I think I’m ready to call the Zbrush phase of the first version of April “complete.”  It’s certainly taken long enough.

Much of her body had already been sculpted prior to undertaking the fiasco that was April’s hair sculpting process, but I’ve since reworked the folds on her sleeves, added more definition to her scarf, increased the wrinkle detail on her kevlar vest, and added seams on several of her garments (most noticeably on her jeans).  Most importantly, though, I’ve spent hours and hours carefully adjusting her head size in proportion to her body size, as well as the thickness and curvature of her thighs and hips.  Despite a significant wealth of literature on the topic, this has continually been a point of contention for me, and I recently made a very big change in her proportions, wherein I increased the overall width of her body in proportion to her head.  I think she appears much healthier and more believable now.

Let me know if I’m wrong.  Click below for a larger version of April from several angles.  I hope you like your images big.

Onward to retopology!

Back in June, I promised that I’d soon post an update on the progress of April’s hair sculpting.  I lied.  Instead, I spent the summer applying for jobs, completing art tests, and trading emails with recruiters.  Thankfully, by the time August rolled around, I had an offer letter from Turbine in hand, and finally got an opportunity to put all of my energy toward… searching for an apartment in Boston.  This weighty task was quickly followed by completing some very overdue illustrations for my much beleaguered children’s book project.

Since starting my professional career, however, I’ve devoted all of my available free time toward finishing up my thesis.  The ideological underpinning of the project is pretty well cemented, and most of my very lengthy paper has already been written, so the bulk of my remaining work relates solely to completing the actual artwork.  Below, you can see the product of many many hours spent in zbrush, painstakingly sculpting April’s hair.  The amount of detail I chose to sculpt (versus simply converting color map data and baking out normal map overlays) was a major point of debate for me.  I probably added more sculptural detail than was necessary, but I generally feel that when possible, it’s better to be thorough than to be fast.

As I’ve mentioned previously, additional volume and a few disheveled locks of hair will be added with alpha transparency planes positioned on top of the base mesh; however, the normal map data for these will be baked from height maps, to be painted following retopology and UV projection.  Here are some  off-angle shots offering a better look at all that glorious detail:

There are actually two levels of HD Geometry detail sculpted on top of this (mostly in the braids), but I didn’t bother to render it for these screenshots.  And sadly, that detail probably won’t make it to the normal maps either, since the geometry doesn’t have any UVs yet.

I should also mention that, for the most part, I’m satisfied with April’s facial appearance in its current state, so this will probably be the final look for the first (koinophilic standard) version of the character.  She’s been modified somewhat since the last renders I posted, mainly through subtle changes to her lip shape and a narrowing of the frontal width of her nose.  If you have any complaints or suggestions, speak up now!

I’ve done some polishing of various components of April’s body shape, as well (images to follow), so retopology should follow shortly.

My “graduation” from SCAD and recent relocation from Savannah, GA to Syracuse, NY has kept me from posting any updates about my progress on April, but those things haven’t kept me from continuing to make that progress.  I finally got my Crazy Computing Rig set up again in Syracuse today, so I figured I’d splash a couple of images onto the web.

The following Zbrush renders represent April’s current state of completion (which is to say that she is still very incomplete).

And here’s a closeup of April’s bust (in progress):

Her hair is obviously very incomplete, and even after sculpting detail is added, it will really begin to come to life once I add alpha planes in Maya.  That’s all for now.

I recently finished the orthographic concept art for my first thesis character, introduced in my last blog post as “April.”  Before analyzing the artwork, here’s a little more information about the character’s in-game role:

April is envisioned as a major non-player character in a disaster survival game, which revolves around the idea of a series of successive, catastrophic terror attacks in downtown Boston, Massachusetts.  As a medical student at Boston University, April is on the T when the first disaster occurs.  After tending to her own wounds following the train’s derailment, April retrieves medical equipment from an overturned ambulance, and begins to treat the injuries of as many victims as possible.  By virtue of story events, April becomes companion to a detachment of National Guard troops dispatched to the city (representing the player entity).  April’s medical background and ‘healer’ disposition incite her to encourage the player to seek non-militaristic solutions to game challenges, adding an additional layer of design complexity.  As such, this role is ideally poised to benefit from the persuasive aspect of the attractiveness bias, whereby the character’s attractiveness will facilitate empathetic connections with the player, and increase his/her propensity to heed April’s advice.

A major thrust of my thesis work has been focused on subverting prevailing female stereotypes in modern video games, and April’s design was largely influenced by these concerns.  While presenting my thesis research, I’ve frequently criticized the portrayal of female characters clad in attire inappropriate for their role within the story context.  Look at the female characters in any fighting game, and you’ll see exactly what I mean.  Their armor seems to cover as little skin (and as few vital organs) as possible.  So when designing April, I challenged myself to eliminate as many gratuitous manifestations of her sexuality as I could.  In doing so, I began to define (if only for myself) the hidden boundary between sexual objectification and the necessary visual markers indicative of femininity.

To this end, I resolved to equip April with a kevlar vest (these are included in some EMT gear packages), thereby hiding her breasts, which are arguably the most steadfast visual indicator of sexuality in female video game characters.  The challenge, however, was that breasts are also one of the most obvious markers of femininity, as well.  In order to compensate for the suppression of this feminine feature, I was required to amplify the appearance of other (less overtly sexual) indicators of femininity.

Click the image to see a larger version:

Such “indicators of femininity” refer mostly to her facial features, hairstyle, and body proportions.  My guiding principle was to not allow these concerns to compromise the appropriateness of her attire, considering the story premise.  Aside from the emergency gear she has collected, April is wearing clothing typical of a graduate student, and believably appropriate for class, a day at the library, or running errands.  My focus on practicality necessitated the absence of apparel designed to enhance the sexuality of the female form, such as high-heeled shoes (apparently the footwear of choice for many female game characters).

While working on these drawings, I received a rather surprising amount of feedback with regard to April’s apparent ethnicity.  Nobody could tell which race she belonged to.  Admittedly, the relationship between attractiveness and multi-racial heredity is a topic that intrigues me, but it had not been my deliberate intention to produce a racially ambiguous character.  The fact that this occurred nonetheless could be a testament to the universality of beauty, or a consequence of the ‘prototypical’ appearance I had intended, or simply a result of insufficient facial topological information contained in the 2D drawing.  Whatever the case, I’ve decided that I’d like to preserve that quality in the final character model, and I’m interested to see whether it requires deliberate artistic direction, or occurs naturally.

I’ve included an initial color treatment below, though I confess I’m not entirely pleased with it, as April’s skin tone seems to betray the ethnic ambiguity that existed in the line drawing.

I hope to have some Maya screenshots available soon.

It’s been a while.  Sorry about that.  The last three months have been exceedingly busy, to such extent that I’ve hardly had time to even process my research, much less post it here.  As a quick recounting of recent events, however, I offer the following:

I passed my 45-hour review in February, which gave me academic approval to pursue my thesis project.  And while the work I presented was considered ’sufficient’ by my professors, it did not necessarily align with my own standards of quality, so I chose to refrain from actually starting work on my project until I worked out some weak points in my research.  While the intense desire to quickly complete my degree requirements made this delay extremely frustrating, it ultimately paid off, culminating in a rather significant revision of my topic as recently as three weeks ago.  The product thereof is represented by my working thesis abstract, which I added to the Overview tab two weeks ago.

With a solid foundation in place, I’ve begun work on my thesis project proper, in the form of concept art for my first character.  The stated intent of this character (codenamed April) is to investigate the feasibility of incorporating neotenous physical features in the design of a female character, in order to maximize her physical attractiveness.  I do not intend to test whether or not these features result in an attractive physical appearance, as a great deal of research has already confirmed that fact with surprising certainty.  And while the quality of my execution is by no means assured, my research goal is more directly concerned with the effective impact that an ‘archetypal’ neotenous face will have on viewer/player affect, and how that methodology may fall short of providing the character design with a sense of individuality.  In other words, how does an archetypally neotenous, anthropometrically accurate female face fail to describe a character’s disposition?

I predict that the neotenous ideal will simply act as a shell — a prototype that can serve as the foundation of a character’s face, but one that will require further manipulation to impart a particular personality to the character and distance its appearance from stereotype.  If this proves true, my next character project will attempt to skillfully and deliberately alter the prototype to more greatly infuse a sense of individuality and originality in the character.

But first, we require a prototype as a starting point.  To that end, I’ve drafted the facial orthographics shown below.  The features seen here were derived by compositing the feature proportions of a number of fashion models, celebrities, and local reference models that exhibited a high degree of neonatal facial proportions.  These were conformed to standards of anthropometric measurement and then subtly exaggerated on a per-feature basis (eye size and distance, lip size, etc.) to form one possible manifestation of an ‘ideal face.’

Naturally, there exists a great deal of room for artistic discretion, and by no means am I claiming that this is the pinnacle of attractive facial neoteny.  And while many of you will undoubtedly perceive a resemblance to any number of actresses, I do believe that the image shown here provides a genealogically-nonspecific facial prototype that can be used as a reasonable starting point moving forward.

As a preventative disclaimer, I’ll simply point out that hairstyle is vastly important in defining femininity, and while cultural associations tend to vary, April’s femininity is inescapably lessened by her current lack of hair.  Further, without discouraging feedback, I must point out that the image you see above is still very much a work in progress, so additional changes are most likely forthcoming.