Major Project: Animation Render Progress 1

 Today I began to render my animation. Even though a few scene still need to be animated, I thought it would be a good idea to get a head start on lighting an rendering to see how the full animation process would worked. I have done the rendering process before however don't have a ton of experience lighting a scene and selecting the correct settings. Before rendering each scene however, there were small details about the scene's background, such as the tree, ground and Japanese Palace which I still had not sorted out. I spent the time playing around with different textures in Substance painter in order to texture the ground and tree. As for the palace I had already textured it however was unsure as to whether It would be worth using in the final animation as it would drastically increase rendering times, even though it is just a background detail. 

After spending the time texturing the ground plane in Substance, I found that no matter what texture I was using, the ground would have a low resolution type of look to it when brought back into Maya. This would not normally be a massive problem as it is only in the background, however certain shots in the animation show the the floor up close (as seen in the image below). For this reason, I wanted to fix this issue as the texturing looked sloppy, which I felt lowered the overall quality of the animation. After trying several different types of grass textures and getting very similar results in Maya, I decided to review the model itself. I tried increasing the size of the models UV in hopes that this would maximise the space the texture would cover in Substance however this did not help. I also tried modelling the grass itself and by using M.A.S.H create a group of grass for the background, however I could not get the look I was going for. Whenever I would change the scale of the grass it would distort into different directions. I used a locator to control the scale and height of the grass, which seemed to solve this problem, however the other problem I encountered was that the grass was facing the wrong way. On a flat plane this would not be a problem however as the plane was bumpy to represent the hills, the grass would not fit the correct placement of the ground if I decided to rotated it after using the M.A.S.H network. No matter what I tried towards changing the direction of the original blade of grass, it would always go back to the same rotation once the M.A.S.H network was implemented. 

As I tried trouble shooting the different methods, I was becoming increasingly conscious of the time as I knew my animation would need a decent amount of time to render in order to make sure it was ready for the deadline. I decided to instead play around with the lighting to see how it effected the animation look when rendering. Once I see what this would look like, I could decided whether I liked the grass in the background. 

                  (What Characters Looked Like Without Spotlight And Directional Light Applied)


While lighting the scene I decided to reflect on what it was I was trying to show from my animation and what the best was of doing this was. My animation is highly focused on the characters and their movement. This reason ha lead me to dedicate a lot of time to their design, as well as taking the additional time I wanted to use for the rendering to smoothen out the animation. I therefore made the decision to shine a spotlight on the characters, the priority of the animation, and reduce the light on their surroundings and go with a minimal background. This way I would not waste any more time messing around with the background and instead begin the rendering process to meet the deadline. 

After comparing the two backgrounds I actually preferred how the animation looked with the nontextured background as it created a really nice contrast of the characters against the harsh white and black. This is especially important as the scenes are so fast paced, the audience needs to be able to distinguish the quick movements. As stated before this also allowed me to put the focus on the characters and animation, the part of the project I spent the most time perfecting, rather than the background which was completed in a much shorter time span. The animation also unintentionally drew a close resemblance to my drawings, which made use of black and white, which I liked the idea of. I want to explore this idea further by possibly making the Japanese Palace also resemble the drawing. 











After deciding to go in a new direction for the project I decided to review my drawings to see what other details I could carry over. I started with the stars in the sky, which I created by texturing a sphere and then applying to the sky dome. When comparing the render of the sky to the original one, I was unsure as to whether I liked the stars as they removed part of the contrast create in the original. I decided to tone down the intensity until it looked like the image below. I was still slightly unsure however it was a small detail which probably would go unnoticed as the shot happens so quickly. 


                                                                      (Sky dome Texture)





I decided to experiment even further by adding similar clouds to my drawings using the cloud brush in Substance painter. The idea was to animate the sky to move very slowly to give the illusion the clouds were moving, however just like the stars, the clouds removed the striking contrast against the characters which I liked. As well as this the shot is far too quick to even see the clouds move so the animation would not really benefit from this. I decided not to include this.




After all messing around with the sliders in the attribute editor, I decided I would go with the minimal stars for an extra detail in the background, even if it would be hard to make them out. However when I started a batch render I found that the shot was completely washed out compared to when I rendered a single frame. I was unsure why this was, until I realised the sky dome I was using was ignoring the emission map and instead lighting the entire scene. I spent a while trying to figure out why this was, but ended up finally rendering my first shot without using the sky dome. It was an annoying to spend so much time working on the sky dome texture to not use it, however it was not a priority in the animation and could prove detrimental to the completion of the project If don't start rendering. I made the decision to remove the sky dome after reviewing the animation I had animated so far and realising that the sky isn't really featured in enough shots to justify spending any more time fixing the problem. 



The final thing I textured before continuing on to the rendering stage was the tree seen in the first few shots of the animation. This was  done very quickly after finding a material which I like in Substance. 



                             

                                                                  (How it looks In Maya)


 Overall the first day of rendering was a long process, however I feel I have finalised the final direction of my project. I had chosen to spend the time I was going to use on the environment, cleaning up the animation. This choice therefore had a slight effect on the modelling of the environment, however I don't necessarily think this was a negative thing. 

Comments

Popular posts from this blog

Toolkit 2: Life Drawing 27/1/22

Soundscape Project: All Original Sounds Used To Create Final Sounds

Introduction To Maya: Camera Movements