Programme Of Study (Draft)

Programme of Study Title

An investigation into the collaboration between art, science and technology (3d computer graphics and animation) in the field of visualisation.

Project Rationale

My undergraduate study provided a strong understanding of 3D computer graphics, and developed practical experience of the most common areas – including modelling, texturing, lighting, rigging and animation – but did not focus on the various applications of these skills. My intention as a postgraduate student is to continue developing skills as a digital artist (working with advanced 3D techniques) and apply skills in 3D computer graphics to research into visualisation. This research is important because 3D computer graphics and animation can now be used to communicate complex ideas (often medical or scientific) to an audience who is not trained in the relevant area, using more user-friendly visual methods of displaying information/data.

Aims

The aim of this research is to explore the use of 3D computer graphics and animation, in the field of visualisation.

Objectives

To explore relevant examples of visualisation which use 3d computer graphics (to help communicate their message).

To research the tools and techniques used in creating 3d computer graphics for visualisation purposes.

To create 3d computer graphics that can communicate scientific data sets, using more visual methods.

To identify the benefits of using 3d computer graphics in visualising complex scientific data.

To explore future applications of 3d computer graphics as a beneficial visualisation tool.

Research Strategy

Research examples of visualisation, which make use of 3D computer graphics, and explore how the use of 3D has improved visual communication of complex information. Sources will include related books, academic journals and conference papers (such as SIGGRAPH). Online examples of visualisation will also provide a sense of what is currently being done, and how. All research will be documented and reviewed as part of the final written deliverable.

Research will be supported with practice-based learning – exploring the tools and techniques used in creating 3D computer graphics for visualisation. This could include exploring advanced techniques in Autodesk Maya, such as using dynamics, particles, and MEL scripting to further skills as a digital artist (dependant upon research project). Additional skills in rendering complex scenes and compositing may also be required.

Although the University will offer some classes, self-directed learning will be part of developing technical abilities – undertaking tutorials from books/DVDs/online articles.

All personal development will be reinforced by creating examples of the methods and techniques used in visualisation, which will potentially applied as part of a directed visualisation research project.

Undertaking a collaborative visualisation research project will provide opportunities to develop the necessary understanding of communicating complex ideas visually, and showcase technical skills in 3D computer graphics and animation - whilst working towards a brief, and within a group dynamic. This work will potentially open up new areas for future development or continuing research beyond Masters level.

Expected Outcome(s)

Strong technical toolset, suited to continuing research into visualisation.

Depth of knowledge in using 3D computer graphics for visualisation.

Series of short video clips, generated from collaborative visualisation projects.

Schedule

Stage 1 - Develop personal skills and understanding, relevant to the field of visualisation.

Stage 2 - Apply knowledge to working with scientific data using 3d computer graphics.

Stage 3 - Critically evaluate the outcome of all learning and collaborative projects.

(these stages do not necessarily represent semesters, and may be cycled for multiple short projects if necessary)

Key Challenges and Issues

Developing a wide range of abilities in a limited amount of time.

Collaboration with other (non-artistic) disciplines – essential for gathering information for visualisation, however difficulties in communication can hinder progress.

Resources

Access to both online tutorials (e.g. Digital Tutors) and offline (e.g. Gnomon Workshop DVDs)

Availability of books/papers/journals/etc. (e.g. In Silico, Studio Projects)

Scientific data-sets generated by collaborative teams (e.g. cell visualisation alongside mathematics division)

Reference Materials

Palamar, T., (2010), Maya studio projects: Dynamics, Hoboken, N.J.: Wiley

Sharpe, J., (2008), In Silico: 3D animation and simulation of cell biology with Maya and MEL, London: Morgan Kauffman

Reflection On "The Story So Far..."

In a recent post, titled "The Story So Far..." I talked about the work I have been doing, and the progress I have made with the various sources of learning. This post aims to expand on this, with some added reflection on my programme of study, and the 'journey' so far.

Initially, I commenced my studies at DJCAD with the intention of focusing on 3D animation - specifically gaining a better understanding of the principles of animation, and how to apply them in being a better animator. This interest came from the work I had undertaken as part of my Honours studies, where I created a short animation depicting a story from Greek mythology - "Theseus And The Minotaur" (this was an individual project).

However, after taking advantage of the opportunity to attend presentations by a wide range of people, I gained an insight into using 3D computer graphics as a visualisation tool - this was thanks to an insight into the work being done by John McGhee and Chris Rowland, although I was more interested in the idea of biomedical visualisation.

This change in direction forced me to change my programme of study to something more appropriate, and as discussed in "The Story So Far..." I have since spent time developing the necessary skills in 3D computer graphics.

These skills have been relevant to my programme of study, as I am currently involved in two projects which make use of this new technical undertstanding. The first project is in collaboration with the University mathematics division, and involves developing more visual methods of visualising mathematical data (more specifically, cell visualisation). The second project is a visual-effects based project, in collaboration with another MSc student. As part of this project, I am responsible for 3D modelling, and creating a dynamics simulation.

Although each of these projects have had difficulties, good progress has been made and the experience has been invaluable. If the projects did not have any problems, I would not have learned nearly as much as I have, and I would not be as prepared for future projects as I am now.

Now that I have a stronger understanding of 3D computer graphics and their use in visualisation, I can focus on the application of these skills, and concentrate on completing these projects.

Moving forwards, I would like to continue developing my 3D abilities, and gear this specifically towards using RealFlow for advanced dynamics simulation (allowing me more flexibility in the type of visuals I can create) and learning MEL scripting within Maya (potentially giving me the option to automatically generate visuals from huge amounts of mathematical data).

Considering the drastic change of direction in my programme of study, I am glad I started developing skills in visualisation. Moving forwards, I am excited by the range of projects out there, and look forward to developing my own visual style.

Visualisation Techniques : Cells (Continued)

Continuing my experimentation with how my cells could look (first post here), here are three more examples which all use a spherical soft-body as a starting point. These examples also show a change in colour - something relevant to the mathematical data I am working with.

The first example combines previous render techniques, and uses a Cloud shader applied to the particles. Unfortuantely this gave the cell a glowing appearance, and had no distinct shape or outline.

The second example instead uses particles which are invisible, using a Blinn shader applied to the soft-body surface directly (the particles are used solely to drive the animation of the cell). A 2D fractal was used as a bump-map, ensuring that the surface was not too smooth and plain.

The third and final example builds on the second, using the same Blinn shader, also applied to the surface. The difference is that the surface material is created using a Layered Shader, which uses the original Blinn (made almost transparent) and a second copy which uses a Ramp Shader to adjust the transparency based on the object's facing ratio (making the shader less transparent towards the object edges). Although there are two shaders layered here, it gives a more interesting look - a transparent looking cell with a clearly defined outline.

Research Skills & Methods : Research Poster

The third and final assignment of this module required me to create a poster which would communicate my research using visual methods. The poster built on work completed in the first two tasks (here and here), and is shown below;

Context & Review : Comments

Please add any comments and/or feedback from the blog presentation on 24-Nov-10 to this post.

Playing With Fire

Adding more special effects to my arsenal, Studio Projects Dynamics moves onto creating and controlling fire (using Maya fluids). Using fire is something which will be particularly relevant to a side-project I am currently involved in, and one of the tutorials involves burning down a 3D model of a house, which hopefully will be useful.

Starting off slowly, the first example introduces using fuel and heat within a fluid container. As this is not particularly exciting by itself, there is not much to show, so below we can see the second example - a simple flame created entirely in CG;

After working with a simple flame, a preset was created, and then imported into a new scene featuring a simple house model (provided on the tutorial disc). The flames were scaled up and tweaked to suit the new size. The house structure was already broken in segments, so these were converted to nCloth components, ready for destruction. Randomness was applied to the building 'burning down' by using keyframed ramp shaders with noise applied. The finished render can be seen below;

This chapter was one of the most important topics I have covered so far (and very relevant to the side-project) which allows me to transfer the skills learned during this tutorial, and apply them in a more creative manner.

Inspiration 1 : Cell Visualisation

In collaboration with the University mathematics division, I am working on cell visualisation - starting off with mathematically generated data, I am importing this into Maya and defining the aesthetics of the scene, making the data more accessible and visually exciting.

Alongside my own work, I have found some examples of cell visualisation that I am particularly interested in. The first of these is a clip called "The Inner Life of the Cell", created in 2006 for Harvard biology students, by a company called BioVisions. Although this animated sequence looks dated, compared to today's standards, the content (and it's importance) are still just as relevant today. A tremendous amount of effort was put into to this project, and those working on it were constantly aware of the relationship between the quality of the visuals and the accuracy of the data. One criticism I would make, is that the scenes are often very 'busy' and feature lots of moving items and lots of different colours. Although this means there is more to look at, it can also make the shots somewhat confusing, as there is no clear focus. "The Inner Life of the Cell" can be seen below;

BioVisions have also continued working on molecular animations, with their latest video titled "Powering the Cell: Mitochondria" (a clip can be viewed here). This video is a significant update to the other one above, primarily thanks to the improvements in technology over the last four years. Although the concept is the same, the video has been output in high-defination, and this is definately a noticeable improvement. The visual style has also 'quietened' down somewhat, and is much more pleasing to the eye, as can be seen in the image below;

Moving away from this type of visualisation, I am particularly fond of "Nature by Numbers" created by Cristobal Vila. This is an expertly created piece of work, and focuses on how nature is driven by mathematics (at it's core). The content is of excellent quality, and there are segments where it appears that some sort of dynamics system has been used to drive the animation - something I am currently developing skills in. The overall look of the video has a very polished feel, something I would certainly hope to achieve by the end of my MSc programme! The video can be seen, in all it's high-definition glory, below;

After looking at other examples of work out there, it is clear to see that there is a great deal of importance placed on both the quality/accuracy of the data, and the appeal of the visuals. Trying to find this balance however, can pose difficult, and it is important for an artist to find an individual style which suits them. As mentioned in a previous post, this reinforces the importance of experimention - practice makes perfect.

Visualisation Techniques : Cells

Experimentation is often the key to success, and computer graphics are no exception...

Recently, I have been manually building a 3D scene from some mathematical data (due to problems using MEL to import this automatically), so I needed to start work on how the data could be represented.

This first data-set features cells which multiply over time, and also change colour (which represents their type, or stage). Previously I had already tested changing particle colours (early results of this can be found here) so it was time to experiment with how these spherical 'cells' might look.

Working separately to the data setup earlier, I started off creating a polygonal sphere, and using a surface emitter to create the particles. I started by key-framing the emission rate, but then decided to output the required amount of spheres and set the initial state (so that we did not see the creation of the particles). I created a simply three-point lighting setup, and animated a camera moving through a 90-degree arc - the scene was now ready for aesthetic testing.

This first example shows the particles, rendered using the Blobby Surface render type. The Radius and Threshold were then key-framed and oscillated, to create a moving, pulsing surface. Ideally I wanted each particle to pulse individually, but I had difficulties in doing this. Using a Blobby Surface created a simple effect, with required very little computational time on render - something which might outweigh the 'awkward' pulse effect when hundreds of cells are required (and can pulse at different intervals to each other).

This second example builds on the first, although uses the Cloud render type. This was combined with a Lambert surface, and used the same animated Radius and Threshold. The cell was animated to rotate on the XYZ-axis, giving more variance visually. I preferred the effect created here, as it seemed more random, but there was not enough definition in the shadows or highlights, forcing the cell to appear flatter than it actually is. Also, several 'holes' appeared in the surface, which was an unwanted effect.

This third and final example is a development of the second, and uses a Ramp shader instead of a Lambert shader combined with a Particle Cloud node. The Ramp shader used the 'glass' preset, and was recoloured to be more neutral. I found that this video looked the best, and gave almost a glass-like look to the cell, with visible shadows and strong specular highlights. When the cell turns green, the glass outer-casing becomes more apparent, something which contributed to the overall style of the cell.

Although great progress has been made here, and I particularly like the third example, it took considerably longer to render. Also, the cell still 'pulsed' in an unnatural fashion - something I would like to correct moving forwards.

Although more work still needs to be done, I happy with the results so far, and it is always good to see progress being made.

The Story So Far...

So far, the majority of my work has focused on developing skills which are crucial in creating 3D computer generated visualisation. Having such a strong background in 3D has helped greatly, and has allowed me to focus on learning advanced techniques and understanding new concepts, as opposed to having to learn new software.

The key goal is to rapidly develop 3D skills this semester ('exploring practice') and although I will start work on related projects, these will take shape later in the academic year, as my abilities grow stronger.

My learning has been mostly self-directed, using a variety of sources. Progress using these are detailed below;

• In Silico - recommended as a starting point, this book gave me a brief introduction to using particles, and how to shade, light and render a scene containing medical data. Most of the content here is relevant to cell visualisation, and will be extremely useful as my projects develop. Some simple examples were created, and uploaded to my Vimeo page. Upon reaching the MEL sections, I realised that these assumed previous knowledge of scripting in Maya - since this was new to me, I chose to put this resource on hold temporarily. Beyond the MEL chapters, awaits some more complex tutorials on working with medical visualisation data.
• Gnomon Dynamics - I started using this material, and soon discovered that the content was of poor visual quality, and at points difficult to follow. As other sources were available, I chose not to continue working on the Gnomon material. However, I may return to this when looking at soft-bodies, as Digital Tutors does not cover this in great detail.
• Studio Projects Dynamics - another recommended learning source, I found this material relevant and easy to learn, developing skills in using particles, nParticles, nCloth and fluids. Tutorials applied skills to interesting examples, such as building a sandbox with moving contents or creating realistic looking tornadoes and volcanoes. I still have to approach the most relevant chapter, titled 'Playing With Fire' - once this has been completed, I will be in a better position to approach 'burning down' a computer-generated house.
• Digital Tutors 'Introduction to Dynamics in Maya' - this has been the most useful source of learning so far. Content is clear, easy to follow and explained in-depth. Although there has been a slight overlap in some areas (the implementation of a particle system mainly), there has been more explanation of working with the features in Maya. The tutorials also discussed the various effects available (such as fire or smoke) and introduced the different particle render modes. These topics have given me the confidence in experimenting with cell aesthetics, something which will develop throughout my programme of study (videos to be added shortly).
• Digital Tutors 'Getting Started With Nuke 6' - although not strictly essential, I chose to gain a basic understanding of node-based compositing using Nuke. This will be beneficial later in my work, when it comes to creating polished videos for presentation. 

Moving forwards, I will continue using most of the above sources, particularly Digital Tutors. I plan to finish off working with dynamics in Maya, and then return to MEL scripting, where hopefully I can begin to make sense of the content. Something of particular interest, is using RealFlow - Digital Tutors has a huge amount of lessons available, so this is something I would definitely like to experiment with.

Looking at the subjects I have studied, and the progress I have made, I am glad I have chosen this programme of study - I have enjoyed this new content, and know that I will continue to enjoy this moving forwards.

Context & Review (Atmosphere/Mood/Design) : The Nightmare Before Christmas

Working individually this time, I was tasked with choosing a short film or clip that showed Atmosphere/Mood/Design in an interesting way.

After much deliberation, I decided to go with the 1993 stop-motion classic - "Tim Burton's The Nightmare Before Christmas". I decided to use the film's introduction, which featured the song "This Is Halloween", sung by many of the characters 'starring' in the movie. Unfortunately, the HD clip could not be embedded, so the SD version can be seen below;

Alternatively, if you absolutely must see the HD version, it can be found here.

Starting with the atmosphere of the clip, the story begins in a abandoned forest which features a narrator introducing the story (setting the scene). This use of narration immediately gives the film a fairytale atmosphere, and (almost) prepares us for the visual diversity and creative setting we are about to enter.

As we travel through the graveyard, one of the most prominent visual features is the carefully controlled lighting. There are lots of shadows and areas of darkness, which add to the anxiety of the scene, playing on the human fear of the unknown - there could quite easily be monsters hiding amongst the shadows. This effect is continued, with silhouette characters appearing on gravestones - we know they are monsters, but without details, our mind creates a stronger monster than is probably there.

After this graveyard scene, we are then introduced to monsters we can see, although these 'conform' to familiar stereotypes - such as the monster hiding under the bed, vampires or werewolves. The careful use of lighting continues here, and spotlights are used to draw focus and lead us through the scene. Coloured lighting is also used to highlight particular characters or props (such as the well, which glows green).

All of these elements contribute to enhancing the atmosphere of the world we are entering, which is clearly a dark, gloomy, monster-infested graveyard.

However, the mood of the clip is very different, and this is primarily thanks to the cabaret-like style of the song and music. Although the lyrics describe the atmosphere, and introduce the monsters, they do so in a light-hearted way. Bearing in mind the likelihood of a younger audience, this ensures that the night-time graveyard does not become too scary, and encourages viewers to continue watching.

The design of the characters and locations is exquisite. Although there are a huge amount of characters introduced in such a short space of time, they are easily identifiable thanks to a large amount of diversity in the character design. Also, all of the buildings are skew, and out of alignment, adding to the visual interest already created.

Most importantly however, is the careful use of exaggeration. Using Jack Skellington as an example, he is a skeleton character - although not dimensioned as he should be, he is 'extra-lanky' which makes him more interesting to look at, and certainly more memorable (he is definately a well-remembered, iconic character in today's culture). Halloween Town's Mayor is also another great example - he has different sides to his personality, represented by a head which has a face on each side, and rotates to a different expression as his emotions change.

One important point to consider here, is that even though we have this extremely colourful musical piece, during which we meet our two main characters (Jack and Sally), neither of them sing. By introducing contrast in this way, it identifies their importance, by showing that they are different from the rest of the characters.

Overall, I chose The Nightmare Before Christmas for many reasons, and without even realising what many of them were. After looking closer, it has only made me want to watch the rest of the film even more...

Snow Is Falling...

After realising Gnomons 'Dynamics' series wasn't going to be a good source of learning for me, I decided to give Digital Tutors a try. After a quick look through their material, I started on their 'Introduction to Dynamics in Maya' lessons (more information here).

Looking at the individual tutorials within this lesson, there was bound to be some overlap. However, I felt that it would provide a good opportunity to consolidate the learning I already have, and fill in any gaps in my knowledge (I find Digital Tutors to be extremely thorough in explaining features).

After completing the first 5-6 lessons, I decided to experiment with some colour techniques which will prove useful for my project with the University mathematics division. Part of this project will feature cells which change type, and each type is associated with a different colour - I needed to find a way to animate an object between colours effectively. More importantly, I needed to find a way to control the colour changes in particles, as these are more likely to be used moving forwards.

The first example shows some simple 'particle rain/snow' which has keyframed colour changes, and works very well. Although the particle effects are not what I'm looking for, I was testing colour here, and this has worked as hoped. Since I already have Lambert shaders setup with the colour changes, I also needed to find a way to use these, as particles use Particle Cloud shaders. Fortunately I can simply plug the coloured Lambert into the colour input of the Particle Cloud shader, providing an additional level of control.

The second example shows something a bit more fun. After my experimentation in the first video, I realised I could make some decent-looking snow. After modeling a basic landscape, with a hill and some simple trees, I created a snow particle effect, with a small amount of randomness applied to it. After creating this short clip, it really made me wish there was snow outside!

Context & Review (Animation) : Ren & Stimpy

Alongside my fellow zookeeper Mark Haldane, we were tasked with choosing a short clip which showed 'animation' in an interesting way.

After looking around online, we decided to go with a clip from a Ren & Stimpy episode, titled "Sven Hoek";

First things first - some of you may not be familiar with Ren & Stimpy (as it premiered in 1991), so here is a brief overview;

The show features two lively characters, Ren Hoek (a chihuahua) and Stimpson J Cat (a cat, as his name suggests). The show was known for its crazy characters, situations and events. It also featured a large amount of comic violence, alongside quite dark humour (for a cartoon). One of the regular featured techniques used throughout the show, was the inclusion of detailed 'paintings' - these would often be far more gruesome and detailed than the animation itself, adding only to the intensity of any on-screen antics.

We chose this particular episode, as it portrays the strengths of the animation style which The Ren & Stimpy Show used.

The most obvious thing to notice, is that all of the animation is extremely exaggerated. We both agreed that this was used effectively, and although it ensured that the cartoon had a crazy look/feel to it, it was not simply crazy for the sake of crazy - the exaggeration was used to clarify all of the expressions and movements, leaving very little to interpretation.

When comparing this over-exaggerated style to a real character, it becomes almost impossible to replicate realistic movement. It is for this reason, we felt that Ren & Stimpy needed an over-exaggerated style, because then their actions and expressions would be even clearer than normally possible. It also ensures that the story is being received as it should be.

Throughout the clip, Ren's anger is very obvious, and this is enhanced by theatrical movement, and strong staging/posing. It is so obvious, that whilst watching this character, it became clear that the clip could also be watched without sound, and the action would not be confused or misinterpreted.

Although this is the case, the show's creators decided to add to the intensity of the situation, by using very focused shots, which used eerie music, and removal of the scene's background to build tension. By removing the backgrounds, we are forced to relate with Ren and his anger'; by alternating backgrounds quickly, it adds to the intensity of the scene, and makes us connect with Ren's emotions quicker.

Overall, we felt that Ren & Stimpy captured the intensity of a characters emotions well, by using exaggerated animation, and slowly building tension. Also, it's hard not to love Ren & Stimpy as characters!

My colleague Mark has also added a blog entry discussing Ren & Stimpy, which can be found here.

For those interested in seeing the full episode, it can be found online in two parts; here and here.

Tornado Warning

Since there has been lots of wind and weather warnings recently, it makes the next chapter of Studio Projects Dynamics even more topical than usual!

It introduces new ways of using Maya Fluids, and how to control them using a variety of fields. These tutorials have helped create some really nice effects (not just in this chapter on Tornado Winds). The final part in this chapter recreates a tornado demolishing a house, but I have still to start this.

The first example shows a Funnel Cloud. This was created by using Fluids and a Volume Axis Field, which spins the fluid. This example was to test the circular motion, and although basic colours were applied, that was not the focus.

The second example builds on the first set of techniques, although it uses a Volume Curve field instead - this gives more control over the shape and size of the simulation. A simple curve was used alongside the volume curve field to define the shape of the tornado. This was then paired with a second, smaller fluid container (for dust and debris) at the base of the tornado. Finally, a dynamic hair curve was used to move the finished tornado around. A directional light was added, and the finished clip was rendered using Mental Ray.

Research Skills & Methods : Writing An Abstract

The second assignment for this module involved writing an abstract, outlining the key aspects of my research, within 250 words. The completed assignment can be found below;

Title

An investigation into the collaboration between art, science and technology (3D computer graphics and animation) in the field of visualisation.

Abstract

The idea of artists and scientists working together to achieve a common goal is not a new one. It is this collaboration that has not only challenged the way we perceive complex technical ideas, but also accelerated pre-existing concepts forwards, into new and exciting territories.

Cox (2008) describes the combination of art, science and technology as a “Renaissance team” – that is, a team that utilises a cross-disciplinary approach, and therefore gains a unique insight into the different ways of approaching a problem and providing appropriate solutions. We are told these ‘Renaissance teams’ often “make significant contributions to new visual research”.

However, this collaboration is not without it’s difficulties. Art and science professionals speak very different ‘languages’. Scientists often undertake quantitative research, and aim to ensure that any data gathered is kept in its purest form, whereas artists tend to be more creative and often leave ideas and concepts open to a viewer’s interpretation. When considering this from a collaborative point of view, we must choose - visualisation or representation.

This thesis will investigate contemporary examples of visualisation, and discuss the techniques and processes involved in creating this kind of work (from an artist’s perspective). Through a combination of discussion and practice-led creation, the author will explore the importance of staying true to scientific data, whilst providing an insight into visualising complex technical data - which would not ordinarily be possible without the collaboration between art, science and technology.

[Word count: 240]

References

Cox, Donna J. (2008). Using the Supercomputer to Visualize Higher Dimensions: An Artist’s Contribution to Scientific Visualization. Leonardo 41(4), 391-400. Retrieved October 8, 2010, from Project MUSE database.

Nuke-lear Launch Detected

Don't worry... it's not an actual nuke - well, not the explosive kind anyway...

Today I started using Nuke, an industry-standard node-based compositing package created by The Foundry. I already have some working knowledge of Adobe After Effects and I am familair with node-based working (thanks to Autodesk Maya), so I decided that being able to use Nuke for simple tasks would be helpful.

I started my learning alongside Digital Tutor's 'Getting started with Nuke' training materials (more information here). Now I know what you may be thinking... did I pay to access this? Thankfully not - DJCAD has kindly provided access to all of the Digital Tutors material, for the duration of my MSc programme.

I am around half-way through this introduction course, and so far the material has been clear, easy to follow and has explained all of the concepts thoroughly - making learning easy. Although I probably won't upload any examples of work, I have been working with some simple resources (provided by Digital Tutors).

Althought I may not use Nuke in my own programme of study, it's certainly handy to have a basic working knowledge, as there is a high likelihood that I will be in contact with those who use it regularly. Especially now, since we are being 'let loose' on the green screen facilities tomorrow morning...

Context & Review (Story) : Second Wind

For this session, we were tasked with choosing a short video which dealt with 'story' in an interesting way, which we would then critique, and present our findings to the group.

Working with Mark Haldane, we decided to use 'Second Wind' (by Ian Worrel) as our short film. According to the creator, 'Second Wind' was created using 2D traditional animation, aswell as being coloured using Photoshop and composited in After Effects. The video can be seen below;

This short film tells the story of a man and his giant cat - loyal companions. They do everything together - travel, eat, sleep and play. Whilst playing together, the man enters a cave and finds a small glowing sprite. He ends up playing ball with this new-found friend, upsetting the cat. The cat then becomes angry towards the sprite and squashes it beneath a giant paw. The sprite is then resurrected but glows red instead of blue and tries to attack the cat, at which point the man sacrifices himself. We are then witness to a 'dream scene' where the man is revived, but with a windmill attached to his back - the cat can then blow on this to keep the man alive, bringing them together once again.

The presentation of this piece was excellent. Characters, although simple, were engaging and well designed. The cat's emotions were 'on show' throughout, creating a strong feeling of empathy - this was particularly important as there was no dialogue. This simplicity was carried throughout the visual style, creating something unique and colourful, but dark and mysterious at the same time (the subject matter of the story being told).

After watching the short film, we discussed the content, and decided that Second Wind was a tale of many things; companionship, loyalty, new friendship, perception of betrayal, jealousy and conflict. Less importantly, we also realised it was also about 'cool eyebrows, football and a bad-ass cat' - well, thats what we thought anyway.

When trying to interpret this story, we came up with several ideas;

• By using a giant cat, we felt that a fantasy/mythical setting was created - this created a 'lighter' atmosphere, and allowed for a small amount of humour despite the seriousness of the story being told.
• The importance of the cat's jealousy/anger - did this create the situation? Without these emotions the sprite would probably not have been killed, and the events that followed would have been very different.
• The idea of shared redemption - the man saved the cat despite being more focused on his new friend, and the cat saved the man becoming his life-support.
• A controversial idea which presented the cat as selfish - the cat struggled to feed itself and relied on the man to look after it. Was the cat keeping the man alive because it needed him? Also, did the cat want the necessity of keeping him alive to make sure the man did not 'stray' again?

An important visual feature to note was the use of colour. Throughout the early stages of the animation, there was a green atmosphere. This turned to blue when the man entered the cave, and to red when the cat 'fought' the sprite. It finally returned to green as the conflict was resolved - this was a great way of signposting the  different stages of the story.

Overall, we found that this short animation was well-received and provided a good opportunity for discussion amongst the group. Mark also added a blog entry discussing 'Second Wind', and this can be found here.

For anyone interested in more information on 'Second Wind', the creator has two blog posts which contain artwork generated during production; found here and here.

In Constant Motion

In Constant Motion. A rather fitting title for a post about using dynamics...

Progress continues on my new source of learning - 'Maya Studio Projects: Dynamics'. In addition to using dynamics within a scene, I have introduced Maya's nDynamics, and made good use of the nCloth system. Although nCloth is intended (as the name suggests) for recreating fabric, it can also be used to create a variety of adaptable and tearable surfaces which rigid bodies can interact with.

I have also now started work on using Maya's Fluids feature, something which will be extremely useful moving forwards, particularly when experimenting with cell aesthetics.

Good progress has been made so far this week, and my understanding of using dynamics is really setting in. Although these seem like simple examples, I am aware I have started with the basics and intend on building strong foundations. Moving forwards, I plan to work on more complicated setups and develop my Maya skills even further.

The first example illustrates a simple 'sandbox' idea, and uses a half-cylinder as a 'bucket'. We can see the earth-like material (automatically generated nParticles) moving around without any key-framed animation - it is all dynamically generated, and moves, collides and flows as appropriate.

The second example takes this idea further and adds an nCloth plane, with the idea of dropping a heavy sphere onto the ground, which then smashes, sending shards flying. The nParticles are used to provide a more accurate simulation - if they were not present, the ball would simply drop through the plane... this would not happen in real life!

The third example shows a volcano erupting. Relying heavily on the use of a 3D container for Fluids,  attributes were manually configured to create the smoke, and an nCloth 'explosion' was used to create the initial blast.