-
‘a painting aims to record the appearance of things only so as to
capture the experience of observing them’
- Roger Scruton
A photographic image is an exact copy, a freeze frame of what the photographer see and captures for preservation or reflection. This image allows other people to see what the photographer had seen in that exact moment in time. Aided by the use of computers we strive to recreate this exact copy, photorealistic rendering focuses on the recreation of real world elements, exactly as they appear in the real world with blemishes and grain just like a photograph.
The birth of photorealism in its truest form occurred in 1822 with Joseph Nicéphore Niépce’s adoption of Johann Heinrich Schultz’s experiments with silver and chalk which evolved into the modern photograph. Before this all imagery was created by people, paintings, sketches, cave paintings dating back ‘over twenty-five thousand years ago, in the caves of southwestern Europe, Cro-Magnon man’ with the single purpose of capturing a specific moment in time through the eyes of its creator. Before 1829 and imagery was non-photorealistic, there was life and meaning engraved into the image by the artist, through their brush strokes, line quality and more importantly what the artist chose to include and exclude in the image.
‘in contrast to photorealistic approaches, aims to produce images’ which
‘stimulate not only the optical nerves, but also trigger the mental abilities
of our higher-level cognitive mechanisms’
- A. Bartesaghi & G. Sapiro
The invention of the photograph completely changed the world not just in the way we communicate but also the way we create. Although the term photorealistic was not defined until the 1960’s with the birth of the Photorealism art movement, and the term non-photorealistic was left undefined until 1994 by Winkenbach and Salesin their
SIGGRAPH paper entitled ‘Computer-Generated Pen-and-Ink Illustration’ (Winkenbach). The definition and methodology behind the creation of images changed and continues to do so the undying principle and purpose of any imagery remains the same, to communicate.
“Visual art is one of the most expressive forms of communication available.
Even a novice can create simple drawings that tell a compelling story
through little more than a collection of rough strokes that suggest
a certain character.”
- Alexander Kolliopoulos -
As technology progresses and time goes on language will continually change, imagery and it’s method of creation will evolve and in doing so become more realistic. Keeping in mind the drastic leaps in technology spanning its early existence, the advances within image creation has been pushed towards photorealism but even with this advances there will always be a need for non photorealistic imagery. Language is constantly evolving, adapting and changing; however the need for communication is not. I am suggesting that the visual world is a clearer less cluttered method communication.
The written word has changed so much through out its existence - it adapts and changes where and when needed, just as people do. Language has segregated the world and is often a slow and painful learning process. However non-photorealistic imagery is pure, clean and simple. It’s a universal form of communication with no need for conditioning, as opposed to the years of conditioning need to understand the spoken and written word. This form of communication is not natural; it was created as a form of consumption - a way of monitoring costs of goods in ancient Iran, documented on Cuneiform tablets. The written word has changed so much through out its existence - it adapts and changes where and when needed, just as people do.
The creation of imagery occurred in the later stages of human evolution, these early cave paintings depicting animals hunted by Cro-Magnon man can be refereed to as the emergence of human consciousness through the development of the brain. These early images are interpreted as being representational and symbolic; keeping this in mind when we look at modern day imagery we can notice that very little has actually changed in the purpose of imagery. Non-photorealistic imagery existed before language, its methods of creation have changed dramatically but it still exits - in fact some examples remain unchanged. From early cave painting to scribbles in a modern day note book, the meanings are the same.
Language is a vehicle in which communication travels. It’s a form of travel which can pollute the meaning, purpose and reason for communication.
Imagery however clear and true to the object it is mimicking needs a leap of faith by the viewer; its creation process is a small percent of the image itself, a large majority of the images importance is contained within interpretation.
As we grow and develop we are educated, taught skills which appeal to the logical side of the brain, but before we are entered into the education system we are free to play, to draw, to experience the world around us through copying… to Mimic. -
When we speak of logic, we often relate it to a small voice in the back of our head. When we work out a problem we have an internal dialogue which takes place where the problem is mulled over, answers suggested and the problem is solved. This internal dialogue or monologue uses language, specifically the spoken word not imagery. Language has given us the power to clarify, explain and elaborate, without it the simple task of making another person to change how or what they think becomes even more complex. The lateralization of brain function leads us to believe that the two hemispheres of the brain control different functions and characteristics, the ‘logical’ and ‘creative’. These lateral dominance is measurable but in fact these functions occur within both hemispheres but if we take the left and right side of the brain at face value we can simplify the process of problem solving and break the process down to ‘a conversation between logic and creativity, with the aim of solving a problem’ (Włodzisław Duch). While keeping in mind the complexity of site, JJ van der Leeuw illustrated the process in his essay entitled ‘The Conquest of Illusion’ focusing on the area entitled ‘world of my consciousness’ aids us in understanding how personal problem solving can be.
JJ van der Leeuw (1928). ‘The Conquest of Illusion’.
Depending on the complexity of the problem, this communication happens within an nth of a second. Meaning is given to objects through the use of written and spoken word, not through imagery, there are very few people who can retain the information contained within an image yet we can recognize them. This logic and quick problem solving is how we piece together the information and build a narrative around our surroundings, through experience we can grow and strength our ability to understand. The connections within our mind become intertwined and a memory is made, but we can subconsciously re-work this memory based on future experience and situations.
“our intelligence reworks the experience”
- Marcel Proust -
If we look into the process of simply seeing is a complex process within itself. From light stimulating the optic disk within the eye to the frontal cortex interpreting the light remained a misty until David H. Hubel and Torsten Wiesel broke the process of sight into two sections, the introduction of the image to the mind and the minds interpretation. The introduction of the image into the mind, this raw information is processed by the visual cortex until the mind has created a sense of sight and processes this information into readable image. The visual cortex interprets the information so how we see the world is in fact a filtered version of our surroundings. This process, just like many of routine tasks preformed by the human body can break; Prosopagnosia is a broken connection between the eye, visual cortex and hippocampus which helps us to understand how we read imagery.
Prosopagnosia is the inability to recognize people’s faces. People who suffer from Prosopagnosia can have no hindrance in seeing, but they can not connect people’s names, personalities or relations when they view faces. It’s as if they are seeing everyone for the first time, even their parents, spouses and longer term friends. The hippocampus is the brains memory center, once the light enters the eye and reaches the visual cortex it triggers the hippocampus and triggers memories which build and form relations ships with other memories. This process allows us to recognize loved ones, tools, places etc. Prosopagnosia does give us an insight into the memory and information within the arrangement and subtle nuances of a face is lost, the ability to distinguish a loved one from a complete stranger is impossible, but it does highly how we see and how important memory is when simply seeing.
If we take a brief look at memory itself, Santiago Ramon y Cajal defined memory in its most basic for as ‘a changed connection between two neurons’. This connection grows, builds, develop so that objects can be named, tools can be used, people remembered.
Our senses help layer these connections and work together creation stronger connections, but as imagery needs to be interpreted by the visual cortex before being passed to the hippocampus sight cannot trigger as strong a memory as say the smell of
fresh bread can take you back to your child hood. Smell and taste have a stronger connection due to the fact that they are connected directly to the hippocampus; their mark on the long term memory is permanent and never truly forgotten. Sound and particularly sight have a filtering process contained with the deferent sections within the thalamus which is structured around language and consciousness.
The process of perception has been finely scrutinized by science; we now have a better understanding how the brain works but not why. Our memories which guide us through the world are how we experience life. Negative memories will resolve in a negative experience. Jonah Lehrer explain this by stating that “the one reality science cannot reduce is the only reality we will ever know”. Experiences can be mimicked, faked or even suppressed but by look at art, whether its simple doodles or complex computer generated imagery we can express our actual experience, it is a map of our consciousness. The combination of simple lines combined with human perception and memory generate a powerful chemistry which can portray a global message yet still manage to connect with people on an extremely personal level.
“The brain-symbol interface is the birthplace of art, science, mathematics, and
most of the great institutional structures that humans have built.”
- Merlin Donald -
Science has provided us with an evolving understanding of how we work, along with guidelines and theories which aid us to appeal to select areas in our mind to evoke a stronger response. These theories also help us create appealing imagery, imagery which at first glance portrays a single meaning or statement but at closer investigation we can view the creator or artist’s personal message. Once the key message has been established and defined, detail can be layered on top of these redemptory forms to add character to the image. For example if we take a cityscape; the clearly defined overlapping forms of the buildings, juxtaposed against a clear and open sky, selling the message of a cityscape. By layering on specific monuments or landscapes the artist can then begin to give the image more meaning, creating a characterised city. With the addition of building characteristics we can also get an idea of time periods to help establish the location. Adding colour adds mood, dark and deep colours reflect a dark city and supporting narrative, light and bright colours reflects a light mood.
As we view non-photorealistic images we firstly see what the artist wants use to see, we are able to lift the initial information from the surface. If the image appeals to use or calls to a distant memory we may take a closer examination of the image and then we can begin to see the subtle nuances and messages that the artist has left behind. This highlights the non-photorealistic imagery, the less cluttered the imagery is the less likely the view is to get lost in meaningless detail and information.
We all see the exact same image; the light bounces and enters the eye exactly the same way as everyone else, but as the information splits along two pathways within the mind we being to see differently. These two pathways create the image we see in our mind, the first path take a blurry image quickly to our prefrontal cortex allowing our conscious to mull over the form and shapes. The second path is much slower, carrying information to the visual cortex which reads the line and light. With the aid of theories such as the rule of thirds we learn to interpret and read the hidden language of imagery.
Iconic memory is a form of sensory memory that occurs in the visual system. In this memory, a brief sensory impression of the visual scene is retained; this memory lasts up to one or two seconds. Iconic memory allows us to view a series of still images being flashed in front of us as being one continuous piece as opposed to a series of individual jerky stills which have no relevance to one another, iconic memory allows us to glue these images together, the more images shown the less work needed to glue the images together and in turn makes the movement smooth.
This method of memory is a key part of non-photorealistic imagery, through iconic abstraction we can take key elements on an image, object or even persons face and use this to aid our memory flow. Iconic abstraction is the process of focusing out attention on specific detail and removing additional details and information which reduces the noise when it comes to memory recall.
“By stripping down an image to its essential meaning, an artist
can amplify that meaning in a way that realistic art can’t”
- Scott McCloud
As Kimon Nicolaides describes in his book ‘The Natural Way to Draw’ there’s “a good deal more than merely looking with the eye.” it’s what the image represents, its information
and narrative. David Canfield Smith coined the phrase ‘icons’ for small for non-photorealistic images within the computer while designing the interface for Xerox’s Star Interface. The term icon originated from the Russian Orthodox Church, an icon is more than an image it embodies properties of what it represents, and it carries the information from the real life object in an artistic form.
“computer icons contained all the properties of the programs and data
represented, and therefore could be linked or acted on as if they
were the real thing.”
- T.S. Perry -
The way we view an object whether it is a piece of art, a finely designed object or a piece of nature dictates the order in which we interpret it’s meaning. Focusing on two aspects of viewing, namely nativistic perception which is based upon basic physical stimuli in relation with how the eye and brain work in matched synchrony. Nativistic perception is based upon visual events and how people pre-set determined method of organising and perceiving visual stimuli and is often referred to being hard wired in the cognitive system.
This first stage of the perception is partly independent of conscious control, we all see essentially the same object, shapes, colours and patterns are all processed in the relatively the same way by everyone. This is due to our genetic makeup, its physical chemistry as governed by the laws of the physical world and is refereed to as nativistic perception.
Once we have seen the image we then begin the second stage of perception, nativistic perception is somewhat generic in the fact that everyone sees objects in generally the same way. The second stage of perception, refereed as directed perception is a more personal occurrence. Directed perception is based upon a person’s history, knowledge and experience.
We direct our attention to the parts of the object which interest or which we can relate a past experience with, ones past knowledge and interest direct ones attention. These past experiences bring with them expectations which in turn influence what we perceive and how we interpret its meaning. These characteristics often overlap but both are necessary to coherently interpret imagery, there is however a need for concentration for the images to be fully understood. -
Our minds search for order, patterns and symmetry. In keeping with Gestalt principles “essence or shape of an entity’s complete form” our mind organized imagery into patterns of perception. As our mind intakes imagery it is searching for patterns of familiarity to trigger memory to aid the understanding of what the eye is looking at. Structuring imagery and incorporating line to guide the viewer’s eye to specific points of interest to emphasize the message. These lines and patterns are not perceived by our conscious, we do not see them as structuring or interrupting the image but we subconsciously allow the patterns to guide your mind.
In the early days of Gestalt theorist scientists discovered Reductionism and began working it into perception and reality. Reductionists understood that we could only understand the whole by breaking it apart, in contrast to Gestalts “The whole is greater than the sum of the parts”. Plato defined Reductionist aims as “cut nature at its joints, like a good butcher.”
Gestalt theorists used optical illusions as an example of perception works, such as the icon ‘Rubin’s vase’. We can view this image two ways in positive space or as two faces or as a vases in negative space but once we accept one we have difficultly seeing the other. It shows that form is dictated from the top down, in one fluid motion and not in small chunks as theorists who followed the Reductionist theories.
Neuroscience has confirmed that many of Reductionist theories are invalid, the visual cortex views sees what the eye’s intake but then process it as upped to the eyes seeing in chunks of image and then the visual cortex piecing the image back together like a
jig-saw. Reductionism has a valid flow and reasoning support but Gestalt is more in keeping with human perception.
Taking a look back over the relationship between smell and memory, this direct connection is due to a minds linguistic view of realism and mental categorization. These connections change as we learn, new connections are made and old ones forgotten, this process is how we can distinguished friends from strangers, tool and the written word. This framework of organized structures and interactions is refereed to as Schema’s.
“the schemas themselves may change with the each use”
- Mark Johnson
The combination of Gestalt theory and image schema’s we can turn minimal brush strokes and into mass stories, a nearly blank canvas in which the artist guides us through their story. We travel through our lives interacting with non-photorealistic images, merely glancing as we go but when we stop we can delve into information. From motorway signs by Margret Calvert, to Pablo Picasso’s Cubist pieces, realistic masterpieces by Leonardo Da Vinci and even animated sequences in Tom & Jerry. -
Imagery as a form of communication is less polluted method than the spoke and written word. There is no need to be conditioned to a point, but can imagery replace the written word?
There have been attempts to create a global iconic language; most notable is Otto Neurath’s ‘Isotype’ and more recently Edward Boatman ‘The Noun Project’. I think that there is indeed a need for a well structured iconic language but there is a need for explanation which needs words. Can there be a universal visual language? In an ideal world I think there could we be. In fact there is a strong structure which already exists. With the advancement of technology and the ever expanding need for rapid information within human nature we have created a strong icon based language. However this changes from technology to technology, device to device.
The physical world has been broken up into segments with oceans of separation, the world of communication has been divide through the language, I think this language divide is unfortunate supported within the image based world but with technology leading the communication world creating a smaller physical world with people being about to communicate with other across the globe with the click of a button and the technological world being monitories the hurdle which standing in the way of one visual language might be too great to concur. However I think that the pre-existing spoken language barrier will translate to an iconic language, as Betty Edwards states “what we see is changed, interpreted, or conceptualized in ways that depend on a person’s training, mind-set, and past experiences”. It’s not so much how we see but how what we see effects us, small changes in culture can relate to mass changes in perception and result in a lose or misunderstanding of information.
“the brain frequently does the expecting and the deciding, without our
conscious awareness, and then alters or rearranges—or even simply
disregards—the raw data of vision that hits the retina.”
- Betty Edwards
Throughout this text I have been referring to imagery in the broadest scene of the term, if we focus upon art. Art emerged as a result of the brain being able to hold images, to create and see images within the ‘minds-eye’ and the ability to recreate these externally. Theories such as golden ratio, face-ism ratio and the rule of thirds are merely guidelines, discovered through the study of perception, this is not to say that these theories are redundant but simple the creation on art should be done first and foremost for arts sake with theories and structure applied throughout out the creation process to help structure and define the piece.
Our understanding of the cognitive neuroscience of perception is constantly growing, which in turn changes how technology evolves and the more knowledge we have about communication. Theories are merely guidelines to aid in the clarity of creation, they are not mathematical formula to create art.
“New scientific theories have come and gone, but this
art endures, as wise and resonant as ever.” - Jonah Lehrer
Non-photorealistic imagery has a specific role to play not only in entertainment and art but also in our day to day lives. It’s an additional language we use to help portray our emotions, feelings and messages in a pure uncluttered form although imagery alone can lead to an abundance of confusion when guidance is needed for clarification.
“An icon is a symbol equally incomprehensible in all human languages.
There’s a reason why humans invented phonetic languages.”
- Jef Raskin
As Jef Raskin refers to ‘an icon’ as something that translates to any form of non-photorealistic images, then phonetic language is there as a safety net allowing the artist to explain their work or other viewers to share in and reflect in a social method.
We also have to take into consideration the detail when creating imagery, using McClouds method of iconic abstraction we as artists have to understand when enough is enough. The additional detail may add to the piece but it may also pollute it. Herbert Simons states that ‘a wealth of information creates a poverty of attention’ and not just focus on the imagery but now how we allocate the viewers attention which leads us back to the theories and guide such as the rule of thirds to aid in the channeling of information. As Jonah Lehrer puts it, ‘great art is not an accident’. -
‘Brutus’ a short animated film created by:
Adam McAllister
Gerard Dunleavy
Priya Mistry
Brief?
The brief was loose but important, based around the Olympics the short had to be no longer than 120 seconds and no shorter than 30. Brutus was part of an ‘Internet Whispers’ project organized by Cinemagic who selected three collages to take part the in series, Letterkenny Institute of Technology based in Letterkenny, University of Ulster and Ex’pression College based in California. We were supplied with Letterkenny’s last scene which we had to work into the beginning of our piece. We had two months to complete the short before handing our last scene to Ex’pression who then used our scene as the intro to their short. As Gerard and myself were the only production people working on our short we had to be smart about story limiting it too one character with a simple style, Letterkenny had a team of twelve and Ex’pression had a team of over 20.
What is the film about about?
The short ‘sketch’ is based around an out of shape olympic hero, set in the beginning of the Olympic era. Brutus, are lead character was once a gold medal winner but the years have not been kind to him. This short depicts one of a series of attempts to keep his golden medals.
How long did it take?
From start to finish the short took just under two months.
Production Overview.
My role within the team was story development at the beginning, we took the time and effort to work as a team to develop the story. As we has such a tight deadline we worked out the narrative and drew out basic storyboard within an hour of our first meeting with Cinemagic and Alaister. Later that day we worked on a digital storyboard and timed it out as a rough animatic. We then meet with a character designer and illustrator Priya Mistry and talked through concept, went over some mood boards and briefed her on the project.
Gerard and myself then talked over the draw backs of The Labyrinth of Language, highlighting the key time wasters namely rigging, rendering and animation transfer. We decided not to use plug=in’s for the rigging and instead I would rig the character from scratch. For the rendering instead of using Mental Ray and out putting as 1080HD which resulted in massive render time, we settled on Vray as a rendering engine which allowed us to shave valuable seconds off a each frame and we outputted as 720HD and upscale using InstantHD, this resulted in a massive reduction in render time, for The Labyrinth of Language we had around 200 hours of render time but for Brutus we managed 48 hours. Although the rigging from scratch can be massively time consuming process we found that mGear was still too unstable to use in production.
As we decided to stick to Maya 2011 to decreased errors when moving files between myself and Gerard we couldn’t us Maya 2012’s Alembic animation baking techniques. We baked the animation with Maya’s default tools, this allowed us to reduce file size, along with the possible loss of animation when transferring files between machines. We could also manipulate and fix small deformations which occurred in the geometry. Animation baking allows us to take the information from the animated rig and apply it directly to the vertices of the model. This meant each vert had a position value which changed from frame to frame allowing us to remove the rig and pass over ‘clean’ meshes. When it comes to rendering Maya runs through the rig frame by frame calculating positions, movements etc which adds to rendering time, by removing the rig it slightly reduced render time, but every little helps.
Once Priya had finalized her designs, Gerard was able to block out the shilluette of the character which allowed me to work on the rig along side Gerard while he refined the model.
I settle on using a ‘flat rig’ system. A flat rig which involved having a three tier joint structure. I set up a forward driven joint structure and reverse driven joint structure, these two systems would be used to drive a third joint structure through a system of constraints and set driven key controllers. This system takes longer to set up but allows rigs to be quickly rebuilt if needed without breaking the geometry as well as allowing the rig to transfer between software.
Once Gerard completed the model I then weighted the model to the rig, while Gerard worked on facial shapes for the character. Once the character was rigged I worked on building a GUI to easily control the blend shapes using expression coding, set driven keys and condition nodes to allow each facial shape to flow into each other.
Even with the tight deadline we decided to set ourselves a challenge of either using cloth simulation or particle simulation within the short. We started with cloth simulation to start, Gerard modeled a toga which I weighted to rig so that it would move with the character, this was a out default option if we couldn’t get cloth sim to work. We ran two version of cloth sim, one which was highly effected by movement and a riged version. We could then blend shape the two simulations into a clean toga mesh and paint the intensity so that the shoulder and waist area maintained rigid and wouldn’t slip off his shoulder and his chest, belly and skirt area had more influence. As we used were using geometry cacheing we could run the simulation once, bake the movement into the geometry which allowed us to control the animation with Maya instead of waiting until the scene rendered and having to re-render due to penetrations etc.
While Gerard worked on textures and look development, I moved onto animation. I started with pen and paper quickly sketches poses and focusing on shillutte and the line of action. Once I was happy with this series of sketches I moved into Maya. I worked on stepping the animation from pose to pose and then on a second pass worked on timing, each pose should tell a story and when viewed this extremely blocky, staggered animation still told the story.
From here I added additional poses, while keeping the animation stepped. I focused on anticipation leading into the key poses. I worked on anticipation, action and reaction with the action being the key poses I sketched earlier. This entire process was a massive learning process, which due to a short deadline had to cut short in order to meet the deadline.
Once Gerard had the geo-caches animation he began rendering the each scenes while I worked on the sound design. I had a small library of sounds which I collected for The Labyrinth of Language including ambient beds, foot falls, rustles etc. I sourced some crowd noises, timed out foot falls and started to layer up the cheers, I then worked on a sound bed to tie everything together so that when there was a lull in the crowed it wouldn’t sound oddly quiet. Some bird cheeps, wind blows and tree rustles later I worked on leveling everything off so there wasn’t any odd pops or differences in noises. I asked Gerard to do the voice over and moved onto cartoon sound effects for when he tugs on the rope, ball moves on grave and a doppler effect for when the hammer moves close to the camera. From here Gerard worked on adding his voice over while I managed the render files and started the compositing process.
We could then start pulling everything together, sorting rendered files, rendering final versions and adding titles. -
My role within the team was story development at the beginning, we took the time and effort to work as a team to develop the story. We spent a couple of days around a whiteboard batting around ideas before forming the initial idea. Once we had a solid idea in place Gerard worked up some story boards.
Once the boards were completed we received some feedback, made some amends and started on a pre-vis. This initially was left with myself Alan and Darren. We sourced pre-rigged blocky characters from the internet and set to work blocking out the stages of the short, and working out timing. After a couple of days we collected the playblasts from Maya and edited them together. The first cut came in at just over 2.30 mins with out deadline being a strict 1.30 mins. The three of use gradually cut it down to under 2 mins and then Gerard focused on tightening the last edit pushing it to the 1.30 mins.
We a pre-vis finalized we moved onto character design and some research and development into software and techniques we could use to product the short. The character design stage was very fluid, Alan first sketch for the grounds keeper was perfect, with a slight tweak made by Gerard when he was recreating him within zBrush Alan started work on the boy character while myself and Darren started on rigging R&D.
Darren and myself settled on using Gear, which is a SoftImage specific plugin which generates a bipedal rig using a small GUI. Having to use SoftImage caused some problems as Gerard was alot more comfortable within Maya for lighting and rendering we started to look into programs which would allow us to rig and animate within SoftImage but bring it over to Maya to light and render. We had a very loose pipeline worked out when Greg mentioned that the creator of Gear had just released a Maya version. This version was still in the very early stages of BETA and had just been realeased a few days but at the time it seemed to cut alot of awkward backing and forting between SoftImage and Maya.
Once Alan had finalized the boy concept and Gerard had built them in zBrush myself and Darren started the rigging procedure with Darren taking the man and myself taking the boy character. We started to notice some bugs with mGear very early on but thought nothing off them as we could quickly work around them. Once we had a guide in place we generated the rigs and bound them to the characters, this is were everything started to go wrong. Alot of the rig worked flawlessly, apart from scaling, inverted wrist rotation, should movement and inverted spine rotation.
These problems made the rig un-usable, we made the decision to spend a couple of days to salvage what we could as there wasn’t enough time to work on building another rig. We enlisted the help of Alan and spent several long days slowly knocking the mGear rig into a working rig. Some of the problems where simple to fix such as inverted rotation axis but it took time to workout why and how they were working, other problems were to complex to fix so we created work arounds and quick fixes. We were left with a rig which we couldn’t scale uniformly and there wasn’t a evenly disturbed twist on the spine but the rest we could work with.
Alan had put together some great blend shapes for the boy character and I took up the role of building a GUI so we could quickly animate with them. It was very rough GUI which controlled the blends using expressions which at the time I could do very quickly but once I had finished the GUI Greg showed me how he builds blend shape GUI’s using set driven keys and nice looking GUI. Unfortunately I didn’t have time to incorporate Greg’s method into our short but I will be using it in future.
This left us only a few short weeks to paint weights, rig additional geometry, animate and render. While Alan, Darren and myself worked on the rig Gerard was busy beavering away with look development and prop building. We looked into ways of geo caching the animation once it was done so we could apply it to a clean model and save render time or take it into SoftImage etc, its industry standard but due to mGear not being build in a style needed for geo caching to work we asked Greg if he knew of any ways to fix it. He very helpfully supplied us with a python script which took the mGear rig and parented the bound joints to a collection of floating joints in a single hierarchy. This worked for the body of the characters but uniformity not for any of the props or additional geo. Time was getting really tight so Gerard came in as team lead and decided to move on and work without geo caching the characters.
Within a week and a half Alan, Darren and myself animated the 33 scenes coming in at just over 1.30 mins, it was a long and painful week and a half but I thoroughly enjoyed every minute of animation. Myself and Alan worked on the boy and Darren on the man, I treated Alan as my lead animator showing him my work at the end of each day as a unofficial dailies. We couldn’t revisit any of the animation once it was signed off by Gerard due to the tight deadline which is really unfortunate.
With the animation completed we help Gerard add textures to the characters in all 33 scenes as well as add lights and background geometry. This took some time due to the massive file sizes and would of took seconds if we could of got geo caching to work correctly but we trucked on through.
Once a shot was lit and textured we started rendering, which was anther mammoth task. As ELIA needed to short rendered in 1080 it took render time up to roughly 6 minutes per frame, with 25 frames per second and 1.30 mins to render we were very glad to have Greg’s animation room with 16 machines to render on.
We rendered out as multi pass images which came in at hundreds of megabytes per frame which made transportation between machine very time consuming so we took it in shifts to gather a scene, bring it into Nuke, grade and tweak before exporting as a tiff sequence which brought the file size down dramatically.
As the scenes were rendering we dropped in the completed files into the After Effects file which we used for the pre-vis and replaced the pre-vis files with the final renders. We then started working on sound while Gerard was pushing through the last of the scene to light and render.Last up was sounds design, with only days left until the deadline this was a rush job. We had originally went to NI Screen who agreed to fund us for sound design and possibly a score, we then went to KaBoom, a local sound studio who agreed to take on the project if we got funding. When we went back to NI Screen they said they could not fund us due to the project having already been started. This left us pretty stuck, so we bought some sounds bites and effects, a little intro music and had a stab a it ourselves. Darren took the role of the man and his nephew helped us by taking the vocals of the boy. There was a little lost in communication and we lost the distinctive ‘Beep’ and ‘Bop’ but time didn’t allow us to amend.
The project was submitted on time with minutes to spare and we had some time to relax, sleep and revaluate. We over looked some massive points which turned out to hinder us and badly effect the project, the main ones are:- Team work - set working times with dailies
- Pipe line - clearly set up a production pipe line and stick to it, plan out the process, problems and steps.
- Play people to their strengths - animators should be lead and sign off shots
- Rig your own work
Having had time to reflect I think the project could of been alot better, especially the animation, two weeks is not enough time to get into the characters head and reasoning. The short went down great in Berlin at the screening and was screened at an event in November held by the ITLG, we even projected it onto the side of ‘The Mac’ at the Art College just like a drive through movie.
I had an amazing time, even with all the late nights, lack of sleep and sun and I would gladly do it again.
Adam McAllister
Notes & experiences on the graphic communication
Thoughts and progress log of Adam McAllister, a student studying for an MA in Multidisciplinary Design.
