Category Archives: Studio 3

Presenting 3D Models In a Showreel and Online

In such a highly competitive industry, it is imperative that your work stands out from the crowd. Unfortunately, great models can be overlooked due to poor presentation. It is for this reason that we should consider how we present our work, especially if we are trying to catch the eye of potential employers.

Let’s look at an example of good presentation:


Hugh Holder’s ‘Gabriel Costume‘ Model

The presentation of this work is excellent. It uses good lighting, is clean and simple and provides all the information that a potential employer may be looking for. To fully understand how to correctly present a model, I will take a look at the different aspects the make up good presentation.


When presenting a model there are two main sets of images that you will be producing: some form of turnaround (with texture, wireframe etc.) and a display render (which may be a posed or action shot render). The latter is more of a splash image that looks awesome and might show a character in a cool pose or a building from a dramatic angle. It is designed to catch the eye but might not convey all the information or even display the whole model: this is the job of the turnaround. In both cases lighting is essential but you might use different setups for each. An example is below:

The display render uses washed out, pink lighting:


While the turnaround render uses natural colored, standard 3 point lighting:


So with this in mind let’s take a look at lighting.

Lighting is of the most critical aspect for presenting a model. Models exist in 3D space so we expect them to interact with both light and shadow. In the case of the turnaround, the lighting should help pick up on the model’s details and give it enough depth while still making the image readable. Conversely, the display render can utilize more extreme lighting in order to enhance certain aspects of the model or to convey something about it.


When lighting the turnaround of a character, 3 point lighting is most commonly used as it nicely defines the model without being too extreme (Marmoset, 2016).

Just as the name suggests, 3-point lighting consists of three different lights: a key light, fill light and rim light. These are defined below:

  • Key light: the main light source – has a high intensity
  • Fill light: the secondary source which ‘fills’ in the shadows created by the key light – has a low intensity (half that of the key light)
  • Rim light / Back light: placed behind the subject, this light lights the rim edges of the model – has a very high intensity (double that of the key light)

(Marmoset, 2016).

The following GIF demonstrates the effect that each of these lights have:


For this to work correctly, the lights need to be placed in the proper way: the key light in front of the subject on a 45 degree angle (left or right of the camera); the rim light placed behind the subject, opposite to the camera; and the fill light placed in front of the subject on a 45 degree angle (on the side of camera opposite to the key light). This is shown below:


Additionally, a coloured ambient light can be used to help create atmosphere. Using this setup allows the model to be shown cleanly while still having depth.


A plain coloured background is standard for the turnaround. Pure black or white do not look as good as they are a little extreme. Additionally, a HDRI map can be used to simulate an environment and give greater depth to specular reflections.

For the turnaround it is much better if the model seems grounded in some way: whether this be by having the model sit on a stand or having them cast shadows onto the ground. This helps to establish their physicality and generally looks neater. See the example below:


If multiple turnarounds are going to be created (one with a beauty pass, one a wireframe, one with the normal map etc.) the same lighting setup and background should be used as it helps with consistency and overall cleanness of the composition.


Just as in film and photography, lighting is one small aspect of the overall piece that contributes to both the implied narrative and composition (Blender Guru, 2016). The choice of colour and the positioning and intensity of the lights helps to create an atmosphere and set the mood. Lighting should suit the subject matter (whether it be a character, object or environment) and convey some additional information (whether that be personality, setting, time of day or emotion) (Blender Guru, 2016).

The Blender Guru (2016) states that we should consider 5 points when setting up the lighting:

  1. What is the purpose of the render?
  2. Is there a implied narrative or story to this image?
  3. What is the focal point?
  4. What mood are you trying to convey?
  5. What lighting style would achieve and suit the above answers?

In the example below, see how the use of purple and teal lighting helps give the astronaut model a sci-fi feel and establishes the outer-space setting. This simple choice helps to set the scene and suits the character of the astronaut:


Model by Frederik A. Plucinski

If a piece is moody and dramatic, dark high-contrast lighting might be the way to go. Note how the shadows obscure Batman’s face to help show that he is brooding and mysterious, while the light catches on his bat ears and muscles thus drawing the viewer’s eye:


Model by Maxime Gainche

Similarly the dark lighting with sharp highlights help to give this sci-fi interior an ominous atmosphere:

Model by Kai Du

Model by Kai Du

On the other hand, a more light-hearted piece might require softer, brighter lighting. Note how the soft lighting on the body and face helps to establish Fiona as a bright and open character:


Model by Vincent Dromart

Similarly, the soft, warm lighting and use of bloom helps convey a calm and pleasant mood:


Model by Encho Enchev

Carefully considering what you are going for with the display render and adjusting the colouring, positioning and intensity of the lights to suit will help to give the final render much more visual impact.


  • The posing of the character (if applicable)
  • The background – this should suit the mood, story, personality or setting.
  • Framing – this should suit the mood, story, personality or setting.
  • Overall composition


Flipping through Artstation, I have noticed a couple of things. I find that the best or most ‘viewable’ 3D character models are ones that contain multiple images. I ones that catch my eye usually contain around 5 – 8  items and include:

1) A display image: a high quality render of the (posed) model with approiprate lighting and background

Model by Per Haagensen

Model by Per Haagensen

2) A full length turnaround of the character with textures and 3 point lighting

Model by Hugh Holder

Model by Hugh Holder

3) Full length turnaround of the character with a wire frame

Model by Miles Wadsworth

Model by Miles Wadsworth

4) Close ups of details

Model by Blair Armitage

Model by Blair Armitage

5) Sketchfab or Marmoset viewer

Additionally, some artists may include a video showing an animation or turn-around, some development work, a breakdown of the design, a breakdown of the textures/maps or some concept art. Including these help show the process by which the model was created.

Model by Andy Chin

Model by Andy Chin

It should be noted that while this format works well for characters, other models, such as environments, may take a different format but usually include a high quality version, textured version, wireframe version and close ups.


Model by Tim Moreels

I feel that this is a pretty good format to present the model as it catches the viewer’s eye, lets them quickly see all the information in 2D and then lets them view the model in 3D. It allows a potential employer to view everything they need without have to scroll excessively.

For this reason, I will try to post my finished model on Artstation using this format (although I may not have the texturing finalised until the holidays).


  • Have your name and contact details written legibly and discreetly on the bottom of the image and don’t use watermarks (Anhut, 2014).
  • Be clear if it was a group project, give credit where it is due and detail your role and work in the project.
  • Keep the overall presentation clean and simple.



There are no hard and fast rules for displaying models in a showreel. However, after looking around a bit I have found what I would consider a good format in which to display a model.

1) A display render (can just be a still) with nice lighting and a background


Model by Aditya Parab

2) A full-length beauty turnaround


Model by Aditya Parab

3) A full-length gray/clay turnaround


Model by Aditya Parab

4) A full-length wireframe turnaround


Model by Aditya Parab

5) A slower, closeup turnaround 

Alternatively, several turnarounds may be shown at once. This is very effective and helps cut down the time of the showreel:


Model by Patrick Kilcher

Overall, it looks professional if the showreel is kept nice and short with clean transitions between the different turnarounds or models.

Of course, what you choose to display may depend on what job you are aiming for or what you are trying to showcase. More closeup shots or even closeup stills might be needed to display a particular detail. Some artists may also like to showcase their concept art, texture maps, rigging or animation.


Model by Aditya Parab

This format allows the artist to quickly showcase the model and display all necessary information, such as topography and textures, to the viewers. I will definately be using this format when putting together my showreel.


  • Keep the overall presentation clean, simple and consistent.
  • Ground the model by using a stand or having them cast shadows.
  • If necessary, include project details such as time taken or polygon count.
  • Have your name and contact details displayed legibly and cleanly without extreme motion graphics or effects.
  • Be clear if it was a group project, give credit where it is due and detail your role and work in the project.
  • Do not use obnoxious music.



For my specialisation project, I am creating a textured character model that will be eventually displayed both in my showreel and online (on Artstation). I am planning on using 3-point lighting and a HDRI map for the environment to render my beauty, wireframe and clay turnarounds. Using Premiere Pro I will be create a nice transition between them to put into my showreel. Eventually I would like to rig and pose my character for a display image but will not be doing that this trimester due to lack of time.

Below is my camera and light setup for the full length turnaround:


I have used three direct lights for the key, fill and rim lights; one skylight (as an ambient light) and a camera. These have all been linked to the dummy in the center so that it is the only object I have to animate in order to create the turnaround.


Below is a render of my (unfinished) model without the three-point lighting:


And the following image is a render with the three-point lighting:


When comparing these images, it is evident how much flatter the first is compared to the one with three-point lighting. I can see how much of a difference using 3-point lighting makes as it helps the model ‘pop’ and gives a greater sense of depth. I will definitely be using this lighting setup to render my models in the future.


Anhut, A. (2014). How To Build Your Portfolio?. Retrieved 7th May, 2016 from

Blender Guru. (2016). How To Correctly Light a 3d Model. Retrieved 7th May, 2016 from

Marmoset. (2015). Character Lighting Tutorial. Retrieved 7th May, 2016 from

A Look at OpenSubdiv


OpenSubdiv is an open source tool that quickly and easily subdivides and smoothes a 3D model through the process of interpolation. It utilises the Catmull-Clark method of subdivision and has been widely adopted by the industry due to its efficiency (Pixar, 2016). 


A simple tube with and without OpenSubdiv

Originally developed by Pixar, it was released to the public as an open source library and, since then, has been incorporated into software such as Autodesk’s 3DsMax and Maya and Blender (FX Guide, 2015).


Geri’s Game (1996)

The first instance of OpenSubdiv was developed by Tony DeRose and Tien Truong for the Pixar short Geri’s Game (Pixar, 2016). Pixar released it as open source in the hopes that it would be adopted by the industry and thus gain hardware support, as well as to increase efficiency throughout the industry (FX Guide, 2015).


As OpenSubdiv was originally developed for character modelling, and because it smoothes and interpolates the mesh, it is most commonly used for organic modelling (FX Guide, 2015). It has been used for many Pixar characters such as Woody from Toystory.


Woody with OpenSubdiv

More interestingly, OpenSubdiv has been used on non-organic characters, such as Steve McQueen from Cars and WALL-E, to achieve a high-poly look while still being manageable to animate (FX Guide, 2015). 


WALL-E with OpenSubdiv and CreaseSet

WALL-E’s many details were achieved through subdivision in combination with creasing.


Using OpenSubdiv in 3DsMax is very simple. OpenSubdiv is a modifier that can be applied to a model via the modify tab.


The model may lose some of its original volume but can still be changed without having to collapse the modifier (and therefore add all those extra polys). This was one of the major advancements of OpenSubdiv and is one of the reasons it is so efficient as it allows the model’s current topology to be edited while also being able to see how it will look once it is subdivided.


The bulkness of the hands is reduced and smoothed with OpenSubdiv

One of the quickest and easiest ones to gain back some of the volume is to add an additional loop to the model.


Using additional edge loops on a tree trunk

However, this may not be what is needed. The creasing of the model can be adjusted in the Edges mode of the Editable Poly modifier. This is done by selecting the edges that require more creasing (i.e. they need to be less smoothed) and adjusting the “Crease” value (Autodesk, 2014).



Uncreased Vs. Creased edges

Alternatively, a CreaseSet modifier can be used to a similar result. The benefit of the CreaseSet modifier is that it allows multiple sets of creases with different values to be managed in all one place (Autodesk, 2014).


I have been using OpenSubdiv while modelling my character and her accessories.


As demonstrated in the pictures earlier in this post, I have been using both creasing and additional edge loops. I have also found that I am using OpenSubdiv on the non-organic elements of the model.


Overall, using OpenSubdiv has allowed me to quickly add detail and give my model a smooth look.


Autodesk. (2014). 3ds Max 2015 Extension 1 – Working with OpenSubdiv. Retrieved 4th of May, 2016 from

Autodesk. (2014), To Work with OpenSubdiv.  Retrieved 4th of May, 2016 from

FX Guide. (2015). Pixar’s OpenSubdiv V2: A detailed look. Retrieved 4th of May, 2016 from

Pixar. (2016). OpenSubDiv. Retrieved 4th of May, 2016 from

The Apprentice: Texturing the Warlock’s Staff

For the game project “The Apprentice” I unwrapped and textured the Warlock’s staff using Quixel. Below is original concept image:


To keep with the style guide and this concept image, I used dark purples and also an emissive map to add the glowing “magical” cracks and a glow to the crystals. The final texture model is below:


The glowing crystals is not very evident here but that will be fixed in engine using bloom.

My Cross-Discipline Work on “The Apprentice”


For my cross-discipline project I have been working on a Studio 2 game called “The Apprentice”. Created for the ‘trust’ brief, “The Apprentice” features a dark warlock who betrays his apprentice by turning him into an unstoppable monster as it is the only way to keep the both of them safe from the angry mob of villagers. Having broken the apprentice’s trust in the beginning of the game, the apprentice, who is now a monster, turns on the warlock and kills him. The game is first person and you play as the warlock. An additional element of trust enters the game when you must sacrifice your own health and power in order to empower your monstrous apprentice.

Our team is made up of two Studio 2 games designers, two games programmers, five Studio 2 animators and myself. Being the most senior animator and the first to join the group, the games designers appointed me as ‘lead animator’. This basically gave me an art direction role but also meant that I had to delegate tasks and make sure they were being done on time. This was the first time I have been in charge of a group this size. It was a little scary as there was more responsibility but with such a large team we were able to accomplish so much more.

My Role and Work

As I was in charge of so many animators, I decided to create a technical guidelines document and a style guide to help keep everyone of the same page and to avoid technical / importing problems. In order to create these I had to do a fair bit of research and testing to work out the proper technical pipeline for exporting assets to Unity and also a fair bit of experimenting to nail down the overall style of the game.

Click the links below to view the PDF versions of these guides.



For the style guide, I created a colour palette to be used throughout the game:

colour-palette-plus-imageThe idea being that the main colours used are deep reds and purples with splashes of yellow and orange. For the most part, both the animators and game designers have taken this colour palette on board and it has been used in the design of the warlocks staff, the apprentice’s design, the menus and also the particle effects. The overall layout of the level also incorporates the colour palette as the yellow highlight has been used for lanterns with guide the player through the game.

I also mocked up a quick concept for the apprentice character:


Although, this was not the final design, it was used as a jumping off point and Jayde, a Studio 2 animator who is interested in digital painting and concept art, has greatly improved on this. The change of design partially came down to the narrative: Jayde made him younger and more impressionable so that the betrayal and power dynamic is more believable. It also came down to practicality and time: his cloak was removed to save on animation and texturing.

Additionally, I collaborated with Andrew, another Studio 2 animator who is interested in hard surface modelling, to design and create the warlock’s staff. Using Andrew’s original concept as inspiration, I created the following design:

staff-redesignAndrew took this design and quickly created a model that everyone loved:

Staff 2


This was imported into the game, as seen below:

05I am still working with the game designers to get the lighting of the staff and the glowing crystals working correctly. At the moment it is a bit dark (especially in the screenshot) but already looks pretty good when you are playing.

Finally, I was in charge of creating the magic effects. In the game, the player is able to ‘mark’ a villager using magic. This makes that villager the target and the empowered apprentice will attack them.

To fit with the dark warlock theme I decided that old alchemy symbols would look good as they have a distinctive magical vibe while still fitting with the medieval English setting (as opposed to something like Nordic runes). Similar, to the magical effect in Dota 2, we decided on two ‘magic circles’ that would spin in opposite directions underneath the ‘marked’ villager while alchemy symbols would float upwards around them. Below is the original concept I created for the magic mark:

MagicMarkV01We tested this in the game and it worked well. The only issue was the faded purple rings. I remedied this in the final version of the mark:

MagicMarkV02This has implemented in the game and looks pretty swell!

At the moment it is still a bit pink as opposed to purple (which has been fixed since this video) and the villager model is a just placeholder. However, it is extremely effective as its movement and glow makes it distinctive and eye catching.

Reflection Upon My Role as Lead Animator

That basically covers my cross-discipline work so far. Although I haven’t worked on many assets myself, I have had a hand at the overall construction, design and production of the project.

When I started this project, I expected that the role of lead animator would mostly entail art direction and would still allow me to work on many of the art assets myself. I didn’t expect for my role to have such a large management component but as our team grew it became a necessity. As a result, the majority of the role was directing the work of the other animators through collaboration, suggestions and feedback.

We used Slack as our main tool for communication and it was extremely helpful to the team. It was good as the animators were able to contact me privately for questions about the design or feedback on their work. It also helped for collaboration, as I was able to delegate tasks in mass on the dedicated animators channel.

I found that the more direction or system that was in place, the easier and more smooth it was for the other animators to work. In this way, the Style and Technical guides were extremely helpful throughout the project.

The main source of problems came from the game designers not liking certain designs or models that the Studio 2 animators had produced. It was up to me to act as a liaison and tactfully work through the design with the students. I found this quite challenging as I was worried that I would hurt someone’s feelings. However, the Studio 2 animators were all extremely flexible and professional and it was actually much easier than I initial thought.

Overall, I found that I don’t mind this sort of a role: I really like creating the overall vision for a project and directing it. I am also not as skilled as some of the Studio 2 animators so I don’t mind delegating the work to them as, honestly, they are able to do a much better job of it than I ever could. It is quite satisfying seeing a project come together around you.

I have actually had a lot of fun with this project: I got to work with great people and create something fun and magical.

Physics Based Rendering


According to Wilson (2015), “Physically based rendering (PBR) refers to the concept of using realistic shading/lighting models along with measured surface values to accurately represent real-world materials.”

traditionalvspbr01Understanding how light works in the real world is essential for creating realistic and believable textures. PBR models, and programs such as Quixel and Substance Painter, calculate the mathematics and physics of how light will interact with different materials and help apply this to the model. This means that the assets will behaviour like the real thing and should look realistic under all different lighting conditions.


PBR utilises the physics and mathematical model known as the Light Ray Model to help prescribe different attributes to different materials.

The Light Ray Model is used to predict the trajectory of light rays. The first rule to note is that a light ray that is travelling through a homogeneous transparent media (i.e. air) has the trajectory of a straight line. When light hits a surface it is either, or possibly both, reflected or refracted.

Reflected light is when a light ray bounces off the surface of the material. An example of this is polished metal. The following rules apply to reflection:

  • A “light ray that hits a surface is called the Incident Ray” and the light ray that bounces off the surface is called the Reflected Ray (McDermott, 2015).
  • The Law of Reflection states that the Angle of Incidence is equal to the Angle of Reflection

UntitledRefracted light occurs when the light ray passes through a material in the trajectory of a straight line. An example of this is clear glass.

PrintUnfortunately, materials are rarely this simple as they are rarely homogeneous. Where the Light Ray Model gets more complicated (and where using a PBR program becomes more handy) is when we look at in-homogeneous or translucent materials.

In such materials, the following factors will affect the movement of light rays and therefore its physical appearance:

  • Absorption – light rays lose energy and intensity and the colour changes
  • Scattering – light rays directions changes randomly

This is an example of how both absorption and scattering affect the movement of light through a material such as skin. This is referred to as sub-surface scattering.

  • The roughness of the material – this will affect the materials over specularity
  • The Law of Energy Conservation which states that the total amount of light re-emitted by a surface is less than the initial amount it received

Additional theories, such as Microfacet Theory and the Fresnel Effect, are included in PBR models to add extra depth to surface irregularities and realism when viewing an object from different angles.


Microfracet Theory calculates the light diffusion caused by surface irregularities on the microscopic level.

This is all extremely complicated and would take an extremely long time to figure out on your own for every different material. This is why so many artists and companies are turning to PBR.


In practice, PBR programs produce texture maps that utilise this theory to give a realistic result. They contain many different presets that are based upon real world materials and how light would interact with them.

For an example I will use the barrel that I textured in Quixel during Thursday’s class:


I decided on an old, weathered look for the barrel with some sort of toxic green liquid. For the metal bands I chose ‘Weathered Brass’. Quixel worked out the base colour, specularity and weathering using its algorithms and scanned material. I was able to tweak the ‘grunginess’ of the material – to make it dirtier and more cut up – and Quixel tweaked its specularity to match. What would have taken me hours to do by hand, just to texture those bands, literally took a minute to do. Additionally, I decided to make the ‘toxic’ green liquid glow. I had no baked in a glow / emissive map but was able to create and edit one in Quixel.


In order to fully evaluate PBR we must consider its pros and cons:


  • Very time efficient
  • Creates realistic textures quickly
  • Physically accurate so the asset will look great under different lighting conditions
  • Quixel and Substance Painter are both extremely cost efficient
  • Allows different team members to produce consistently styled textures


  • Doesn’t really work for super stylised textures
  • Programs can be buggy or heavy to run
  • Less unique than ‘handcrafted’ textures
  • May not suit the project

PBR not only saves time (so much time!) when it comes to texturing but also creates physically accurate textures that will look great under different lighting conditions. Additionally, programs such as Quixel and Substance Painter are both extremely cost efficient for both small and large teams. Overall, this makes PBR the sensible choice for creating realistic textures.

This is becoming apparent to the industry, as video game companies have started to incorporate PBR. Several high profile games have already been using it: Remember Me, Ryse: Son of Rome, Battlefield 4 and, most famously, Star Citizen. For good reason, PBR may be more than just a trend but rather the direction the industry is moving into.


McDermott, W. (2015). The Comprehensive PBR Guide by Allegorithmic – vol. 1. Retrieved from

McDermott, W. (2015). The Comprehensive PBR Guide by Allegorithmic – vol. 2. Retrieved from

Wilson, J. (2015). PBR In Practice. Retrieved from



klimt-ref-zoomThe Kiss
Gustav Klimt
Oil and gold leaf on canvas
180cm x 180cm

Painted during his Golden Period, The Kiss is one of Gustav Klimt’s most recognisable works (AS, 2016). The painting is highly decorative and depicts the abstracted form of a man and woman in a passionate embrace (TH, 2012). The couple kneel on a patch of flowers before a golden expanse. Overall, the piece has a warm and serene mood which embodies the theme of unity (PA, 2011). Through the subject of the painting, and elements of design such as shape, texture, colour and tone, Klimt carefully constructs balance in order to demonstrate the enlightening quality of a harmonious duality.

Through the use of shape and colour Klimt strikes a balance between the two figures. The man’s form is constructed almost entirely through out of tonal rectangles, while the woman’s form consists of coloured swirls and circles. This choice of shapes reflects the binary Structuralist thinking of the early 20th century (Briggs & Meyer, 2009). Rectangles and squares reflect the masculine qualities of strength and solidarity while the rounded shapes reflect more the feminine qualities of softness and flexibility (AS, 2016; Davari, S; Echostains, 2010). This contrast makes their unity interesting: they are not one in the same but rather two opposing forces that have united (Harper, 2011). Additionally, both figures have a small amount of the other’s shapes within their form.

This, again, amplifies their unity, as they are absorbing each other, and has been interpreted in many different ways: as a sexual act, as spiritual unification, as reconciliation and as a representation of ultimate love (As, 2016; Davari; PA, 2011). I believe that Klimt was demonstrating all of these and more. The unification of two contrasting elements could also be interpreted as the balance between life and death, good and evil, Mother Nature and mankind, and so on. Through this piece, Klimt explores the notion of a harmonious duality.

Klimt uses decoration and tone to create balance between the embracing couple and the golden background. The couple is depicted almost entirely through detailed decoration and, compared to the speckled background, is high in complexity. The background is much darker in tone than the couple and contains little visual information. This juxtaposition draws the viewer’s eye and helps make the embracing couple the focal point of the piece. It gives the sense that, to the embracing couple, the only thing that matters is their unity. This harmony has set them apart from their surroundings: the rest of world has fallen away and is immaterial and inconsequential (PA, 2011). This helps to establish the idea that ultimate enlightenment comes from uniting a duality.

Through the use of decoration, colour, tone and shape, Klimt creates balance within his work, The Kiss, in order to demonstrate that insight and clarity is achieved through the reconciliation of contrasting forces. While the message is subtle, it was poignant at the time and is still relevant today.


Well, yes.

Just as Klimt used different forms to convey different emotions and meaning, so too does graphic design, advertisement and character design. This has become known as shape language – the idea that different geometric shapes have different emotional properties (Anhut, 2016).

The idea is to break down the concept or character to its simplest form (e.g. strong, stubborn) and use rough geometric shapes to represent this (e.g. a square). Nowhere is this better seen than in animated children’s movies. The image below demonstrates the shape language used for the characters in Up.

upshapesWhen we look at the final version, we can see how squares are used throughout Carl’s design – in his face shape, glasses, eyebrows and chair – and how this matches to his stubborn disposition. This contrasts the Ellie’s more rounded form – as reflected in her face, hair and chair.

Disney-Guide-to-Getting-Over-a-Breakup-UPJust as in The Kiss, the contrasting shapes of these two characters helps emphasise the notion of a harmonious duality.

Similarly, the character designs for the emotions from Inside Out began as a series of simple shapes that reflected the personality of each character (TFD, 2015). Again, we can see how these suit the final characters:


Butler (2013) expands on this and demonstrates that constructing a character from a shape not only emphasise their characteristics but also gives cohesion and consistency to the overall design. Additionally, the contrast between different characters helps the audience to distinguish them.

tumblr_inline_mwqvzesYdW1qdm9r3Additionally, Klimt’s use of decoration has inspired modern artists such as Jen Zee: the art director of Supergiant Games who incorporated some of Klimt’s techniques into the game Transistor. Zee (2013) states that Klimt’s use of “interesting shapes” and his “flare for dramatic presentation” seemed to work well with the cyberpunk aesthetic.

1407982520654_Transistor_19-mar-2013_04The use of decoration and Art Nouveau shapes is present in both the actual game and the promotional art and helps create the game’s atmosphere.


Anhut, A. (2016). Art Style Atoms. Retrieved from

Art Story (AS). (2016). Gustav Klimt. Retrieved from

Briggs, R. & Meyer, J. (2009). Structuralism. Retrieved from

Butler, T. (2013). Notes on Character Design. Retrieved from

Davari, S. (NA). Analysis of Klimt’ artworks. Retrieved from

Echostains. (2010). Behind the Paint: ‘The Kiss’ by Gustav Klimt. Retrieved from

Harper, S. (2011). The Universe in a Kiss. The Montréal Review. Retrieved from

Painting Analysis (PA). (2011). Klimt, ‘The Kiss’, Painting Analysis. Retrieved from

Pelissier, S. (2014). Portrait of an Artist: Glustav Klimt’s Controversial Pattern in Painting. Retrieved from

Tips From Disney (TFD). (2015). D23 Expo 2015 Gets Turned “Inside Out”. Retrieved from

Totally History (TH). (2012). The Kiss. Retrieved from

Zee, J. (2013). Unicorn Sparkle: An Interview with Jen Zee of Supergiant Games. Retrieved from

World Builders Project: A Post Mortem


The World Builders Project was a seven week project in which we were to design, develop and create an environment based on a description from a novel for a ‘teaser trailer’. I decided to base my environment on a scene from Deep Secret, by Dianna Wynne Jones, in which the characters travel along a garden path strung high in the air between three impossibly tall towers. I decided to create the scene in 3D in the Unreal Engine. Below is the final video:

From here, I will look into what went right and what went wrong with this project and ways that I can improve my processes for future projects.


In my eyes, there are many more failures than successes however I did learn a lot about documentation and using the Unreal Engine. Additionally, I believe that parts of my final piece reflected both the intended narrative and Art Bible.


The most successful part of this project was the documentation I created during pre-production. Not only did I learn how to create an Art Bible and a Project Plan but also researched pipelines, workflow, techniques and styles that really helped to inform my work. This was extremely useful during production as it gave me a clear guide and structure to what I was creating. In particular, the dissection of my references in the Art Bible really helped me to define the look I was going for.

Final Piece

Although the final piece is extremely lacking, literally in terms of assets but also in detail and style, I believe that I successful captured some of the intended narrative by following through with what was outlined in my Art Bible. In particular, I managed to follow through with my colour scheme and atmospheric effects which did help to create a dreamy atmosphere and the feeling of being high up in the sky, both of which were in the novel excerpt. This is evident below:


The first image is a test render before the final lighting and colour adjustments were made while in the second these have been adjusted to suit the mood. This was done in reference to the Art Bible.

Unreal Engine

Through the production half of this project I learnt a lot more about the Unreal Engine. In particular I learnt about:

  • Re-importing objects
  • Lighting
    • Editing the direction and colour of lights
    • Editing the sky box and what effect it has on lighting
    • Using lighting to give the illusion of a particular time of day
  • Matinee
    • Setting up cameras
    • Animating camera movement
    • Rendering through the Matinee system
  • Materials
    • Manipulating complex materials
    • Creating blendable materials
    • Using vertex painting to blend two materials

This knowledge will be extremely helpful for future projects.

Technical Framework

Another successful part of the project was the technical framework which I used. I outlined the rules and guidelines for this technical framework in my Art Bible. The key guidelines are as follows:

  • When opening a new Max file, change the Unit Setup to Metric Centimeters.
  • Texture files are to be in PNG format, 1024px by 1024px.
  • Files are to be saved in the correct folder and named according to the naming convention.

Following Brett’s advice about file structure, I created my own which allowed me to easily to update assets and re-import them into the Unreal scene. It is pictured below:


Each asset had its own folder, inside which were a ‘Master’ FBX file and also a Maps and a Mesh folder. These sub-folders corresponded to the 3D and texture files. Inside each of these folders were a Superseded folder, in which old ‘Master’ FBX files were kept, and a Working folder in which all the working MAX files were kept. Once an asset was updated, the old ‘Master’ file was moved to the Superseded folder and the updated model was saved in its place.

These ‘Master’ files were the ones that I used within my Unreal scene. After updating an asset, this allowed me to simply right click on an asset and select re-import. This would pull the updated asset into the scene without messing up the positioning, scale or rotation. I had not used this system before and found it much, much more efficient.


This project failed in several major ways particularly in the production phase, resulting in a final piece that was not up to standard.


For me, the biggest issue with this project was time: things came up, I worked longer hours, other classes required extra work and, most importantly, I am horrible with time management. This is largely due to my overambitious scheduling: overestimating my available time and underestimating how long things would take me. I was too relaxed during the creation of the documentation, meaning that it took longer than it should thus leaving me with less time for production.

Incomplete Final Piece

My scheduling issues resulted in me being unable to finish of all the assets. As shown in my work breakdown structure below, the assets marked in red were incomplete or unfinished:

WBSMore critically, the final product is not what I envisioned: it is far blander and empty. I wanted it to feel overgrown and heavy with plants, vines and trees. In particular, I really wanted to experiment with modelling some surreal trees but, again, I ran out of time to do so.

To me, the biggest failure was the path. I had wanted it to look as it had originally in my concept:

new conceptUnfortunately, the final path is extremely plain compared to this one. It lacks depth, the root system and surface details, like trees and rocks. Instead it is boring, ugly and out of place. Overall, I am quite disappointed with my work on this project and will try to push myself for the next project.


This project was the longest solo project I have worked on for university. It was quite an interesting experience and I have learnt a great deal that I hope to carry over into future projects. I have also discovered several issues that I hope to improve upon.

For future projects I will definitely be doing the following:

  • Conducting research into the workflow, pipelines, style and design
  • Creating comprehensive documentation
  • Using an effective file structure and technical framework

However, to ensure that future projects do not fail as badly as this one, I will need to improve my scheduling. I can do this by:

  • Creating the documentation more efficiently (this should be quicker the second time around)
  • Accounting for potentially unavailable time: I can do this by leaving in some buffer time
  • Overestimating how long things will take (Chris suggested doubling my estimate)

With these aspects in mind, I will be aiming to improve my production during the next project.


The Word Builders Project was extremely interesting and fun to work on. To me, the most enjoyable part of it was actually researching and deciding on the look and feel of the piece when I was putting together the Art Bible. I probably spent too long on it as I didn’t want to put it down but rather wanted to keep tinkering. This may have contributed to me falling behind in production but, if anything, the comprehensive documentation actually helped me a lot. Unfortunately, unforeseen circumstances meant I had much less time during the production than I expected resulting in a lackluster final piece. I am quite disappointed with myself, as there is not much I would want to use from this project in my showreel. However, with the work I have put into it, I will definitely be coming back to it with the goal to create what I originally envisioned.

Modular Workflows

Last week in class, Brett took us through a modular workflow exercise. This involved us using a set of small modeled assets, such as a part of a wall, that could fit together in a multitude of different ways. We were to use them to construct an entire building.

Below is an image of the modular pieces that we used:

02By duplicating, rotating, mirroring and occasionally scaling the original pieces I was able to construct the following building:

renderAs you can see, this building has been made of many, many smaller pieces:

04To ensure that the pieces fit together correctly, we used vertex snapping and axis constraints. I had never used this before and it was extremely helpful.

05The only issue we encountered was that the assets had not been modeled to the grid, which made it harder to fit them together in some circumstances. For my own projects, I will definitely be constructing the models to the grid.

From this exercise I could already see the benefit of using modular pieces in order to construct larger objects and environments. I will be using this knowledge for both the world builders project and for the cross-discipline project, Orient.


From the lesson on modular workflows I could see how it could be applied in my own projects.

In the world builders project, the most obviously application would be for the pyramid’s steps. Below is a mock-up of a modular step piece which I construct to the grid:

06Using this I was quickly able to construct part of a staircase:

07Obviously, this is unfinished as it would need sides or a railing. However, as each stair piece is an instance of the original, this was a very simple and quick fix.


09Super simple and quick. Again, this will need some adjustment once I determine the steepness of the pyramid.

For the Orient project I will be creating many different low-poly buildings so using modular pieces will speed up this workflow dramatically.

To begin with, I decided to start with some modular pieces for a square building (10m x 10m for the base and 5m for each floor). These were modeled to the grid.

11The windows and door are not necessarily to scale as the game designers wanted the buildings simplified and stylised.

From this, I can make different buildings with a varying number of floors.
12Of course, these are very basic and very repetitive: they lack variation.

13However, it is a good place to start and I can always adjust things later. Similarly, as the walls were made separately, I can also construct rectangular buildings from these pieces.

Deep Secret: Improved Concept

Taking the feedback from the gallery walk on board, I created a new concept that was a paint over of a 3D mock-up which helped me enormously with the perspective.

new concept

This is still pretty rough as my final scene should be much more overgrown, particularly where the towers connect to the path.

Deep Secret: Development Work for the Path

One of the biggest piece of feedback I got from my initial concepts was that the path was unclear in how it was structured and seemed completely disconnected to the rest of the scene. With this in mind, I created a couple of concepts which explored different styles for the path that would make it seem more plausible:


The initial concept was just a exploration of my original idea of the path: a grassy garden path that literally just hangs in the air. This concept is the least interesting and also the least plausible.


The second concept is based on suggestions that the path could be connected to rocky outcrops or floating rocks. I really like the image of a garden path atop floating rocks.


My third concept was based on a suggestion that the path could be made from tangled roots. I really like this idea and tried to explore how the trees might hold the path together. I feel that this concept works the best as it has a very surreal tone to it and the trees can be used to add variety to the scene and to break up the path a bit.

These concepts were also displayed at the gallery walk. While the feedback was mostly positive towards the rocky and root path concepts I was also advised that the top of the path was too plain and boring – that I could break it up with rocks and foliage.

Deep Secret: Initial Concepts

For the Deep Secret world builders project I created some initial concepts. Below is my first attempt at a ‘photo-bash’ concept for this project.


While this does kind of show the positioning of the props within the scene it is also very confusing. The photos I chose were from differing (and incorrect) perspectives which makes this concept very unclear.

To remedy this, I quickly coloured one of my thumbnails in the hope of making something more usable.


Again this does not display the correct perspective and gives no information about the details of path or the towers. So again, this was an entirely unusable concept.

I presented both of these concepts at the gallery walk and got a good amount of feedback on how to improve: in particular, I was told to focus on the perspective and give more detail to the path. While it was nerve-wracking to display concepts which I knew were ineffective, it was still a good experience and I have learnt a lot from it.

Colour Studies

A couple of weeks ago, we did a series of colour studies in class. Mine are included below:


This was a really hard exercise for me. However, it was quite good as it forced me to consider how light and colour work in different environments and in different types of lighting.


I think my second attempt was more effective, as I had begun to get a sense of how to construct an environment.

Deep Secret: 3D Mockup

To help block out my idea I mocked up some 3D models of the three towers and positioned them in varying positions to create some 3D thumbnails.


Deep Secret: Initial thumbnails

For the world builder project I am creating an environment for the novel Deep Secret by Diana Wynne Jones. My plan is to create this environment in 3D in the Unreal engine. The environment is in the land of Babylon – a mystical and hidden place – where a swaying garden path climbs upwards among st three tall towers. The hanging garden is high up in the air and the towers are each very distinct: one like a rook, one like a pagoda and one like a pyramid.

Below are my initial (and very rough) thumbnails:



Second Tone Study

Below is the second tone study from this image. Again the study used 7 tones, one layer and took 30min.


This one was much rougher than my last study but I felt that I was quicker and had a better understanding of what to do this time round. I notice from both my studies that the absolute tones (black and white) are used in very small amounts and usually for focal points.

Tonal Study

Below is a tonal study from this image. Study took 45 mins, used 7 different tones and was painted on one layer.

Erika_TonalStudyThis is very rough but it was a really good exercise. Upon reflection, I should have put more detail into the sky and some more highlights on the rocks.

Digital painting photo study

As I have been trying to improve my digital painting, I have been doing a couple of photo studies. Below is a quick colour study of a pot plant:

Practice_PlantsOverall, I am happy with this but I should have put more detail on the leaves as they look very flat. Also, the pot does not look grounded and I couldn’t work out how to fix it. I tried adding a shadow but it did not work very well.

Landscape Colour Study

Below is the colour study I did today in class. I had a couple of computer issues but I am still pleased with how it turned it.

ColourStudy_ErikaThis was a really fun exercise and I definitely be doing this more often.

More Painting Practice

More painting practice, this time a plank of wood that I created with from this tutorial.


Holiday Digital Painting Practice

This coming trimester I would really like to get better a digital painting and actually create a finished piece that uses a “painterly” style. So I have been trying to follow some tutorials and improve my technique. Here is a strawberry I painted while following a tutorial.


The leaves look too shiny so I will try to improve on that.