Creation Of A Humanoid Character

Published Date: 02 Nov 2017

An investigation into the creation of a humanoid character.

Focusing on effective design, efficiency in the games engine, and maintaining the level of detail displayed on characters throughout cinematic trailers.

Thomas Watson

Games Concept Design BSc

Supervised by

Tom Elgon

Contents

Creating an Effective Character

Introduction

NEED THIS

Gender

It is generally suggested that the gender of a humanoid character can be quickly defined using basic shape alone. In her book Women’s Sports Medicine and Rehabilitation, Swedan (Swedan, 2001, p. 3) outlines a number of "Anatomical Gender Differences" (p.3) between both the Male and Female genders, identifying Women to be at an overall physical shortcoming to men. These factors also play a large role in the external appearance of both genders; Su and Zhao (Su & Zhao, 2011, pp. 60-63) describe the physique of a woman to be far more elegant than that of a man, explaining that typically a male character should be represented using "geometric shapes" (Su & Zhao, 2011, p. 60) and "strong lines" (Su & Zhao, 2011, p. 60), allowing for the masculinity and perceived heroism of the character to be portrayed, whereas a female character would be depicted using far more gentle shapes and flowing lines, emphasising her superficial beauty. This is, however, not a solid rule; it can be beneficial to use a combination of rigid and curving lines on a male character (Cantor & Valencia, 2004, p. 79), "[enhancing] the characters’ vividness, liveliness and aggressiveness" (Su & Zhao, 2011, p. 63) and "[highlighting the] masculine toughness" (Su & Zhao, 2011, p. 63). NEED MORE REFERENCES HERE

JOHN CLEESE’S THE HUMAN FACE

Along with the overall shape of the human figure, it is widely agreed that the proportion of certain elements of the human form can help recognise the gender of the character. In his book Figure Drawing For All It’s Worth, Loomis (Loomis, 1943) explains that the "ideal" (p.26, p.27) proportion of both a male and a female to be eight heads tall; the difference in width of the figure, however, is what’s most gender defining. Su and Zhao illustrate this point (Su & Zhao, 2011), showing that "from the front, both men’s and women’s physical structure can be presented by using two triangles of different sizes. As for women, the smaller one is on the top, while for men, the bigger one is on the bottom" (p.46). IMAGE HERE

Hard Surface Modelling Techniques 2

Even if the character is a soldier, add gender weighted elements, ie a female may have high heels

Alive Character Design

More to say about Male Characters

OBVIOUS GENDER INDICATORS - CLOTHING HAIR ETC

Symmetry

It has been shown that the physical features, especially the use of symmetry, displayed on the character can strongly influence how the character is perceived, whether it be in an attractive or unattractive light. In her book Better Game Characters by Design, Isbister (Isbister, 2006) describes that generally all heroic characters will wield attractive features, such as strong level of symmetry along with "straight profiles" (Isbister, 2006, p. 7), "encouraging the player to see them as smarter, stronger, kinder and more socially skilled" (Isbister, 2006, p. 9); both Tsai (Tsai, 2007)Cantor and Valencia (Cantor & Valencia, 2004) emphasise this point, explaining how it is typical to include symmetry within a design meant to be ‘attractive’. Classically, villainous characters are designed to be asymmetrical, with an "unhealthy complexion" (Isbister, 2006, p. 10); however Kotaki (Pogoda, et al., 2011), Cantor and Valencia (Cantor & Valencia, 2004) all agree how it can also be extremely beneficial to introduce asymmetry within heroic characters, giving them a further degree of depth. Despite this, Tsai (Tsai, 2007), explains that if asymmetry is introduced on the design it should made to be very apparent, as "subtle asymmetry can look like a mistake" (Tsai, 2007, p. 83).

101 Quick and Easy Secrets - Shaping Up with Symmetry

Motive

Describing the motive of the character within the design can be achieved through their clothing, composition, body language and even the way the shadows are arranged in the concept (Pogoda, et al., 2011).

Kotaki (Pogoda, et al., 2011) explains that to create the aura of arrogance and greed surrounding his villainous character he donned him with a costume covered in precious metals, signified by the strong gold colour. His organic flesh has also been replaced with both magical and mechanical parts, suggesting that he wants to be immortal, further strengthening the arrogant and conceited nature he portrays (Cantor & Valencia, 2004). Kotaki (Pogoda, et al., 2011, p. 124) also goes on to explain how he decided to keep the palette of colour as minimal as possible, keeping him "pure and lacking in personality other than his arrogance and need for power". Despite typically being evil, Cantor and Valencia (Cantor & Valencia, 2004) explain how a lot of villains aren’t entirely so, and it therefore can be beneficial to give them "empathetic qualities" (Cantor & Valencia, 2004, p. 65) that will appeal to the audience. Seegmiller (Seegmiller, 2004) suggests that to ensure the villain you are creating is effective, you should place it in a completely different context, and if it is still noticeably villainous, the design is functioning effectively (Seegmiller, 2004, p. 11).

Kotaki (Pogoda, et al., 2011, p. 110) describes that he directed his heroic character away from being a "pretty" design, explaining that if he had done this, the character wouldn’t have looked "rough and ready", a characteristic he still wished to portray; Cantor and Valencia (Cantor & Valencia, 2004, p. 63) support this statement, explaining how it is rare for heroic characters to be completely flawless, and is therefore effective to give them a few imperfections, making them far more authentic. To help direct the concept, Kotaki (Pogoda, et al., 2011, p. 110) illustrates that, like the villainous design, he used the colour scheme of the character; by dressing him in colours usually associated with royalty, the viewer will immediately assume this is his role.

To help "ground [a] character in reality" (Gahan, 2009, p.163) and give it some personality, Gahan (Gahan, 2009), Spencer (Spencer, 2012), Cantor and Valencia (Cantor & Valencia, 2004, p. 66) explain how it can be extremely useful to ask some questions surrounding the characters characteristics and motives. These questions could also help inspire visual elements of the character, and vice versa, assisting in the overall development (Le et al., 2005, p.26).

Anna Pogoda shows how a characters motive can be easily changed by editing very subtle elements

Detail

It is widely agreed that the use of detail on a character can both make and break the design. Cantor, Valencia (Cantor & Valencia, 2004), Valve (Valve, 2012) and Tsai (Tsai, 2007) all employ the "less is more" (Cantor & Valencia, 2004, p. 74) approach when detailing their designs, explaining that large areas of heavy detail can ruin the concept, making it very overwhelming to the eye. To refrain confusing the design, Valve (Valve, 2012) suggest combining areas of detail with plainer sections, helping achieve an appealing equilibrium; making the retained detail much more of a focal point (Tsai, 2007).

Cantor and Valencia (Cantor & Valencia, 2004) both agree that applying detail to a concept without any justification for doing so can be dangerous, and in some instances can destroy the overall look of the character. Tsai (Tsai, 2007) suggests taking "visual cues" (Tsai, 2007) into consideration when designing the character; extremely useful when determining how the character is viewed. Both Isbister (Isbister, 2006) and Tsai (Tsai, 2007) describe that use of visual cues will allow for the viewer to effectively place the character, helping direct their initial perception. For main characters, however, Isbister (Isbister, 2006) suggests slightly veering away from the stereotype; helping to differentiate them from their team and therefore making them far more visually interesting, giving the player a stronger experience (Isbister, 2006, p. 16).

Familiarities

Along with visual cues, it has also been shown that familiarities play a considerable part in the appeal of the character; Le (Le, et al., 2005, p. 32) Cantor and Valencia (Cantor & Valencia, 2004) claims that for a fantastical character to relate with the viewer, they must have a selection of noticeably familiar human characteristics. Spencer (Spencer, 2012) reinforces this point in his book ZBrush Creature Design, explaining that by using "real world sources to ground the design in reality" "helps make the creature much more believable to the viewer" (Spencer, 2012, p. 3).This is strongly portrayed in games such as Gears of War (Gears of War, 2006); no matter how alien the Locust are, they still have very strong human features. IMAGE HERE

Despite explaining the importance of using existing themes to influence the character, Spencer (Spencer, 2012) describes how for the design to be effective, changes to the positioning of certain elements can be altered, "breaking expectation" (Spencer, 2012, p. 4) and ultimately gaining the interest of the viewer. Cantor and Valencia (Cantor & Valencia, 2004) backup this statement suggesting that a character does not necessarily need to be completely relatable; the only requirement is for the viewer to be interested (Cantor & Valencia, 2004, p. 57).

In his book ZBrush Creature Design, Spencer (Spencer, 2012) explains the importance of being able to distinguish your design between a character and a creature. A character will be relatable by the viewer, displaying a number of human traits, whereas a creature "only needs to appear to be a viable life form that is suited to the task it is portrayed to undertake" (Spencer, 2012, p. 4). Cantor, Valencia (Cantor & Valencia, 2004), and Spencer (Spencer, 2012) continues to describe how for a character to be effective, it must be able to create a sense of empathy between it and the viewer, usually communicated through the eyes (Spencer, 2012, pp. 4-5). Seegmiller (Seegmiller, 2004) emphasises this statement, describing that for a character to seem appealing to its audience, it must have expression; the most of which being derived from the face, namely the "eyes, nose and mouth" (Seegmiller, 2004, p. 30).

The Importance of Anatomy

It is widely agreed that to successfully create a believable humanoid character, a strong knowledge of the Human anatomy is required (Seegmiller, 2004) (Gahan, 2009) (Spencer, 2012). Both Gahan (2009) and Spencer (2010) state that even on an alien character correct usage of anatomy is vital, as similarities between that and a human allow the viewer to relate to the design and ultimately find it more visually appealing.

It can be beneficial to employ certain techniques, allowing for a quick understanding of the basic forms, which will go on to provide the building blocks for the finer details. Loomis (1943), Goldfinger (1991), Spencer (2010) and Kingslien (Kingslien, 2011) all display similar methods, utilising planes to build up the structure of the human form; describing how it is effective at splitting the figure up into its fundamental forms, ultimately allowing the creation of a much more realistic character.

Simpson (2010), however, criticises this method, explaining that dividing up the body into simple geometric forms can be very unrealistic and it is far more effective to use what he calls "The Process" (Simpson, 2010, p. 11) claiming that using this technique allows the artist to create a human figure without reference, therefore suggesting it would be extremely applicable when designing a fantastical character. The route followed by this method begins by creating the skeleton and then adding the muscle, fat and skin respectively. Although, despite having the potential to be extremely effective when creating a character without available reference, this method will require a far more developed knowledge of the human anatomy.

Along with having a solid knowledge of the human anatomy, Goldfinger (Goldfinger, 1991), Gahan (Gahan, 2009), Hylton (Character Modeling for Production. Vol 2, n.d.) and Kingslien (Kingslien, 2011) all stress the importance of having strong anatomical reference; helping the artist accurately recreate the human form within their work (Goldfinger, 1991, p.xi), along with allowing anatomical masses to be visualised underneath clothes and armour (Gahan, 2009, p.187) and helping pick out subtle details that may not be initially obvious. (Fedor et al., 2007, p.22).

In his book ‘Practical Anatomy for Artists’, Simpson (2010) explains how by employing proportional theories, along with some personal judgement, the quality of production can be greatly influenced, giving it a far more "professional look" (Simpson, 2010, p. 19). However, he also warns that relying too much on one system can cause problems; not following it close enough, causing the image to become disproportionate, or trying to be too precise and potentially taking a large amount of time (Simpson, 2010, p. 19).

Loomis (Loomis, 1943) and Kingslien (Kingslien, 2011) both suggest splitting the height and width of the figure into a number of sections, the size of which being dictated by the characters head, with each section landing on an anatomical landmark, such as the nipples and navel. IMAGE HERE Simpson (Simpson, 2010) contests this method however, suggesting that there is far too much variety in the positioning of these elements throughout various people, and instead the artist should use the lengths and widths found within the skeletal structure to help dictate the proportions of the character. Kingslien (Kingslien, 2011, p. 50) agrees with this, explaining that these skeletal landmarks give something for the artists to visually relate too, helping them "ensure the [anatomical] shapes maintain their proper relation to each other". IMAGE HERE

Cantor and Valencia (Cantor & Valencia, 2004, p. 78) argue these statements however, explaining that these rules should not be depended on, as in some cases they can be very misleading; "if it looks right, it is right" (Cantor & Valencia, 2004, p. 78); Tsai (Tsai, 2007) reinforces this point, explaining how in some cases it can be effective to even exaggerate the proportion of the character, helping "communicate the character concept" (Tsai, 2007, p. 81).

Along with skeletal landmarks, Loomis (Loomis, 1943, pp. 58-65) and Spencer (Spencer, 2010, p. 91) explain that there are a number of dominant superficial muscular landmarks. The shape and positioning of which has a huge impact on the shape of the body’s surface, adding depth to the planar form and ultimately influencing the realism of the character.

Creating the Concept

Introduction

NEED THIS

Concept Workflow

Inspiration

NEED THIS

Thumbnailing

There are several methods of beginning the character design process, each with their own variations; however the most prevalent seems to be using silhouettes and value. Zhu (Episode 30 Character Silhouettes part1, 2010) explains that the silhouette of a character is what will provide the viewer’s first impression, and therefore its potential to quickly describe the character is crucial. In his article, Tsai (Tsai, 2007) claims how the silhouette of a character is undoubtedly the most important aspect of a character, explaining that it is the only element of the design which isn’t affected by external sources; constantly allowing for the viewer to evaluate it. Seegmiller (Seegmiller, 2004) emphasises these statements, describing how he believes that an effective silhouette makes the character far more aesthetically appealing, along with helping them clearly portray their motive; consequently strengthening the viewer’s opinion.

Due to the effectiveness of a characters silhouette being crucial, the process taken to create which must be thorough, allowing for in depth design exploration, ultimately resulting in a successful outcome. In the ‘Skillful Huntsman’, both Yamada and Robertson (Le, et al., 2005) illustrate how it is initially beneficial to create a large selection of primitive silhouettes, eliminating any detrimental preconceived ideas, along with allowing a variety of possibilities to be scrutinized, thus concluding with a refined selection of designs which can be taken further into the refinement stages. Tsai (Tsai, 2007) and Archer (Archer, 2013) support this claim, advising to refrain from applying restrictions throughout the initial design process, allowing for a large amount of silhouettes to be rapidly prototyped; Archer (Archer, 2013) also suggests the workflow adopted should be very bold and loose, potentially causing unexpected brush strokes thus influencing elements of the concept. Seegmiller (Seegmiller, 2004) emphasises that the silhouettes created initially should be solely concepts, not be made to look attractive or appealing; it is in the later stages where the aesthetics are a focus, "You are not drawing pretty pictures, but generating different ideas" (Seegmiller, 2004, p. 25). Even so, Zhu (Episode 30 Character Silhouettes part1, 2010) has a slightly different approach to this process; rather than collating a large bank of possible designs, he begins by creating only a few silhouettes, individually refining each of them up to a presentable standard. It could be argued that because he is an industry professional he has the experience to confidently skip this initial stage, however to make sure the final outcome of this project is as effective as possible, it will likely be desirable to make sure it isn’t looked over.

With pure white being the highest value and black being the lowest, contrasts in value can be used to draw the eye around the image. Once the silhouetting process has been completed, value needs to be added to the design, illustrating the concepts internal forms. Both Valve (Valve, 2012) and Keyes (Color Theory: The Mechanics of Color) describe that strong value contrasts should be made in areas of high interest, as the eye is "naturally drawn towards these areas" (Color Theory: The Mechanics of Color, n.d.). Shown by Zhu (Episode 30 Character Silhouettes part1, 2010) and Valve (Valve, 2012), value should be gradated throughout the design, with the lowest value at the bottom of the character and the highest at the top towards the head; again, directing the eye through the design.

Described by him as "chaotic and sketchy" (Pogoda, et al., 2011, p. 170), Ordonez’s (Pogoda, et al., 2011) method is far from the methods previously described, however still allows for a lot of design variation, and refrains from focussing on one particular route. Ordonez (Pogoda, et al., 2011) begins his design by dynamically laying down a large number of lines onto the canvas. From this, he continues to further his design by refining the work to developed line art; picking his desired path within the sketch. CHECK TO MAKE SURE QUOTES HAVENT BEEN COPIED and REFERENCE

Concepting within ZBrush

The benefit of creating the character concept in ZBrush is that it completely negates the reason for having a ‘traditional’ 2D design (Maia, 2012), therefore allowing for the high-polygon character sculpt and diffuse texture to be created alongside the character concept; considerably reducing the time required for completion of a unique game ready character.

As previously discussed, Archer (Archer, 2013) begins his concepting with the silhouetting technique; unlike the other examples, however, he then takes the design directly into ZBrush. Utilising tools such as ZSketch, Unified Skin and Dynamesh along with the Clay Buildup, Inflat, Dam Standard and Move brushes, Archer (Archer, 2013) swiftly creates a 3D representation of the 2D concept; this can then be continually refined within the software. This method, however, is most applicable when working with the anatomical base of a character, as the ZSketch feature quickly allow for muscular volume to be realised. When working with an armoured character, however, Archer (Archer, 2013) uses a duplicate of the anatomical base which he then pulls out, using a combination of the Clay, Damien Standard and Trim Dynamic to create rough forms on the exterior of the character. To refine the design, Archer (Archer, 2013) explained how he worked into the forms using the Polish brushes; creating a crisp final outcome. Jensen (Jensen, 2012) displays an almost identical method to Archer’s, however instead of solely using the ZBrush’s Polish brushes to refine to concept, he creates brand new geometry on top of the existing form by using the Topology and Crease tools; producing a far cleaner, and consequently more appealing, final outcome.

When concepting in ZBrush, Seaut (Seaut, 2012) explains that it is desirable to have a few preconceived ideas of how the character is going to look, usually achieved by creating a few thumbnail sketches. Once this been decided, he then proceeds to create a simple base mesh and uses the Clay, Inflat, and Smooth brushes to create the characters main forms, defining them with the Dam Standard brush. When creating external elements, such as armour and clothing, for their characters, Seaut (Seaut, 2012) and Silva (Silva, 2012) utilise very similar methods; using both Adaptive Skin and the CurveSurface brush respectively, external geometry is able to be created on the surface of the character which can then be sculpted into to form the desired element.

Within Spencer’s book ZBrush Creature Design, guest artist Bryan Wynia (Wynia, 2012) details his approach to character design. Initially creating a series of speed sculpts; he then proceeds to render the meshes into Adobe Photoshop, using the Liquify and Brush tools to experiment with the forms of the character. Along with using this technique to initialise the design, utilising Photoshop’s tools throughout the entire process could be extremely beneficial; allowing for forms to be realised that may not necessarily be initially obvious.

Throughout the later stages of the design process, it is possible that the overall silhouette of the character can get gradually overlooked; to cease this from happing, Maia (Maia, 2012) suggests changing the material of the mesh to a flat colour, allowing for the silhouette alone to be displayed, giving the potential for it to be altered if necessary.

ZBrush Experts

Page 144

Page 168 ZSketch prototyping

WRITE ABOUT ANY REALLY REQUIRED CONCEPT TOOLS and how they benefit the concept, DYNAMESH ETC

ZBrush Professional tips

215 Character concept zBrush

Ryan Kingslien

Panel Loops Concept

Detailing within ZBrush

Once the concept is complete, it can be beneficial too....

Colourizing the Concept

Colour Theory

Spencer (Spencer, 2012), Cantor, Valencia (Cantor & Valencia, 2004), Keyes (Color Theory: The Mechanics of Color, n.d.) and Valve (Valve, 2012) all suggest having knowledge of colour theory when creating the schemes for a concept, as it colour plays a large role in how the character is read by the audience and therefore understanding the theories behind it allows for successful application.

The colour wheel plays a huge part when determining the colours to be used in design, and therefore having an understanding of which will potentially greatly benefit the final concept. (Spencer, 2008, p. 156). Using its three main sections, Primary, Secondary, and Tertiary; a variety of algorithms can be employed by the wheel, helping suggest various complementary schemes.

Analogous

Complementary

Compound

Monochromatic

Triadic

Saturation is the strength of the hue in a colour; the higher the saturation the greater the intensity of the colour, and vice versa (Spencer, 2011). Both Keyes (Color Theory: The Mechanics of Color, n.d.) and Valve (Valve, 2012) explain how the use of saturation within a concept can help define focal points, drawing the eye around the image. However Spencer (Spencer, 2012) and Valve (Valve, 2012) both describe that the using a large amount of intense saturation can be very overwhelming to the eye, and therefore its use should be sparse.

Application

The application of colour to the concept can play a huge part in appeal of the final piece; research suggests that an effective character should be able to be recognised by its colour scheme alone. In 2010, Browning (Browning, 2010) recreated a selection of the characters from the popular video game ‘Street Fighter 2’, using only block shapes and colours; IMAGE HERE this artwork proves the initial point, showing that the characters are still completely identifiable, without the use of their physical forms. This finding is reinforced Monahan (Monahan, 2009), who conducted a similar investigation; displaying that no matter how simple the physical forms of the Left4Dead Characters are, the colours make them completely recognisable. Along with using these techniques to determine the effectiveness of the final design, this method could also be utilised when experimenting with the application of colour to the character, allowing for quick prototyping of different schemes and their success.

When creating the colour scheme for a character, Su and Zhao (Su & Zhao, 2011) suggest initially picking the character’s most prominent colour, and then choose the secondary colours based upon that choice. In their DOTA 2 character design guide, Valve (Valve, 2012) explain that initially the focus should be on creating a colour scheme directly adhering to the rules above, however suggest that once this is done, it can be beneficial to slightly experiment with the colours making sure that they effectively fit the concept. Valve (Valve, 2012) also advise to "avoid using colours which are dominant in the [game] world", adding an element of contrast which will allow for the character to stand out from its environment.

Despite the character being from a fantastical setting, the colour scheme does not have to be also. Spencer (Spencer, 2012) explains how there are a wide variety of appealing schemes in the real world, and taking from these is an efficient method of quickly creating an appealing palette that can be applied to a character; for example, "Aquatic animals have vibrant and interesting skin colorations that lend themselves to use in alien life" (Spencer, 2012, p. 18).

Workflow

There are a number of different methods of applying colour to a mesh in within ZBrush; hand painting, using the Polypaint feature, or from photographic reference utilising either ZApplink or Spotlight. Features in ZBrush to transfer texture to diffuse

Polypainting

Research suggests that the predominant function of ZBrush’s polypainting feature is to realistically recreate organic objects, such as human skin. The use of the brushes, in conjunction with Alpha’s and the layer palette, allow for ‘colour zones’ and ‘mottling’ to be built up, giving the skin depth and a natural "sense of translucency". IMAGES HERE – COLOUR ZONES Spencer also illustrates the power of using a base layer, especially for fantastical characters; changing its colour allows for the entire subject of the paint to change. IMAGE HERE – GREEN/BEIGE

As well as the previous, Polypainting can be used to help greater define both recesses and raises in the mesh; using Cavity Masking to simulate ‘dry brushing’, painting dark, cooler colours into the recesses can give them depth, along with painting lighter colours onto the raised areas, highlighting them.

Despite this, in his Masterclass, Archer displayed how it is in fact possible to successfully texture hard surface objects with Polypaint; using alphas in combination with a base layer to achieve a weathered feel, along with other fine details.

Photo-texturing

Photographic texturing is a far more artificial technique, using the visual information gained from images to texture the model, rather than hand-painting it. The benefit of this method is that it enables very high frequency detail to be applied that may be difficult to paint by hand; however, the texture is greatly dictated by the image used and is therefore not as flexible. For example, Kingslien explains how that humans tend to have relatively oily skin, which can reflect a considerable amount of light when photographed, causing problems when applied to the model. Although, as displayed within the Essence of Face, unconventional images are very effective for texturing; suggesting these should be used instead of a typical source. IMAGE HERE

There are a couple of methods of texturing from photographic reference in ZBrush; Spotlight and ZApplink. Spotlight uses tools directly in ZBrush, and is praised for which, as it allows the user to effectively apply images to the mesh without exiting the software. ZApplink, however, combines both ZBrush and Photoshop and the tools are claimed to be slightly more powerful than Spotlight’s. Spencer explains that to ensure the texture created is most effective; a combination of Spotlight, ZApplink and Polypaint should be used.

Presentation

Composition

The composition of the final concept is ultimately how it will be seen by viewer, and therefore it is important that a number of individual elements are correctly executed; ensuring it evokes the desired emotion, giving them an "exciting and pleasurable experience" (Spencer, 2012, p. 21).

Focal points are specific elements of the image that a designed to interest the viewer, however can be wasted if they are not positioned correctly. Busch (Busch, 2006, p. 130) and Spencer (Spencer, 2012, p. 21) both agree that a central composition is particularly uninteresting; employment of certain techniques can help prevent this from happening, making sure the final composition portrays the concept in the most effective manner.

The ‘Rule of Thirds’ is a technique in which the image is equally split into nine sections to provide a grid which is designed to guide the composition of the image. Both Spencer (Spencer, 2012, p. 21) and Drew (Drew, 2005, p. 51) explain how by positioning the subject of the image and/or the focal points on these lines and their intersections helps guide the eye around the composition, making for a much more pleasurable experience. Bamberg (Bamberg, 2009) however, contests this point; describing how these rules should not be completely adhered too, explaining that it is sometimes not desirable to place the subject directly on these points, and can be just as effective to slightly offset their positioning.

Keyes (Color Theory: The Mechanics of Color, n.d.) Spencer (Spencer, 2012) and Busch (Busch, 2004) all agree that the use of lines throughout the composition direct the eye to focal points; however Präkel (Präkel, 2010) explains that it has been proven that the use of lines throughout the composition have no effect on how the eye navigates the images, and instead is much more attracted to areas of strong contrast; suggesting that these are areas which should be applied to the ‘Rule of Thirds’, if used.

It is agreed (Spencer, 2012) (Tsai, 2007) (Cantor & Valencia, 2004) that posing the character in the final concept can be greatly beneficial to its outcome; playing a large part in helping direct its motive and ultimately deciding how they are perceived by the viewer. By adding elements of dynamism to the pose, it can be greatly improved, helping portray emotion which in turn suggests a large amount about the character; although, Cantor and Valencia (Cantor & Valencia, 2004) both agree that it can sometimes be beneficial for the pose to be stiffer, as this can highlight elements of the characters personality that a dynamic pose might not. Regardless of the stiffness of the pose, Bamberg (Bamberg, 2009, p. 122) explains that it is very undesirable to use vertically straight lines throughout the final composition; instead, the character should be positioned in an asymmetrical manner, which is far more aesthetically appealing.

Adding a background to the final concept can be a great addition to the design; along with the visual appeal, Su and Zhao (Su & Zhao, 2011) describe that placing it in an environment can help explain elements of the character. Consequently, Kotaki (Pogoda, et al., 2011, p. 124) explains how a simple background can work just as well, giving the final concept some depth. Both Bamberg (Bamberg, 2009) and Busch (Busch, 2006) agree that to make sure the foreground isn’t detracted from, some depth of field should be added to the design; with the subject of the image being sharpest and any other elements being slightly out of focus.

Creature Design

P19 Mixing a complementary colour into a composition can create a focal point and draw the eye; along with making a fairly dull area look more vibrant

P20 A colour will get saturated the further it gets from the eye, this is useful when trying to separate the figure from the background as it can add depth to the image

Colour Theory DVD

Use value to bring the character out from the background

Use value to dictate where your eye looks on the painting, will naturally be drawn towards highest value, as well as contrast in saturation

Colour Theory DVD

Use colour in elements other than the character to draw away from them, maybe if you want them to look pale

Secrets of ZBrush Experts

9 Use a contrast between the foreground and backgro the lattund

Creating the Game Character

Efficiency

Topology

Batman Tri Count (Coombes, 2010)

25k limit on main characters – Batman Arkham Asylum http://www.cgsociety.org/index.php/CGSFeatures/CGSFeatureSpecial/batman_arkham_asylum

Polycount Wiki (Polycount, 2012)

Re-topology means creating a new mesh surface from scratch that conforms to an existing "messy" mesh often created via sculpting. Messy meaning that it either has too much detail or has a topology that's not the best for texturing, shading, or animation. The details from the "messy" mesh are transferred to the new re-topologised mesh by baking a Displacement Map or Normal Map.

Polycount Wiki (Polycount, 2012)

Poles should be placed in flatter areas, along with areas that don’t deform

Male and Female Torso (Williamson, 2012)

Use the flow of the anatomical structure of the character to direct the topology

Use poles to help dictate the flow of topology, allowing for extra loops to be inserted without using excess topology to fit them in

Use poles to "control the topology"; change the direction of a loop and add separate elements to the topology (breasts)

Shows the loops which are important to have on the torso

Street Cop Workflow (Mishu, 2008)

Followed creases of the clothing with topology

Uses triangles a lot to optimize

Face Topology [Community Breakdown] (Del, 2011)

Use topological loops to dictate muscular flow/areas that will deform

Don’t want triangles on the face as "they just don’t deform well at all under animation". Use poles instead, they can effect animation slightly but not as dramatically as triangles, vital to position them properly.

Ancient Pig’s basic deformation tutorial (Rush, n.d.)

By correctly employing the use of animation loops within the topology, the model will deform far smoother when animated

"Collapsing only increases as you add more and more polys into the equation. Better deformation for fewer polys"

"Edges provide the model for something to collapse on", therefore fewer edges means fewer collapses

"Think about a character running: if there’s no loop of polys in the crotch separating the two legs, then when one leg goes forward and the other backward, they’re going to be pulling on each other, resulting in horrible deformations and shading. The loop in the crotch acts as a buffer zone and is there to be stretched."

Unwrapping

Textures

Two 2048x2048 maps, one for head and one for body; Specular, Specular Power, transmission, reflection, color masking http://www.cgsociety.org/index.php/CGSFeatures/CGSFeatureSpecial/batman_arkham_asylum

http://udn.epicgames.com/Three/MaterialExamples.html

Replicating a High Level of Detail

Diffuse/Spec Maps etc., Normal maps, DX11 Tessellation, SSS etc.