to get some food. I weirdly related to that.
That is how the rough synopsis came out. I wanted to establish one strong character, a cat, with so much personality he would be the story. I wanted him to resemble me, in its flaws mostly. I wanted him to be lazy, fat, annoying and loving food but also stubborn and determined. A cat who has an idea and does not give up. A metaphor for how I never give up on what I want. A metaphor on how I got into this program in the first place.
I cam up with what may seem a simple problem that would be meaning but also joy to the piece. This cat is bored and he is dreaming to be something more, to do more (just like me), he may just be too lazy for it. But then his owner brings him a challenge, a box. This cat is wondered at this box. What is it? Do I fit in it? Should I try? Why is it so hard to try and fit in it?
The cat tries every way to fit in this box but he is too fat. I found that also very poetic. We all try to fit in a box after all. Finally, after twisting his body hard, he fits in. He is proud. But so much of him is covered. He is stuck.
Then the owner comes back and puts a plate a food down in front of him. The cats wants to get to it, but he cannot move out of the box. He tries and tries but nothing gives. So he starts giving up, maybe this box will hold him until the end. Nevertheless, the cat thinks about his inspiration, the tiger, the one he wants to be and in his last attempt to get out, breaks the box, completely breaks free of it…. to land face first in the cat food thus sending it all through the window.
I thought this simple story was holding so much of me and what I believe in , I wanted to pursue on it.
I consulted my tutors and mentors and them gave me advice, on the storytelling, the 3D process and the character. Most told me to look at Simon’s cat for reference, especially in terms of timing and behavioral animation. Simon’s cat are little animation episodes about the same fat cat and his owner Simon. The stories are usually hilarious and effective.
I then researched more into strong cat characters in animation such as Lucifer in the 1950 Disney’s Cinderella, or more recently Puss in Boots in Dreamworks’ Shrek or Chloe in the Secret Life of Pets. I looked at a character design book and magazines I own from 3D total publishing.
I led extensive research in cat’s anatomy as well as pictures and videos of cat expressions, manners, habits and videos of them climbing into boxes,… Thankfully I also happen to have a fat cat at home, which turned every moment of me seeing him a study session. I would be moving his legs around to figure out the degree of movement or stretch his face to see what expression it could shape.
I also looked at cat 3D rigs and models to get an idea of the sculpt I wanted to make.
Upon starting, I had to look at tutorials online, mostly on Lynda.com and consult google constantly to start shaping the 3D sculpt. I started weeks early since I knew it would take me longer due to my inexperience. Shaping the main character was fun as well as learning a new software actually. Zbrush offers an interface that allows an artist to actually sculpt without worrying about the coding.
I presented this to my mentor as well as a lady in the industry who has given me great feedback since I started. When sent in, she sent me back this screenshot of my model, with all these notes to make him more balanced. So I created a new model respecting her indications. Then I compared both versions. The one she suggested looked more natural, however I was biased the first version had more personality, with a bottom so big and far back, it looked like the cat would tip. I was so torn I made a poll on social media to ask people to help me choose. The first version was very tightly declared winner.
The harder step was to move the sculpt from an artistic software to Maya, doing the retopology was quite a challenge since, in order to make the sculpt lighter in terms of polygons to reduce the file size while animating , you have to create a 3d grid around your sculpt. I have now been on the retopology for a week, reworking the polygons so nothing goes wrong later on. This step has taken me longer than it should, however that is why I wanted to start early and get as much advice as possible. I have been talking through the issues I could run into while modeling, rigging and animation with Kevin Rowe, our 3D tutor to get a head start in researching and listing those issues. There are three main challenges to this animation: the way the head and body are going to squash, stretch and deform in order to accomplish a cartoony animation, the way a packing peanut is going to have to be stuck to one of the cat’s paws and how spatially accurate the animation needs to be. Thankfully, Kevin gave me advice on these issues.
Moreover, talking through my animatic with Steve Roberts, Sarah Woolner and Clare Murphy allowed me to improve the overall story and rythm.
The retopology was quite the challenge and I spent most of the time correcting my mistakes. Everytime I would fix something, something else went wrong. I finally made a proper mesh with less polygons and go on to the UV mapping. Another challenge, everything is new, so I make mistakes, however this character cannot have any mistakes on it or I will not be able to properly animate him. After many trials, I finally was able to export it on a new software, substance painter, to paint the model with a bit of texture on. However, another problem of exporting and importing is the loss of data, which I ran into many times. The problem is usually the compatibility between the two softwares which can result in a loss of texture.
Upon starting the rig, I knew this would be the most difficult part of the process. The rig allows all the parts of your character to move properly. The more the character is cartoony and exaggerated, the harder it is. My character is basically a mix of 2D and 3D. I wanted a character that used all the personality and characteristics of 2D animation into a 3D environment.
My character has a lot of animation that resides in the squash and stretch and extreme actions and faces, typical of 2D animation. So when making a rig for it, the software basics will not understand what you are trying to do. I would have to understand exactly how the cat’s body works and reacts and moves but I also need to know Maya perfectly to make this happen. There are so many options and ways in Maya, and this being my first time, I am having quite the trouble. On recommendation, I used Advanced Skeleton, a plugin that allows you to make a rig easily. The software is great if I were using a normal cat, however, mine is very different. A very expressive face, a fat body on very small legs, that is the source of many problems for a rig. The only thing to do is ask for advice and be patient and meticulous. I have asked the software team as well as my mentor for options in my case. I have been on this rig for 4 days now and have not found a solution. However I do not despair, the more I try, the more I understand the issues and the what the body needs.
I am a very stressed person so failing so many times is quite a toll on me. I am also not really the tech person so this is really a personal challenge. What I would do to get it through is usually work on another task for a day so my mind can breathe. For instance I was able to correct my animatic that way and build the room in 3D.
Building the room in 3D was a fun breath from what I was doing because the room I designed is very simple with few elements. It also gave me the opportunity to experience with modeling and textures in Maya and Substance Painter which is a 3D texture software.
However now I am back to rigging and hopefully will find what works for my character.
After a lot of fails, I finally have my body rig as well as solutions to my body rig problems. I used twist/bendy and free orient options for my joints and I will paint the weights on the legs to avoid further deformation. Since the legs are so small, the deformation needs to be exactly the size of the joints. Moreover, on my mentor’s advice, I will add a squash nonlinear deformer on the body which will allow it to squash and stretch properly.
I finally got on the most dreaded part of the animation: the facial rig. With my animation relying heavily on facial animation, I had to take this part very seriously. Once again, many versions saw the light. In the beginning, for about a week, the facial rig was appearing on the wrong part of the body. It took me many days to understand the body and the skeleton were not aligned properly for Maya. The cat’s body was on 0 on the x axis while the rig was at -180 on the same axis. Everything needed to be at zero. I finally found a way to align them and started on experimenting on the facial rig. A facial rig is a lot of points that need to be aligned properly with the right stretch/squash and deformation for it to work. It can get quite confusing and overwhelming.After two weeks of trying to work out the Advanced Skeleton facial rig, I understood my character was somewhat too cartoony to properly work with this plugin. It still got me help on the rig, however, I cannot keep wasting time trying to make it work. I am going to have to work with what I have. Tomorrow is the official day when it is recommended to start animating. And while I do know animating in 3D should require less time than 2D, I would like to follow the schedule provided by the tutors. My character is not quite ready yet, I believe I will need one more week to finish it. I now realize it will not be perfect, and that is frightening, I put a lot of pressure for myself on this project, but the control freak in me needs to keep on moving to deliver a complete animation. I am now sculpting my own blend shapes, a series of 3D models with different expressions. This will allow me to help with the mistakes on the facial rig. It also allows for more extreme expressions. The problem is it takes quite a lot of time, nevertheless, it is relaxing to model for a change. After that I will give a last attempt trying to fix the mistakes on the facial rig. I have rigged the ears and need to rig the tongue for some of my scenes which should not take long. Finally I will experiment with non linear deformers and fur textures.
Then I will move on to animation. The room for my animation is ready so I will only need to place the lights and decide how I want the rendering to be. From what I heard, rendering takes quite a lot of time, so I plan to do it scene by scene. There is a lot more I am going to discover along the way I am sure, but I am happy to get to and end for the rig. It has been a lot of work and downs and time, about two months now trying to figure it all out. I have learned immensely during that time, people do wonder why I did not take a model ready or hire a rigger, however tome it was important to go through all these steps to truly learn about the 3D process.
Upon starting the animation, I realized my topology and rig were all wrong. I went back all the way to the beginning and restarted on the topology. I made sure everything was even and smooth. Later on, I reworked on the rig. I finally understood what the problem was. Because the cat I created is very cartoony, it should not move lie a cat. Its legs are short and straight. So I created my body rig using a blend of cat and human skeleton. While the spin works like a cat’s one, the legs work like a human’s. In addition to that I added a freedom of orient on the knees to allow for more motion.
When this was figured out I moved onto the facial rig. I wanted to do this without using an automatic rigger. I experimented with soft mod controls, no controls and finally chose to manually place the joints on the face, parent them to the head joint and use nurbs curves as controls for the animation later on. Moreover, the first year student helping me for two weeks will create the blend shapes we agreed on to complete the facial expression palette.
During this time, we were introduced to the Royal College of Music students who will compose our music. This meeting was hugely beneficial and allowed me to define a bit more the tone of the animation. The students all had the idea to attach a melody to the main character to add more personality to it.
Later on, we were introduced to our assigned first year students. I was very excited about the possibility of working someone’s strengths into the project but also hopefully teaching one or two things. It started really well until the student I was assigned to was reluctant to helping because of his assignments. In the end, no help was given and I redid the blend shapes myself as well as experimenting with the fur. I completed the facial rig and tomorrow will be my second trial at animating in 3D with the new version.
3D animation revealed to be more of a challenge than expected. The whole process of animation has to be learnt again from zero with a different perspective. I now animate with controls and less freedom. Yes the software will in-between, however it will probably not do exactly what I want. The difficulty moreover is working with a rig that is not 100% correct. Maya does not realize or account for those mistakes. While I used to freely draw extreme keys, then do my breakdowns and in-between and later on easily correct my timing or movements, now the process is completely different. I use auto key to automatically save the movements as soon as I choose a control, however, many times I have found one had not been keyed. Moreover, the more you add keys in Maya, the worst it is, so all you want to have really are strong keys and in-betweens with an ease in and ease out if needed. Because Maya adds automatic curve to it, you might find yourself with difficulties you would not have in 2D. For instance, I downloaded a female rig for my minor second character. When using this rig, I realized there was a problem on the knees. The deformation was also impossible to correct since the rig was not mine. This you cannot really work out as adding more in-betweens to try and correct it will look ragged and incorrect. Another problem I had with my rig is that the eyes, tongue and teeth were not attached properly to the mesh. It meant extra keys for the teeth and tongue, which is a problem I can handle, just requiring more time and attention. The eyes however were acting like they were not part of the body. Thankfully this could work thanks to the cartoonish style of the film but presented me with a lot of challenges.
The animation is moving along slowly, slower than what I am used to. However, practice makes it easier and easier everyday. I finally understood I need to quickly animate a first version with no fuss before I can go back and refine the timing, poses and camera angles. This is important to finish the project in time but also to have a version to present to the composer early to get the tune in accordance to the animation.
3d animation feels as if learning animation all over again. I expected it to be similar in some ways, however the more I go on, the more I understand how completely different it is. I move controllers one by one, making sure to key all of them properly in accordance to the others. I am mostly experimenting with the timing since I found out it is completely different from 2d. Everytime I think about a 2d timing, it turns out too fast in 3d. What I do is I put two keys, put a breakdown and see how it works. If it is too fast, then I space the keys more until I am satisfied. It is very much discovering and exploring 3d, which means the more I go, the better the animation gets. However, the beginning of my animation is sloppy and I would like to go back and correct it if I have time. Another thing with 3d I learnt is you need to go fast and finish a basic version before going back and fixing it. In the beginning, I would just stay hung up on a scene for more than a week trying to figure out the subtleties. After loosing a lot of time I understood the way to do 3d is work on a basic version of your keys, breakdowns, timing and spacing, then put everything together and correct camera angles, transitions and animation.
I have finished a rough cut of the entire piece. It is now five minutes. From what I can see this is because some of my scenes are too slow. When animating on Maya, the only way to see the real speed is to play your animation on playblast. If you play it on your animation timeline, the result will usually be faster. This is something I did not realize until later on, explaining why the first scenes are very slow, and others are very fast. The other big problem I noticed are transitions. My change from one scene to another are not smooth. Moreover, some scenes I feel are not necessary so will need to be cut out. It feels like a gigantic task to do in only a week since these are little things here and there to be take care of. I have decided to proceed two scenes by two scenes. Checking the timing, cutting if necessary, correcting the animation, camera changes and transitions.
The editing is a very important part of 3d. To make my animation more effective, I have cut from 5 minutes to 3. I took off everything that was not working and fastened the pace so the comedy is snappier.
Checking every scene and correcting the animation is also an essential step to the 3d process. While I correct every scene, I work with my composers to create a soundtrack that will define “Fat Ass Harold”. Hand in hand with the sound designer, we are working all together to get the perfect sound to enhance the animation.