Idea / Powerpoint Presentation
Download the full powerpoint here Grotesque introduction
Artist Research
Research is an on going process throughout the creative workflow. I have looked at various artists, sculptors and 3D animators to help me find inspiration and ways I could apply methods to my work.
Jonathan Payne
Utterly surreal warty, nipple-y, toenail sculptures that induce anxiety and horror.
Payne uses super sculpey, polymer clay, acrylic, and human hair.
http://streetanatomy.com/2013/01/09/the-fleshlettes/
If I were to take inspiration from Paynes work it would require me to explore heavy 3D modelling. I could possibly take elements from it such as teeth or pubic hair as these are things that could possibly have good animated potential.
Paul McCarthy
McCarthys work is audibly questionable and often exaggerated. The projection room exhibition is an interesting take on showing clips in a chaotic and convoluted environment.
John Walter
Walter is an artist, curator and academic from London. He works across a diverse range of media that include painting, moving image and installation. I recently visited one of his exhibitions in Home, Manchester during the Manchester Animation Festival. Walter latest series of artwork is titled ‘Capsid’
“Capsid is the latest multi-media maximalist installation by British artist John Walter. Presenting a compelling sometimes riotous and often surreal world, CAPSID is the result of collaboration between Walter and molecular virologist Professor Greg Towers of University College London.
This major exhibition addresses a crisis of representation surrounding viruses such as HIV, by bringing new scientific knowledge about viral capsids to the attention of the wider public…” – John Walter, Capsid exhibtion booklet, Home.
Walter had a few video demonstrations shown across tv’s in his exhibition. They were often moving versions from his still paintings. Interestingly though he used a lot of disgusting sound effects from his mouth such as sneezing and coughing to accompany the visuals. Which it got me thinking into possibly incorporating that technique but taking it a step further and using all sound effects for my final animation by my body and mouth. This could potentially give my animation the dynamic it needs.
Nicolas Gaspardel ‘Yuck’
Gaspardel’s Yuck is a series of photographs of unconventional objects in a some what normal placement. The use of wet chewed gum layered onto a piece of toast creates a discomfiting aura.
I really like how the background of each piece is a pastel colour. This contrasts the subject from itself and presents it in an inviting, relaxed manor.
Incorporating this style background could be something I incorporate further down the line.
Extraweg
https://www.instagram.com/extraweg/
Oliver Latta is an motion designer under the alias Extraweg. He is famous for his work posted on Instagram.
Latta uses a very muted, soft colour pallet consisting of mainly pinks and whites. His work explores surreal human elements that often leave the viewer grossed out but somewhat oddly satisfied. Latta’s work has set a trend for Instagram 3D art as more and more artists are using his colour pallet and visual ideas.
I find Instagram a good way of finding similar artworks and inspiration as there are millions of CGI artists that upload their work.
I saved hundreds of grotesque themed renders to my account, all of which I have taken various amount of inspiration from. However I feel I have already divulged enough into artist research. I will continue to use instagram and source more research along the project.
Idea
Like all good ideas, it came to me in a back of a taxi which were loosely dictated into my phone. I originally thought of having a CGI urinal with someone urinating in it with cigarette butts and urinal cakes floating along the stream. Narrative animation isn’t my forte so I know I wanted it more various scenes than an interlinked story driven animation.
With the urinal idea in my head I was figuring out how I could make sense out of it and give it purpose. I toyed around with building a completely CGI dirty bathroom and having various objects in the bathroom its own independent event and the bathroom being the environment that ties them all together.
Initially I liked this idea as it would push me to do more modelling and create a realistic textured and accurately lit environment that would require a bespoke HDR map.
But dissecting the idea some more, I thought I could possibly 3D motion track gross objects into a live action bathroom. This would require a lot less work creating the environment and allow more time to perfect the gross subject. However, location scouting and filming in a bathroom could pose its own issues.
The idea developed more and more as I progressed with this project. I work better with starting something without a lot of planning and it all falling into place, rather than planning every step I do, before I do it.
I will document the work on a scene by scene basis as I was working on them all at the same time.
Urinal Scene
Wanting to put pen to paper (if you will). I started modelling a very rough urinal model with dimensions from a urinal manufacturer.
Modelling of the Urinal scene. Dimensions from here
Visual test to see how the urinal would hold dynamic simulations. Cubes being a place holder for liquid.
Inital realflow water simulation tests using the Realflow C4D Plugin
Various more simulation tests. Played around with the idea of interactive urinal cakes that would be buoyant on the water. Decided against as it was more strain on the render for little gain.
Initially I wanted the grotesque animation to sit within a single 3d environment; a dirty public toilet. But through continuous experimenting i opted to not make it look real but to take the “real objects” and put them in a surreal scene.
Experimenting with colours and various studios to make it obvious that it is trying to be cgi and not pretending to be real.
I decided to switch render engines and use/learn octane renderer to improve the visual fidelity of my renders. I have no experience using octane so I will have to learn to texture and light using it from now in during the project. This was a big step in the right direction for me during this project. I felt I have always been limited by the in-built render engine within C4D. I hope by making this switch and learning the plugin will push my renders onto another level.
Along with changing the renderer, I re-modelled the urinal to a high quality. I bevelled edges along the faces as with a metal texture applied, it will give the urinal nice edge reflections.
Through experimenting with background and lighting I was left with two styles of background for my animation. These background textures were from a texture pack “Render King Texture Pack 2” I managed to get a copy of the pack for free. I will be using pre made textures for the time being until my skills at octane improve.
I was unhappy with the source of the water being visible so I experimenting and came up with this solution of the source of the water emitting from inside the urinal. This gives it a exaggerated disgusting look. The water was made using octanes subsurface scattering texture, which would otherwise not be possible using Cinema4Ds standard renderer. I really like this approach to the effect as its exactly the surreal look I was going for.
Im happier with how the animation works for this scene. The background and lighting still need adjusting however. The properties of the water are not perfect also, the water slides along the floor with little to no friction but it is a step in the right direction.
Ive decided adding some kind of action to make the urinal flood with liquid will make the animation stronger visually. For this, I modeled a old fashioned toilet chain. The chain is fully dynamic so when the handle gets pulled the chain springs back up.
I am still playing around with various backgrounds to complement the subject. You can also see how the chain will be positioned and utilized.
Here is a render with the new changes.
An issue with realflow plugin that I have had for a while is the inability to create a foam effect that acts as a secondary effect with the water simulation. The standalone application Realflow 2015 has this ability but is much less streamlined as I have to import data back and forth rather than it all being in one program.
I attempted to find a working foam solution with standalone software however my scene involves an emitter within the body of water and thus makes it impossible for some reason to use the solver that creates the foam. This was very frustrating for me as water without secondary effects such as foam or splashes make it look poor quality.
Heres an example, albeit not a good one.
Im hoping I can someway fake foam in post production as that is my only solution it seems.
To add more nuanced detail to the urinal scene, I modelled a grid/drain that would drain out some of the urine.
Originally, the drain was supposed to function and some water visibly flowed through the small holes. However, adding small objects with tiny holes / small cell sizes increase the render time 10 fold. I tried increases the size of the holes and making it just an empty square but they didnt add as much as the render time warranted.
I opted to remove the colider tag from the drain and make a square with the same dimensions. So the liquid would interact with the shape of the square without draining anything which ended up looking great and without a big load on the render time.
I also changed the background into a pinky soft style material which was a direct inspiration from Extraweg’s work. I feel this contrasts the gross subject matter which inherently makes it more visually appealing.
add final images !!!!!!!!!!!
Poo Scene
Created using very viscous liquid properties in C4D’s Realflow plugin. The poo is very irregular looking therefore more realistic. This was a rough initial render but I was very happy with how it turned out, shape and texture wise.
Here is an animation test to see how objects would interact with the poo. Initially I wanted a shoe to step into the poo but to take it one step further removing the shoe and replace it with a bare human foot or skin contact.
The model of the human foot was a premade asset from a human body pack I downloaded previously.
I tried doing several more varied poo shapes but I feel the initial test is still the strongest.
A problem I ran into was that I wanted my camera to be a fixed constant throughout the animation and a scene that built like this makes is difficult to maintain that without dramatically increases the size of the poo to fill the frame.
I could not get a good press into the poo to leave a foot imprint and after many failed attempts this scene was not doing it for me, even after such a promising initial start.
I scrapped the scene I came up with ideas that I could push the surreal theme into.
Still wanting to use a human element I kept the model of the leg, but I tried rigging it this time to give it actual animation.
I introduced a toilet model from here to give more of a subject to the scene and so the camera had something big in the frame. I re-textured the toilet and widened it for my scene.
I came up with the idea of the human foot breaking the toilet somehow and poo gushing out onto the studio floor.
The poo simulation is still very much early stages but you get to see the framing and angle of the impact.
In this video you can see the progression of this scene
To save myself animating the foot doing a walk cycle into the frame, I decided to attach it to a hydraulic type arm that brings it into the frame. This saves me time creating a detailed walk cycle.
Animation test 1:
The leg was not rigged properly so the toes were left out of the link, however when animating it it gives the walk this uncomfortable, creepy walk which reinforces the underlying theme. But I felt it looked more broken than purpose built.
Animation test 2 – 3:
Shows the toilet breaking with impact of foot and also the toilet lid reacting to the impact, hitting the back and then shutting. This was all keyframed animated as nothing dynamic is actually happening.
Animation test 4 – 5:
Rough simulation of liquid in the bowl and escaping when the wall is broken
Animation test 6:
Increased viscosity to make the poo more thicker.
Animation test 7 – 10:
Added debris that falls out of the toilet to sell the impact of the foot on the toilet bowl. Added the hydraulic arm to the foot. The debris is a separate effect added during the impact
Animation test 11 – 12:
Animated the hydraulic arm and polished the keyframes of the other interactions.
Animation 12 is the final iteration of the scene.
Here is the final render with added lights and the new soft pink studio.
Boob Scene
For this idea I was thinking of more scenes with human elements. With still using the 2 high detailed human models I have, I separated the breasts from the female model.
I immediately thought to use the soft body dynamic tag for the boobs as it gives them life like jiggle.
Initially my idea was to have the breasts expand and explode showering milk everywhere. I did a few render tests but did not save any unfortunately so I can not document them
This shows the portion of the female body that I removed. I closed any open polygon holes to make it a solid object.
The trolly the subject sits on was initially created for a previous scene that ended up being not used in the final.
Brainstorming the idea, I figured the milk was not that strong of an idea and it could be changed into something better.
I then came up with the idea of using surgical breast implants exploding out of the model. This is something that can be argued a lot more gross than breast milk.
Surgical marker lines would be important for my model to include as this reinforces the idea that the breasts are undergoing surgery.
I went into the texture map of female body model and hand drawn surgical lines on them, similar to the ones researched above.
With the texture replaced for the new one you can instantly see the effect it has. The hand drawn element sells the idea more as apposed to perfect pen tooled lines as a surgeon would often do it free hand.
Using the pressure tab in the soft body dynamics tag I am able to expand and compress the subject to make it seem like there is pressure building up inside of the model.
I created a thin wire box to confine the subject as it expands.
As the subject reaches its max expanded state, it tears apart and hits the trolly like a cloth object and 2 silicone implants fall onto the table.
Cinema4D has difficulty working with soft body tags and cloth objects in this same object. A work around for this is creating two identical objects, one with soft body tags that inflate, and another with cloth dynamics that tear apart. I switch the visibility and turn on/off the dynamics on apposing keyframes to make a seamless transition between the two separate but identical objects.
In this video you can see the effect in action.
Here is the final render with the added soft pink studio
Masturbate Man Scene
I got inspiration for this idea by a fairly recent trend involving a dancing motion captured man underneath a cloth dynamic plane.
This effect uses motion captured data to make perfectly animated characters with various overlayed effects on them.
There is a fairly large movement with Instagram artists using this trend. In particiular Antoni Tudisco uses this effect a lot in his work.
I have explored using Motion capture data before in my other work.
This video was coursework in my first year of University.
With prior knowledge of using motion capture data, I set up the new scene accordingly.
Mixamo is a free online store to download thousands of character models and motion captured animations. You have full control of the speed and the arm length of the data you want. You can import the data into various 3D programs and assign the keyframes to a preexisting model or the default one.
My idea was to have movement of a man under a sheet masturbating. The concept is pretty out there and could be quite controversial if done correctly. Which seems fitting into my underlying theme.
Instantly I knew there would be no motion capture data of a man doing what I needed, so I would have to alter the keyframes of a pre existing animation.
I found a model of a man doing a fist pump which was the closest thing I could find. I figured if I moved the arm down 180 degrees or so and have him repeat the fist pump animation I could pull it off.
In this render you can see a rough version of how the animation would look without the sheet. The sheet will interact with the body causing a outline of the event.
Here is a video showing the progression of the scene. (for some reason the video isn’t embedding)
Animation 1:
This shows cloth dynamic object overlayed the maximo character. The cloth picks up the outline of the animation.
Animation 2:
This shows a human skin like subsurface scattering texture on the cloth. This allows you to see the silhouette of the character underneath which gives you a better representation of what is happening. I have also joined another maximo animation to the end of the “fist pump” one. This makes the character lie down of the floor out of exhaustion.
Animation 3:
Changed the background to an HDR enviroment to give realistic lighting to see how the texture reacts with the light. I decided to remove the visibility of the character underneath as it was clipping through some parts of the cloth. I also removed the sitting down sequence at the end.
Animation 4:
This shows first use of liquid simulation which will be used for the climax of the event. The texture is okay in some parts where the liquid is dense however in thinner parts it is too transparent.
Animation 5:
Changing some parameters of the liquid. Not liking the outcome as it is too bubbly and rigid.
Animation 6:
Added a larger flow of liquid. Don’t like how it pours, too much at once.
Animation 7:
Tried to make the liquid come from the tip of the object. Didn’t like how it looked.
Animation 8:
Thinner stream of liquid, better result.
Animation 9:
Changed background to something more relevant for my animation. Lighting needs adjusting as it was set up for the HDR environment previously but it needed to be changed to fit with the style of the other animations.
A big issue I ran into with this animation was the relationship between the cloth being a moving object and the fluid needing to collide with it. Fluid collision works best when its collides with a non moving non high detailed surfaces. The scene I was attempting to make is a moving surface with tens of thousands of polygons. I needed to figure out another way to make the fluid collide with the cloth without crashes my computer.
I needed to lower the polygon count on the cloth surface to allow the fluid to collide with it. I tried various methods but when decreasing the polygons in the cloth it gave me unwanted results.
I spent weeks trying to figure a solution out and I eventually cheated a method and it pretty much worked.
I created 12 spheres that intersected one another along the rough outline of the subject. These spheres had the collider tag instead of the cloth. There was an issue of the spheres not having the motion captured keyframes, but when placed under the hierarchy of the motion captured body, it worked!
So the fluid interacts with the spheres, not the cloth. But when turning the spheres invisible, it creates the illusion that the cloth is being interacted with.
However, as you can clearly see the shape of the sphere objects is different that the intricate details on the cloth object, so the fluid doesn’t align perfectly with the cloth but it isn’t noticeable unless you look.
As this was my only option, I decided to progress with this workaround.
For the final render I changed the liquid texture. Originally the texture had subsurface scattering properties but I found removing those properties and adding opacity to the texture created a more realistic look.
Balls Scene
For this scene I got inspiration from the Boob scene I did previously. I really like using the soft body dynamics tag as giving objects squishy properties gives them a sort of personality that otherwise would not get from a rigid body dynamics tag.
Macomoroni, another Instagram artist whose work constantly featured deformed human elements is another artist I have taken inspiration from.
https://www.instagram.com/p/Bt8ah-kg16w/
This video shows a soft body dynamic sack being interacted with a paint brush. The sound design with the video really adds to the animation also.
Initially I wanted to create something similar but with a pain element. I tried creating a scene with suspending balls and a hammer hitting them and they possibly bounce around after the impact. This could provoke a sort of ghost pain for my viewers which I something that would fit in with my theme of being uncomfortable.
I had extreme difficulty recreating a dynamic sack with objects inside it, so I scrapped the initial idea and came up with another one.
I came up with the idea of several soft body spheres falling into a container. This would allow all the spheres to squish into each other and potentially bounce around the container.
I would be adding hair dynamics to the spheres as that reinforces the aesthetic that they are meant to be testicles.
I wanted the container to be a medical / science related object as it gives the animation a science experiment feel. I have also used similar objects throughout my scenes such as the medical trolly object in the Boob Scene and some other objects that I did not make the final cut of the animations.
Click here https://vimeo.com/333097039 to see the render progression with the Balls Scene
Animation 1:
Initial test with hair dynamics and collider tag on beaker.
Animation 2:
Higher quality initial render.
Animation 3:
Added keyframed shake to the beaker to reinforce the idea that the balls have some mass to them and the beaker is shaking from the weight. Keyframed animation did not work that well. Looks faked.
Animation 4:
Added rigid body tag to the beaker so the soft body spheres now interact with it correctly and it shakes depending on the mass value of the balls. More accurate render.
Animation 5:
Added more spheres, increased the size of spheres and added more hair. Too much hair, will reduce next render.
Animation 6 – 9:
Changed the soft body properties to make them less bouncy and deflated. Changed render setting and removed hairs to speed up test render time. Similar result with animation 7 8 9
Animation 10:
Added colour variation to the spheres.
Animation 11 – 12:
Adjust soft body properties again.
Animation 13 – 15:
Dramatically increased size of spheres to fill the beaker. However, bounces out too much at the end
Animation 16:
Bounce is getting better.
Animation 17 – 18:
Final animation test. Happy with the result. The bounce of the spheres gives me a good opportunity to add some sound to each bounce event.
I created this high resolution render during one of the tests which I did not fully render out. I felt the image works better as a still render than a animation.
There is an issue with the scene that the high dynamics do not collide with the beaker. You can see all the hairs are clipping through the glass. This is something I will fix in post production.
These are stills from the final render.
Animatic
This animatic was created in November to demonstrate the rough idea and style of my final animation.
Discontinued scenes
In total I have created 10 or more scenes to have a wider choice for my final few animations. I believe the animations I have chose are stronger and are more visually appealing than the discontinued ones.
Spot scene
Initially I had this spot animation as an animation I could use, however I could not create an accurate spot pop animation. I discontinued this scene as it is quite static. Although it is visually interesting, I feel I made the right choice having fewer animations but a higher quality overall rather than more animations with less quality. This animation was shown in my animatic but not my final.
Skeleton Scene
This scene consists of a pre-made skeleton model sitting on a tray that fills up with a blood like liquid. I discontinued this scene as again, it lacked a lot of motion and the skeleton model was a pre made model. I tried to avoid pre made models as much as possible in this project as it could potentially miss lead someone into believing I created it. The times I have used pre made models are when the model is not the main subject of the animation, and there is other things happening around it.
Tampon Scene
For this scene I wanted to have a dynamic tampon on a suspended string being dipped into a puddle of blood and it soaking it up and increasing in size. However, when making the scene it was extremely difficult for the string to connect dynamically to the tampon for some reason. Running into this difficulty so early on in the creation of the scene made me stop pursuing this idea. Looking back on it I feel it was a good idea that if executed well would have made it into my final animation.
Teeth Scene
This scene was featured in my animatic and consists of loose teeth in a mason jar rattling around as the jar shakes up until the jar drops and shatters causing the teeth to escape onto the studio floor. I wasn’t very happy with this scene from the get go. I feel it had not enough interesting elements to make it visually appealing.
Vein Scene
This scene also featured in the animatic. It is a vein on a medical looking support that a hypodermic needle penetrates the skin and draws blood. I discontinued this scene as I found it hard to demonstrate the needle penetration from a static framed camera position.
Audio
Audio has always been a crucial part of any of my animation work. However I feel I do not spend enough time mastering the audio as much as I should do. I have been excited to work on the audio for this project as it gives me the opportunity to take my time and perfect it.
As the audio solely relies on sound effects and atmos this will really bring the animation to life and make it the best project I have worked on.
Attending the Friday audio sessions in uni have taught me crucial information about audio that I otherwise would have disregarded. In those sessions we were using Avid Pro Tools, an industry standard software for mastering audio.
I tried using Pro Tools for this project but was quickly overwhelmed by its sheer complexity. Instead I opted to use Adobe Premier Pro. This is also a new software for me as I normally used After Effects to master audio however I quickly learned this method is far from good.
Learning Premier was a fairly simple process as I am very familiar with the adobe suite family.
This picture shows the workflow inside of premier pro for the masturbate scene.
I wanted to record all the sounds myself where possible to make the animation as original as can be.
For this I used a H5 Zoom Recorder loaned from my universities equipment loans. The H5 is an amazing handheld recording allowing for clear crisp audio with the addition of a external boom mic and the ability to do duel channel recording if need be.
I have had to use some pre existing audio clips as some of the sounds are not possible fore me to create or record. For example hydraulic arm sound for the toilet scene.
I find creating a list of all the audio helps tremendously. For some of the soft body animations I went to the local shop to buy props to record their squishy properties.
The props include a squidgy orange ball filled with yellow liquid for the balls scene.
Balloons for the inflate of the breast scenes.
Putty slime for a squidgy noise for the balls scene.
A super memory foam ball for a soft bounce noise for the balls scene.
Final Renders With Sound
I think the audio has worked rather well. It could definitely be better in some places but overall I am happy with the outcome.
The urinal scene I feel the water audio does not have enough depth and weight to it as my only source for the water was using my bath / shower which was in a small reflective room.
I also felt the Balls scene lacked somewhat. I feel the impact of the ball sounded too slappy but I did not have an object with similar properties to record. I think the best object would have been an exercise ball as it is large, hollow ad compresses when under pressure.
The best scene audio wise in my opinion is the masturbate scene. This used 3rd party sourced sounds such as the kettle and groans however the rest of the sounds were originally recorded. I really liked the kettle sound, I feel it added a layer of humor to the animation.
The Exhibition
A plan of mine since the very early idea was to separate the animation from traditional media and instead, display it on something nontraditional to emphasis the meaning of the animations.
The animation studio has an assortment of old crt monitors which will be the displays for my animation.
Recently, I was able to test some of the animations out on the screens and the results were brilliant.
This is the first time watching my animation in this format. It gives it a really nice warm-ness that the 4K colour accurate screen I created it on lacks.
The screens will be stacked on top of one another in a public place looping each animation independently.
The Exhibition Film
Conclusion
This project has been the most successful and ambitious project I have worked on to date. I have been incredibly enthusiastic for each idea and for the message as a whole, making the project not feel like work.
Of course there are parts I wish I spent more time in and other parts I feel could not be improved but as a whole, the project has been amazing to work on.
The biggest challenge and turning point for me was to switch render engines from the standard C4D renderer, to Octane 4.0. I had to pay for a subscription license for octane which was around £20/month for a total of 4 months. Without doing this, I feel my work would look completely different than how it looks now, and not for the better. Learning Octane has always been a desire for me and i’m grateful this project gave me the opportunity and allowed enough time for me to learn it.
I love how I have incorporated a different medium to display the animation on i.e the CRT Monitors. This creates new challenges for that I have not encountered before such as working in both a 16:9 and a 4:3 format. The former so I can upload them online for people to view (traditional format is 16:9) and the latter to use with the Monitors.
I have many things I would have changed or added looking back on it, particularly the thought of incorporating some Virtual Reality animation to accompany the screens as a means to escape etc. However making animation for VR is very render intensive and takes a lot of time. I also feel I should have developed more of the discontinued animations as I believe they had some potential but I didn’t pursue them.
For the 5 animations I do have I am very pleased with the results. I would have added foam to the urinal scene if possible as that would have made the animation so much more realistic but after finding it wasn’t possible I stopped. I like how each of the animations explores a different animation technique, whether it be soft body dynamics, cloth dynamics, liquid simulation and keyframed rigged animation.
I have really demonstrated my capabilities using Cinema4D on this project and as I will be using this project in my showreel, which could potentially set me up for good employment.
As for the audio, I feel it is not as strong as the animation. Partly because my knowledge of audio is a fraction of my animation knowledge and also partly I didn’t give myself as much time as I should have. I completed the whole audio in about 2 weeks with little location / object scouting. If I planned what the objects would sound like in a similar designed room that would have pulled the audio quality up a lot higher. I did contemplate getting in contact with a professional audio mixer who specifically designs audio for 3D animations (there a lot of freelance audio artists on Instagram). However part of the submission was to create your own bespoke audio solutions to accompany the animation, which I feel I have achieved.
DSpall Animation Blog 