Evaluation: Look Development

Shot Development

Look Development

December

As part of the Animation Process and Practice unit we had to take on the task of ‘Look Development’. This involved us assembling in groups to collaborate and produce a final composition inspired by an original source. Using the knowledge we had gained from the previous unit, we would take the technical aspects further to produce the final piece.

As I was in the 3D group for the character performance piece, I chose to continue in the 3D medium. My partner Siobhan also chose 3D before and continued with the medium in this project. We chose our group because we believed our combined strengths would produce an effective result. We understood our strengths and weaknesses and used this knowledge to assign our roles within the group.

Siobhan’s strength lies within pre-production/production including concepts, modelling and texturing whereas I discovered my strengths lie within production/ post-production during the technical phases such as animation, lighting and compositing. I only discovered this late in the project, which did cost the group time. However we adapted to the situation and worked to what we knew.

Our original ideas were too demanding and our tutors advised us not to proceed with the underwater settings. This was due to a great number of potential problems and the difficulty of an underwater scene. We had to consider the physics of water and particle effects and how all of these things would affect us down the pipeline i.e. rendering.

Siobhan then produced a swamp concept; inspired the illustrative styles from the likes of: ‘Scooby Doo’ and ‘Disney’. The style as seen in the concept art is very illustrative with a well throughout colour pallet. The lighting focusses the audience on the shack in the distance to convey mystery. The trees enclose the audience with a sense of thickness and depth to the dense swamp.

Moving onto the scene, we chose 3D as our medium to produce the final scene. We intended to create a 3D scene with an illustrated style with soft shapes and water colours much like a painting. This was inspired by the concept art produced by Siobhan, and we chose to keep this style for the final piece. We chose the 3D route because of our acknowledgement of the progression of the industry, with the rapid advancement of technology we believed we could achieve an effective result.

I experimented with a number of tools in Maya to achieve the result I intended: illustrated yet believable trees. Late in production I had to hand over the role to Siobhan in exchange of the role of lighting.

Pre Vis: https://www.youtube.com/watch?v=IsZNIgdshtk

As we wanted to capture an atmosphere within the scene of ‘spookiness’ and ‘mystery’ we thought a lot about our composition including: lighting, spacing and the colour pallet. I believe we could have created a better atmosphere with the addition of lights and fog effects to add further depth to the scene. However due to time constraints we could not have added these elements without being late.

Late into production during research of sources such as films and games, I discovered that game trees are flat textured with a painting UV. These tress looked fine from a distance, it was upon close inspection I discovered this. If I had knew this before I would have completed the texturing much sooner and incorporated this knowledge in the final piece.

Final Video: https://youtu.be/fC-HxQlUCa8

The final scene is not as effective as I had intended. This is because of the style of the video, it has not rendered as intended and does not match our expectations. This is to be expected in a creative production, but does not make it less frustrating.

There would be a number of elements I would change to the final piece. For one I would have spent more time with the texturing, giving feedback rather than struggling throughout the project. I would have handed the responsibility over sooner to then support with feedback. Because I am not pleased with the final look of the trees as they appear rather flat and plain. As shown in the concept art, the trees are covered in moss which adds the elements of time and life to the work.

I would add additional effects such as fog which would be an appropriate atmosphere for a humid setting. In addition this would have aided the ‘spooky’ theme we were trying to achieve. To have achieved this we would have had to research the effects within Maya and how long a particle effect such as that would have taken to render.

I would have spent much more time on lighting, experimenting with the colour, the effect through fog and on surfaces such as the moss growing on the trees. Again this would have taken too long in the time given. Both learning how to apply these effects and the significant increase in render time this would have added.

Next time I may do a project such as this I would plan to use a dedicated render and stick with it, such as Renderman. I intended to work for those results, and I must plan ahead in good time to understand the requirements of such results.

Overall I have found this project enlightening and challenging. I have learnt that projects do not work as intended, no matter how prepared you believe you may be. Despite our preparations of role assignment, we still changed in favour of completing the task on time. I found it especially difficult accepting I did not have enough time to learn texturing and give the responsibility to Siobhan; by the time I discovered this it was too late. If I had accepted this earlier I would have had much more time with work I could understand and experiment further with improving our final piece. There have been elements I have enjoyed including the animation of the camera and the fireflies within the scene. I have learnt that Look Development is a very long and difficult process of experimentation and tests, which I did not enjoy all of the elements such as the texturing. As a group we worked effectively together offering feedback and support for each other, I should have communicated effectively with my team about how difficult the texturing was for me. I will listen to the advice that Tim Searle gave me at ‘Manimation’ in November, to keep things simple, understand the limits and make the best of them. Understanding my limits as a student in knowledge and time, and develop these by understanding how to use these effectively. 

Compositing

December 

After I had rendered the scene in Maya, exported and encoded the media it was time to continue to compositing. Compositing is the process of combining all of the elements within a production into the one piece. This could include: images, sound, music, video, effects, etc. 

For the post-production phase, I used Adobe After Effects, Adobe Premiere and Adobe Media Encoder. 

Within After Effects I added effects to the render including elements of rain, camera blur and particle effects. These are all effect incorporated within the composition as I believe these are all suitable elements to make the scene effective in its presentation. I explored and experimented with other effects before I discovered the right ones. These other effects were not suitable for this particular scene; however they may be useful in the future. 

Throughout this process I discussed with my team about feedback which meant going between multiple versions and software to achieve the result we wanted. 

After I had complied and rendered the effects in After Effects; I imported the elements of sound, video and image into Premiere. Using free sourced sound from ‘Sound Bible’, I incorporated natural sounds to the swamp environment such as frogs and rain noises. This adds depth to the scene as the audience uses multiple senses which emerges them into the setting; becoming an experience. 

The post production phase was interesting as I did expect it to be as delayed as it would be. This is due to the feedback and communication within the team. I believed the process would be rather linear, moving from one process to the next. Instead I took a step forward then two back, to change element’s for the team and improve the composition. I have learnt to adapt and predict the unpredictable in results; any process can be rather fluid, be it going backwards or forwards. 

Render Development

December

Once I had completed the lighting of the scene, modelling and texturing was completed. It was time to compile these elements together into the single scene. Siobhan had lead the pre-production stage, now I was leading the post-production including rendering and compositing. As mentioned in a previous post, this required a technical understanding to ensure the scene could be rendered in good time.

I faced many problems regarding the rendering. I had to change a number of elements within the scene such as the number of trees, the size of the whole scene and the number of lights. All of these elements increased the time of the render for each scene. Initially I intended to use Renderman for this project; because Renderman is one of the better renderers with the best results which I believed would enhance out group’s work. However I found this needed complete dedication and a deep understanding of the software; which I did not have time for within this project.

My second choice of renderer was Mental Ray, because the results can still be good quality. I used mental ray for a while throughout the project; however towards the end of the project the render times became very long. Even after reducing the file size in a number of ways, Mental Ray continued to have problems with rendering. On multiple occasions the render would stop for no reason with no warning leaving me to believe it had rendered.

As time grew shorter I had to make the decision of using a different renderer, to ensure the scene was completed before the deadline. I began to explore other renderers:

As shown different renderers provide different results. For our project I chose the Maya Software renderer because it had one of the shortest times, and the results weren’t as dark as the others. The results weren’t exactly what the group initially intended; however we understand the choices we needed to make both creative and practical to ensure the project is completed on time. 

Pre-Vis

November/December

Part of the project requires the group to produce a pre-vis of the scene. This allows us to see the fundamentals of the scene without concerning ourselves over details. Allowing us the time to critique and change anything that does not work before committing to it.

With the created scene I created a playblast of the scene. This is a low resolution playback of the scene with hardware texturing.

I found problems with lights, models and composition. These changes I made with no severe problems. The group agreed to proceed with producing the final scene. This would move into details such as the quality of the render to ensure the best image is produced. I find the Pre-Vis a crucial part of any creative production to ensure the final product is what we intended as well amending any changes now rather than later. 

Lighting Development

November/ December

After I had changed roles from Texture-er to Lighting, I had to learn how to light a scene effectively. I incorporated what I had learnt from collage on the photography course.  Many of the principles are the same much like a set, with main lights and highlights etc. In addition to memory I researched professionals experience from films and games which I found very helpful.

I found lighting was key in a number of ways, and that there is a lot of meaning in every light within any scene. Applying this to the work I tried to light the scene in such a way the audience would be directed to the house in the centre of the scene. The house is the centre point of the scene, surrounded by mystery and kept secret of by the swamp.

As the swamp is enclosed I considered the light would be diffused through the leaves, no harsh shadows. The lights would be ambient to illuminate the scene; allowing the audience to differentiate shapes and the space. Fireflies would fly around the scene giving a mystical essence to it with the soft glow hovering above roots and water.

I found lighting a very technical role with values within the editors. How these affect different materials, as well as monitoring potential render times. If I made the scene too complicated the render time would not be possible to complete before the deadline. I have learnt a great deal about the importance of light within a scene.

In addition to lighting, my role included maintenance of the scene. I had to monitor the poly count to ensure the render time would not be too high. Create the setting for the scene, creating the composition. This required a lot of problem solving, patience and technical understanding of Maya. I have learnt a lot about the technicalities of Maya, solving problems and applying them to my work. 

Texture Role: Changed

Look Development: Swamp

18th November 2016

Towards the end of the project, I found the texturing task very difficult. I couldn’t achieve an effective style which matched the concept art. I am still not sure if this was due to art itself, Maya or myself being over-critical (most likely the last one).

Discussing with my group, we decided to exchange roles of texturing and lighting with each other. We found that I understood the technical aspects of the project such as lighting and setting whereas Siobhan preferred the creative aspects such as modelling and texturing.

After we had exchanged roles I felt I could progress with the work effectively. Despite our original agreements at the start of the project; we saw a problem which could hinder our project and we adapted our roles to complete the project effectively.

From this I have learnt that roles aren’t necessarily ‘set in stone’. At first I beloved I could learn how to texture effectively, with time and effort I can learn and apply these new skills to the project. However we saw it was taking too long for me to learn and apply these techniques; if I continued the texture role I could have jeopardized the project for both of us. I have learnt to accept defeat for the benefit of the team, this means I can learn these skills in my own time and not the expense of a group project.