My involvement in this project:

• Storyboard – Act 2
• Techvis – Act 2; Shot 10-18
• File referencing setup and initial pipeline preparation.
• 3d models – Background
• 3d models – Space Colony
• Rigging – Space Buggy
• Simple model modification & rigging – Spaceship
• Animation – Shot 06, 07, 08a, 08b, 08c, 08d and 09
• Fixing scene with VFX rendering problem
• Full render for shot 03, 04, 05, 06 and 07
• Final video editing

Our Team and Collaboration

First of all, I want to thank everyone in the team, Emma, Antoni, Ghera, Cally, and Ben. They have done phenomenal job in this project and I feel very fortunate to be able to work with them who are very talented in their respective fields.

When we started this collaborative project, we were a stranger to each other. Even with Emma, which is on the same course with me, I never talk or meet with her before. But throughout this project, everyone gets to know and get along with each other to complete the project. Everyone was very open to suggestions and easy to collaborate with.

Group Leader: Emma Copeland
Animator: Emma Copeland, Kamil Fauzi
VFX: Gherardo Varani, Antoni Coll
Sound Design: Cally Spence, Ben Harrison

During our initial meeting when we had our storyboard ready, we set some targets for what we should at least achieve by the end of the project. Based on the storyboard, there are 3 acts (Act 1, Act 2 and Act 3) with 35 shots for the story. Technically the project is quite heavy, so we targeted to complete everything for Act 1 first, in terms of modelling, animation, lighting, visual effects, sound and music. And if we able to complete Act 1 quickly and have more time, we will proceed with the remaining 2 acts. We acknowledged that everyone in the team was also busy with their respective course assignments.

At the end of the project, we managed to complete everything that we targeted for Act 1, while sound department managed to produced complete sound and music for all 3 acts. It’s a bit sad that we were not able to proceed with the remaining acts, but I think everyone was proud with what we accomplished. The end product might not very brilliant, but the collaboration and commitment given by each team member were amazing.

If I can describe everyone on the team from my perspective, Emma is the group leader but she was very open to any suggestion. She always tries to finish her job quickly and very quick to learn new things. As you can see in my previous posts, she was active in our group’s conversation to give respond and feedbacks. She is also very good at explaining something and I wish I could be like her.

Antoni is a very hardworking person. He did many research and experiments on his own to get all the visual effect and lighting needed for each shot. He was also active to give his work-in-progress updates in our Discord channel. Ghera is the same as Antoni, and he even willing to do the animation for the first shot at the beginning of the story. The visual look that they setup for all the shots were very good. I personally love the atmosphere in the final render.

Ben is a very creative composer, and he has produced 3 amazing tracks that fit extremely well with the story. I especially like the track he did for Act 2. Cally on the other hand is a visionary foley artist with unique vision with the sound effects. When I received the audio effects for editing, some of the sound feel quite weird to me at first and I can’t imagine how it should be implemented. But when Cally did the audio mix on my video editing, the effects match very well with the scene.

With all that said, I also want to thank our supervisors, Luke, Christos and Ingrid, who has been very helpful in providing guidance and assistance throughout the course of this project.

What I Learned In Terms of Technical

I have learned several new techniques and tools in Maya. I can still be considered as new with the software and there are a lot of things that I need to learn and understand.

At the beginning of the project, I learned about how to properly setup and use referencing system in Maya. I had several experiences working with reference or proxy pipeline using 3dsmax and Blender before to understand how important and useful this method is. So when we started building assets in our collaborative folder, I quickly began testing and experimenting the feature to see how it works in Maya. I watched several video tutorials when exploring this method to check in case I missed anything. I found that the reference system in Maya is quite easy to understand and powerful to use in project. Despite that, there are small things that irritated me about how Maya project, working folder and referencing system works when compared to other software. As far as I discovered during this project, the features not working as seamless as 3dsmax where it always finds files that we throw inside the working folder without the need to ‘repath’. But maybe there are still somethings that I don’t completely understand about how to use the feature in Maya. I will explore this more during the Easter break.

During this term I also learned more about rigging in Maya. In this project, I rigged the space buggy so I can animate the vehicle with moving wheels and suspension for one of the scenes. I also did a very simple rig for the spaceship model so all the 3 spaceship’s doors can be animated.

What I really explored deeper about rigging in Maya was when Crystal asked for my help to rig one of her group’s characters for their game demo project. I’m not fully involve in her group but only to help her with the rig. I have done various character rigs in other software before to understand at least the very basic of it. But I don’t have experience to do a full character rig in Maya, so I did a quick research and watched several tutorial videos to understand the tools and process. I found that the character rig in Maya is not very hard to understand. So in this regards, I’m very happy that I learned so much things with Maya rig from placing bones, controllers and weight paint. The only thing was that I’m quite slow while doing the rig as I’m not yet familiar with the tools in Maya.

Throughout the process, I was actually very concerned about the rig will break when imported into a game engine. I have faced and learned many character rig problems since I was working in game production for almost 3 years. I knew that a lot of things must be setup properly like root, bone directions, and parenting for the rig to work correctly in game engines. I know what I should prevent and do in 3dsmax and Blender, but feel clueless when I rig the character in Maya. The rig I did works properly in Maya but it’s not guarantee will work in game engine. I don’t have much time to test it as I also had other tasks with my group project. Unfortunately, what I afraid turned out to be true when Crystal mentioned to me that the character legs broke when imported into Unity game engine. She said they proceed to do another rig themselves as she don’t want to interrupt me again. Even though the rig was not use in their final product, I’m so grateful that I have learned so much in this matter. I may want to revisit this again during the Easter break so I can understand more about rigging in Maya and how to prevent the problem.

There are a lot of other technical aspects that I learned with Maya during this project. So to list a few; I explored more about modelling, properly separating and combining meshes, grouping, modifying Arnold materials, lighting, using layers, understand more about outliner (which is something I’m quite confused during the last term), creating and using ‘Quick Select Set’ feature which is very useful during the animation process. The most important was, I learned how bad Maya 2020 is! So I’m going to stay with Maya 2018 as suggested by Luke.

Reflection

I have learned a lot throughout this project. One of them was I’m improving one of my behaviours which I think can be pretty bad sometimes. I’m a bit too perfectionist person where I always concern about everything must be done precisely and organised accurately which can cause delay to my works. One example is you can try to save any image and video in this blog to see that I even renamed everything properly, which nobody will actually care or see. In the past, my obsession was much worse. I will measure every 3d object and align vertices precisely to not have any random decimal in coordinate when I do 3d modelling, even that kind of measurement was not needed or necessary.

But as I work more and more with other people during my previous works and in this collaborative project, I learned to control and apply this behaviour for more appropriate situations. I can see perfection can be useful but too much picky can cause problems to the project. This is something that I continue to improve more when working in this project. I managed to accept and weight which things are more important. One example was I’m not become too fussy when I saw minor inconsistency in render quality between different shots. In the past, I would reject it and will ask for re-render again and again. I know that everyone is learning as much as I am. I feel calmer when I managed to do that and grateful this collaborative project happened. I still have this obsessive behaviour, but it is much, much better now.

I also realised that I communicated more with my friends whether using text in our Discord channel and verbal in our weekly video meeting. They were all very nice and there are times we become pretty chatty in the channel and make jokes. I joined the conversation whenever possible since I really want to improve myself to write and speak fluently in English. Sometimes I had difficulties when I want to explain somethings quickly. I admit that I had a lot more to learn but this collaborative project has somewhat helped me in this regard.

I talked about my thinking for future path in my first post during the first week of this term, https://kamil.myblog.arts.ac.uk/2021/02/01/self-reflection-and-future-path/ . My heart was torn into two. I’m very interested with cinematography and character rigging (and animation). To be honest, after doing this collaborative project, not only I still cannot decide which side to focus, but it was getting harder to determine. I did a character rigging tasks and animations, and also attempted to do my animation shots with cinematic elements. Although probably what I did with these areas was not very much in the project, I genuinely love to do both of them. I will have to do more thinking more about this during the Easter break since I have to decide and blog about this in the first week of the next term. Maybe I will also need to ask Luke for his opinions and suggestions for this.

22 March 2021

I have finished all my animation shots (shot 06, 07, 08a, 08b, 08c, 08d and 09). At this moment, our team was still struggling to get all the animated shots with visual effects rendered on the render farms without errors and crash.

Since the deadline is coming in the next 3 days, I’m thinking to compile everything we have so far from renders, playblasts, storyboard and sound into a complete video because at this moment we still don’t have anything concrete yet. In worst case scenario if the final render cannot be completed, we at least have something in a full proper video form for submission. I let Emma, Antoni and Ghera pushing the render jobs since they were already much more familiar with the errors that currently happening and render farm than me.

I gathered every latest playblasts and renders that we have from the shared OneDrive folder and university network. I then started doing a rough edit using Final Cut Pro. In this early version of editing, all of the Act 1 animation were technically there but with a mix of playblasts and broken render frames. We only managed to finish animation for Act 1 only, so for Act 02 and 03, I used drawing frames from the storyboards as animatic with motion to tell the scene. As more renders become available, I will keep updating the editing clip with the latest one.

Emma suddenly mentioned that she had a problem with her Shot05 where the character legs become wonky. I offered to look into her file and fix the problem so she can send the shot for render as soon as possible. After spending a bit of time, I finally found the problem in which she accidentally animated the bones instead of controller. While I’m at it, I created and animate a camera for the shot and send the file back to her.

We still had time to have some fun and laugh with each other in our Discord channel.

• 

• 

Later at night, Emma and I jumped into a quick session with Luke where he try to fix our problems with the render farm. And after numerous tries, Luke found out that Maya 2020 could be the culprit because the original animation shot created in Maya 2018 can be rendered without problem, but animation shots with added VFX that were created in Maya 2020 crashed on the render farm. Luke and Emma explain it to the other team member.

After the session with Luke, I asked Emma to stay in the voice call to discuss what options did we have to solve the problem. We finally decided that Emma will render all the shots without the VFX since they can be rendered without problem, while at the same time I will look into the problem and taking the turn to get it rendered with VFX. So if I failed at my render, at least we have all the normal shots rendered in time by Emma before the submission dateline.

I stopped doing the editing at this moment and since Emma was still logged in to the university network, I asked her to send all the latest animation files with VFX into our shared OneDrive folder. I started by checking shot 06 and 07 as soon as I received the files. There were several missing reference items created by Antoni and Ghera when I opened the file. So I keep asking Emma several times whenever I need another missing references.

I did several initial test renders and Maya did crashed several times. When I managed to get a render, I asked Antoni if I got the render look exactly like how they setup and some of them turned out different on my computer. After several tests I finally understand how they setup the HDRI and lighting references in the animation files.

23 March 2021

Since its already 2.00am in the morning and I didn’t have enough sleep for the last 2 days when pushing my animation shots, I proceed to let my computer doing the render for shot06 and see how it goes.

When I woke up, the computer was still rendering the shot about halfway done without crashing. I didn’t cancel the render to see how long it will take to finish. I took the opportunity to continue the video editing using my other computer at the same time.

The render finally completed, and it took about 7.5 hours to finish. So it can be rendered successfully without any crash and broken frames. The only problem now is the render time. I did some calculation and its impossible to render all the shots in time. So I went into the file again and doing some optimisation to the lightings, settings and anything that could make the render time faster while at the same time retain the visual to look as similar as possible to the original. I finally found the way and managed to render the same shot again within 2 hours compared to the previous 7.5 hours. I then make a new template for the lightings and settings to be used for other shots.

I spent the whole day for this and managed to render shot 03, 04, 05, 06 and 07 with fairly short time; about 2 hours for each shot. While the renders running, I continue doing the editing on my other computer and updating it each time a new render finished. Emma also has managed to render all the shots without VFX, so I put some of the shot into the editing.

24 March 2021

I sent the first version of the final video with all the current rendered shots to the team in the Discord channel. With the render possible now, Antoni, Ghere and Emma procceed to render the remaining shots which are shot 02, 08a, 08b, 08c, 08d and 09.

• 

• 

16 – 21 March 2021

This week I focused to finish my remaining animation tasks for shot 06, 08 and 09.

Action for shot 08 was quite long, so I decided to separate the action into 4 shots to have better angles to deliver all the actions of the scene. While doing these shots, I also made some adjustments to the Space Colony model since the character has some interactions within the area. The original space colony I made several week before was a bit large for the character so I scaled it down about 10% smaller to make it look more proper and realistic for the character to interact to. All of the shots took a bit longer to finish than expected because a lot of things going on at the same time.

Playblast

Below are still renders (without VFX) from all animation shots that I finished this week.

While doing the test render for shot 08d with the character’s helmet off, I noticed that her eyes are black in render, but not in viewport. I proceed to repair the problem and thinking it would be a very quick fix, but it turned out to be a 2 hours work. There are 2 layers of material for each eye and there is one material that was not shared for both eyes. It was a bit weird. I did a lot of back-and-forth changing settings and doing renders again and again to get it right. I even imported the character into new file to check if the problem persists. Almost every setting was changed from reflection to specular, roughness etc and I was still unable to fix it.

It was quite frustrating when the eyes sometimes turned into completely white, or completely black with the space colony reflection in it. I also do some lookup to forum discussions to solve the problem. At the end, I cracked it by a straightforward method by just ignoring the settings and replaced both material layers with just a pure texture. I think it is not a correct solution if we want to have a very close-up shot to have realistic look and reflection of the eyes, but it will do for now especially for shot 08d.

After everything ready, I sent the Maya files to the visual effect department for lighting and effects so that they can be sent for render. I also sent the playblasts to the sound department so they can compose and arrange the sound effects needed.

8 – 11 March 2021

This week I focused to update the space colony building model, so I could animate shot 06, 08 and 09. The previous space colony was not finished since I only used some dummy objects for the interior.

I spent a bit of time searching for free premade models at several places including online. And later I found a very suitable assets for the building interior from the university library. I downloaded the textures and placed them in the project’s ‘sourceimage’ folder. Similar to the previous models that I setup, the first step I did was to re-path the textures to have dynamic address to our collaborative folder.

The assets were modular and can be combined to produce different objects. I chose the objects that I thought could be used for the space colony and saved them in separated files.

I made some adjustment to the entrance model to fit the space colony. I extracted the door and fixed its pivot position and orientation so it can be opened, closed and animated properly for shot 08. I also modified the space colony wall at the entrance area to have proper path for the entrance model.

After that I proceed to arrange the other interior models to make it looks more like a real space colony with tents, generators, water tanks etc.

I separated the tent’s door and fixed the pivot so it can be opened and closed for shot 08

Although the process seem simple, it took me almost 4 days to finish the interior from searching the free models, choosing the suitable model, fixing polygons, shapes, material, textures and lot and lot of arranging assets until I got the final version.

I updated my progress to our Discord channel but I didn’t upload the model to our collaborative folder yet. I want to animate my shots first because the character will interact with some of the interior props. I might want to change some of the prop arrangements later according to the suitability of the shot and camera angle.

2nd Collaborative Project

Back in February, Crystal invited me to join her project, but I was already in Emma’s team at that time. So, she asked if I have some free time to help her for character rigging work only.

I told her that I will let her know if I can join after the first meeting with Emma’s team to see the workload of the project. After the meeting, I informed Crystal that I can do the character rigging for her project.

I waited for Crystal to finish the character model first and last week on 27^th February, she sent me the final model of the character. I do a quick check on the model to see if there is any problem with it.

I then decided to do the rigging work for this whole week as fast as possible so I can get back to work on my primary group project.

1-10 Mar 2021

I started by preparing the model for the rigging. The file has some unnecessary objects like wall plane, lightings & camera, so I did some cleaned up and deleted those objects. I repositioned the character to the center grid and scaled it according to the real human size which is around 160 – 170 cm tall. I also renamed every object in the outliner properly so I can easily identify them when I do the rigging process later.

Next, I did some adjustments to the model topology and shapes so it can be rigged properly. Since Crystal mentioned that the character will be imported to a game engine, I did some optimisation by removing unnecessary segments on body parts which has too many polygons. I also changed some of the character shapes and poses like the mouth and hands, so they are symmetry and easier to rig.

Model optimisation

• 

• 

• 

• 

• 

• 

• 

• 

5-8 Mar 2021

I’m not very familiar with rigging in Maya when compared to 3dsmax and Blender. So before I proceed, I did some research and watched some tutorials on this matter. After several testing and learning, I decided to use Maya’s standard Human IK to rig the character instead of using the Advanced Skeleton plugin. I found several videos showing exporting to FBX for game engine is simpler with Human IK then with Advanced Skeleton where I need to change lot of things before it can be used in game engine. I’m not really sure, but don’t have much time to test everything as I need to finish this quickly so I can back to work on my primary group project.

Bone placement process

Skinning Process

Poses Test

It took me more than a week to do the whole process. I also did lots of testing and experiments while doing the rigging to better understand the Maya rigging which make the process longer. I actually did try to rig the face but I encountered some problems and not satisfied with it. So I sent the character rig without the facial controls first to Crystal for her to test.

Crystal was concerned about my works with my primary group project so she said she can do the facial rig. I suggested to her that if she want to do it, we can both do our own rig in case I fail to do mine, then she will have a character with facial controls.

Facial Bones and Controllers Setup

I proceed to finish the rig for the character face. I don’t use any plugin for the facial bones and controllers, but manually creating and placing them using the Maya standard ‘joints’ and ‘curve’ shapes. I watched several video tutorials for this and throughout the process, I can say that the concept of bone / ‘joint’ in Maya works quite different than 3dsmax and Blender. I’m not really sure how to describe it but in Maya, it treats bone as point and the connected line between each point is a visual to show its parent-child constrain. In other software, that connected line is actually a bone. I’m a bit struggle at first to understand the Maya joint especially the point’s orientation. After several try and errors I managed to grasp the concept although maybe not fully master it.

• 

• 

• 

• 

• 

• 

Skinning Process

Facial poses test

About a day later, I managed to complete the rig and sent the updated version to Crystal.