Major Project Submission Post: Reflective Statement + Art Of

 Reflective Statement For Major Project

Overall I am really pleased with the outcome of the animation. This is mainly due to me feeling that I was able to include everything I set out to do, all the way from the beginning of Premise. It has been a very long process, with different routes taken in order to bring me to the final product. At times I felt I had set myself a far too large goal to achieve and was unsure that I was able to complete the project within the timeframe. However I kept going and I am proud to say I completed everything to the best of my ability. I believe the key to succeeding in the completion of this project was good time management. Had I not completed certain section in time for the next then I would have run out of time nearer the end of the project. 

At the end of the Minor project, I had completed the redesign of my animation story and characters and applied my ideas to create a Pre-viz which depicted the events of the second iteration of the animatic. I then set myself the goal of completing the texturing, rigging/skinning, final modelling of the environment, animating, lighting and editing for the Major project. I had a lot to complete however had a solid foundation to continue from, having modelled the final version of both the characters and a clear guide of how to animate thanks to the detailed second version of the animatic.

Final Animation

UV + Texturing

I found that I really enjoyed the texturing process, once I had got over the hurdle that was UVi'ng. On paper I understood how to UV, however applying the processes to your own model is a completely different thing. However being at the very start of the project and knowing how much I needed to do, I did not waste time in learning methods to completing the UV section. I have definitely grown more confident with the process of UV unwrapping as I have managed to establish a workflow which I found worked for me. I tried my best to complete a UV unwrapping with as little cuts and distortion as possible, and for the most part feel I have achieve this.

I was then able to move on to texturing which I really enjoyed doing for both of my characters. It was really fun to try different combination of materials together in order to see what worked and what didn't. Having the characters all in one Substance scene helped a lot with the process as I meant I could make very quick changes based on how the materials interacted with each other. This is the part of the unit which I was I was able to spend more time in just playing around with different objects and seeing what could be made. It was really satisfying to find a material meant for something else and then repurposing it to fit with your own personal needs. A good example is this was the Samurai's hair which is made from 'dog fur'. By rotating the texture around and shrinking it I was able to create hair which worked well for my characters design. Extra details like this really elevated my original ideas for the characters and I would have like to see how far this process could have gone if I had chosen a more texture focused project.

Rigging + Skinning

From the start of Premise, this was the part of the project I was not looking foreword to completing. I was very aware of how much time the rigging process took up, with it being one of, if not, the most time I spent on in Major. This was due to not only having to create a rig for a character, but two, which were good enough to be used for a highly animated animation. In order to complete this section of the project, I heavily referenced the Jetpack Jones video tutorial series. This was essential for me to completing both characters in time.

The process as a whole was strange as it was stressful, however I knew as long as I followed the videos, the characters would work as intended. However it was still the unknowing factor as to whether I did something wrong, which could potentially cause me to have to start a section over again. This section was mainly hard mentally as I knew whatever action I had taken in completing the rig for one character, I would have to go back and do the exact same thing for the other. I had to therefore establish a strict workflow, which forced me to complete a section of a video and then instantly go back to the other model and do the exact same. This was fine for the most part, however got tough for sections such as rigging the arms and legs, where there is not mirror function. This meant I had to essentially do the same thing 4 times in a row. With that being said, the positive outcome of all of this is that I understand the processes a lot more clearly. After completing part of the video for one character, I challenged myself to go back and try to complete the same section for the other character without looking. This forced me to actually applied what I learned, helping me to actually understand why I was doing specific things. It got to the point near the end of rigging the control rigs that I was doing it without the video because it was quicker and I understood what I was doing. The final few videos I completed only had to be done for the Samurai character as the Cyberpunk character did not have eyes or eyebrows. I am so glad that I thought ahead all the way in Premise not to include a mouth on either character and to limit the eye movement to 1 character.

Animating

I was eager to begin the animation section of the project, however at the same time worried as I did not know how to go about it. I didn't have much experience animating prior to this project, with the only other time being in the group project. I purposefully chose an animation heavy project for Major as this is something I wanted to focus on improving. It was a very ambitious thing to set myself, however I feel I have completed it to a decent quality considering the time frame.

At first the only real guide I had was my Pre-viz and animatic, which did well depicting the basic events of the animation. However I quickly learnt from completing the first shot that translating a drawing into 3D is a completely different process. As the Pre-viz is very slow due to its limited animation, I failed to see the problems with the animatic up until the final animation section began. I quickly learnt that a drawing which can last 6 seconds can easily only take about 1 second once its animated properly. This is even more true for a fight scene due to its fast paced nature. I found that if I was going to follow the pre-viz shot by shot, I would be left with a very quick and uninteresting fight scene. This went against the point of the project, with me choosing to do a fight scene as they are normally the best part of the film. I needed to fix this problem however was running out of time.

I decided to refer back to videos online from films and animation which depicted fight scenes. I did my best to see how they portrayed a fight scene in an interesting way. My main focus was getting the timing right as this is something I previously failed to do. As I got more familiar which what to do, I began taking ideas from shots I liked the most, trying to understand what made them so interesting. I then applied the ideas to my shots as I was animating. them as I did not have time to do another pre-viz. This resulted in an animation which has completely different choreography to what I initially set, however this is for the betterment of the animation. It wasn't until I got to the end of the process that I realised how much of my original ideas had been altered to fit the new version of the animation. 

A good example of a shot which had been given a much needed lift from the pre-viz version, is the scene where the Cyberpunk character face is first revealed. Instead of simply crossing to the the side of the bridge and pushing the Samurai away, I decided to add in a dramatic shot of it flying closer to the camera. I really like the idea of this shot as it really conveys the over-the-top-ness of the character. It shows the audience there is a clear different in power between the two characters and emphasis the Cyberpunk character is something other than just human. More than anything though, it is simply cooler to look at. This became the philosophy when animating. Does it look cool? Would the audience appreciate it? If the answer was 'no' I sought out ways to improve the shot. Changing a single shot however naturally created a knock on effect on the shot which preceded it. If a character is not longer standing in the same position which flows to the next shot, I then had to come up with a reason as to what happens between the two shots to get the characters to be where they should. This is seen very clearly after the Cyberpunk character lands. As the character was not long on the right side of the map to make the shot which followed it make sense (shot where Samurai dodges under his weapon), I therefore had to add in the whole section where the characters cross past each other. This added a lot of additional work to the animation process however it was worth it if it meant I could keep in the shots which were more entertaining. 

I feel that my animation (movement) is as good as the time limit I had allowed me, however there are still parts in which I would have liked to improve upon. I learnt about auto framing quite late into the process, which made my life so much easier for adding in the secondary movements. I managed to get two iteration of the the animation done with improved movement, however if I had the time, I would do this again. I feel this is always inevitable within the animation industry, so I am proud of what I was able to achieve given the time. I feel like my animation, regardless of the quality, brings a lot of interesting concepts to the fight scene. I did not what the characters to simply swing their weapons at each other for a few frames and then end it. I instead chose to fill the time by displaying the range of fighting abilities the two characters have gained in their years of fighting. The animation includes swings, jumps, rolls, running, ducking and takedowns. I used a range of compositional techniques in order to showcase such actions. I also included a narrative which I told through the characters actions as they fight over the item. From Premise I like the idea of having shots which showed the item being thrown around midway through the battle and characters stumbling to prevent it from landing on the floor and breaking. When I re-watched my Pre-viz I was disappointed by the lack of this idea and so made the changes in the second section of the animation to showcase this idea.

One of my favourite shots I animated was when the Samurai tackles the Cyberpunk character, who falls to the ground however managed to instantly recover. I like this shot as I feel it encapsulates my main goal for the animation to create something which is visually entertaining. I love how its such an over-the-top action, which really showcases the skill of the character. It also transitions flawlessly into the next action of the animation. I hoped to create a fight scene which seamlessly flowed to the next shot while offering entertainment, and feel it is shown best here.   

Lighting + Rendering

I began animating at the beginning of April, hoping that I would be finished by the end of the month, giving me 3 weeks to render. It ended up taking all the way up till around the 6th of May to finish. Although not too far off, time is essential when it comes to rendering. However after getting a decent amount of the animation done, I decided that I would rather spend more time having cleaner animation than I would on the environment. I therefore used the time I would have had spent adding in additional little detail to the environment on animating the characters. The small details would have been a nice touch, however at the end of the day I set out to complete a character project which main focus is animating. After spending the majority of the project battling with the rigging, I wanted to properly show my hard work off. The best way I could do this was through the lighting, which I would focus all of the attention on the characters. This was not out of place in the world of the animation either as it was intended to be set at night. I used a spotlight and directional light in order to light each scene, in which the character was the main focus of the shot. I was originally planning on adding grass however I felt that it took away from the characters as the white of the ground create a nice contrast against the characters.

As I got further along in the project, I decided to make the directional light orange in order to represent the Palace which was on fire for the backdrop of the scene. This was a nice touch and switched up the colour scheme of the animation, distinguishing it from the first half. I found the rendering side to be stressful as I was unsure whether I had left enough time at the end for the animation to be fully rendered, however this turned out something not to worry about. However at the time it added another layer of stress while working. I rendered the animation with Tiff files and set the settings to 4, 3, 3, 3, 0 , 0. I wanted the animation to look the best it could while also having settings within reason. I was able to bump the setting up more as there wasnt too much geometry in my scenes. Its not until I got to the second half of the animation with the Japanese palace, that the render began to struggle. This left me with less time to do the art of however I feel it was worth it as I finished the animation in time.

I then imported the animation into Premiere Pro where I added the final touches. I used Audition in order to produce the sounds.

Problems/Challenges + How I Overcame Them

Eyebrow rig and eyelids rig

The process of actually rigging the eye brows and eyelids was relatively simple and mainly required time to spend on using the brush to make the skin naturally fall into a natural position. However upon completing both processes I found that when the character head would rotate, the eyebrows and eyes would move in way which different from the rest of the characters body. This was a big problem for the animation as it was essential for the character to move his eyebrows to express his emotions. It would also be impractical if his neck did not move. After explaining the problem to my tutor is turned out the reason for this was due to the original head model not facing directly foreword in the viewport. Therefor when I duplicated the head to use as a blenshape, I had to rotate it slightly so that the mirror skin weights tool would work. However by rotating the head slightly, it meant that when the blend shape was connected in it would rotate along the axis of the rotated duplicate head. In order to fix this the original head had to be unbound from the the rig and rotated into the correct position. After that it needed its transformations cleared. I then had to go back through and remake the eyebrows and eyelid rig. 

                                       

Character Blenshapes

This was a hard process within the skinning, mainly for the Cyberpunk character. As the characters shoulders are made out of tubes, they would distort in strange ways when the arm was moved. I attempted to help the situation as much as I could in the skinning process, however there was only so much I could do. It is a very noticeable part of the character and so I was worried that it could not be fixed and would look strange for the duration of the animation. However I was able to fix the problem by using corrective blenshapes. 

Render Crashing

This seemed to be a bigger problem at the beginning of the rendering process and I am unsure as to why it stopped. However the first few scenes which I animated would crash, therefore wasting time if I left them on over night. Luckily this did not end up happening for most scenes as I got further on with the project. I did however have a lot of trouble with the scene in which the Japanese Palace is on fire. I was aware that by making the fire effect so big it would significantly impact the render times. However I liked the shot so much that I wanted to still include it in the animation. I ended up setting the render up for 100 frames at a time wit its settings on 2, 2, 2, 0, 0. Each frame would still take 3 - 6 minutes and there were 300 frames. I just left it on overnight where I wasn't using my pc. However  If the scene was not limited to 100 frames it would just stop rendering. It did not crash it just stayed still on a single frame. I expect this was due to being overloaded with information, however I thought it was strange that it would just stop. 

Samurai Weapon

The idea behind giving both the character a different type of weapon was to grant me the freedom to explore different fighting styles. I do not regret doing so, however it was significantly harder to make the Samurai look like he was attacking with his weapon. This was due to whatever hand was holding the to of the weapon not being able to reach the weapon if it was swinging forward. This became challenging with certain shots as I need to show that he was using his weapon to attack. In order to fix this I would usually angle the camera in a way which prevents the audience from fully seeing both hands at one time for shot such as this. This definitely helped, however I feel like there are some parts where his pose is awkward as a result. 

Samurai Legs

The Samurais legs often look awkward in the animation due to the way I bound the foot geometry to the skeleton joint. I tried correcting this with blend shapes however the geometry was too thin to really do much without it looking odd and I had already gotten too far into the rigging process to go back. This is a small thing which annoyed me about the character, however I don't think it takes anything away from the animation. I feel like this sort of thing is inevitable when your juggling creating two characters at the same time. 

Environment 

I initially felt that my environment was lacking, however after making the decision to move on from it and focusing on the character, I came to realise that the whole second section has got a significant background. The Japanese Palace is a much more substantial environment then what I was planning on doing for the first half, which is important as its a key element to the story. It wasn't until I got to that part of the animation that I realised this achievement and now feel no need to worry over not having grass and a few buildings to the background. The buildings may have looked nice, but overall add very little to the final product as it is a fast paste fight scene. Removing the grass was a creative decision as I decided that the animation looked better without it. 

Lack Of Computers

A clear problem which effected the whole course was the inability to render on the computers due to the screen timeout preventing renders. Although this problem was resolved it was done significantly closer to the deadline and I would not have been able to render my entire animation that way. I instead got a head start and began rendering my scenes at home which turned out effective enough that I did not have to use the University computers when they were fixed. I did however effect how I was able to work as I was unable to do anything on my own computer while a scene was rendering. If I had the opportunity to render at University. I would have maybe spent more time animating scenes or focusing more on the art of. 

Art Of