One of discreet’s newest Montreal neighbours, Di-O-Matic is also a fresh face on the 3ds max developer scene. Despite being a relative newcomer, the quality of its max plug-ins thus far should ensure that it is already welcome in both communities.
 

Its first two plug-ins, Morph-O-Matic and Cluster-O-Matic, both brought to max powerful tools that are particularly suited to facial animation. With the first of these products providing progressive morphing and the second equipping max with a tool for animating different complex selections, the fact that Di-O-Matic’s newest product again concerns the facial form.
 

Facial Studio builds on these core technologies, bringing a whole suite of facial modelling and animation tools to max. At the heart of the plug-in is the Facial Studio object, which is accessed from a new drop-down in the Create > Geometry section. One simply drags out a default head in the front viewport to begin with. The Modify panel then shows just two rollouts, which are used primarily to control the level of detail that the head is displayed and created using. The option to create the head at low-res and high-res is given, with the more detailed version generating just over 10,000 faces and the low-res just under 8,000. Swapping between these two levels of detail is possible at any time. Both the low and high resolution versions are designed to be used alongside the MeshSmooth modifier, which works just as you’d hope.
 

Also located within this area is the Apply Head Materials button, which generates a Multi/Sub-Object material consisting of eight materials, one for the face, two each for the eyeballs and corneas, and one each for the teeth, tongue and gums. Stepping down into the face material reveals nothing more than a standard skin coloured Blinn-shaded material, but changing this shader will reveal a new subsurface scattering model: the Hanrahan-Krueger shader. This provides two layers, one for the epidermis and one for the dermis, which gives a nice scattering effect, softening the shadows, where the red tint of the dermis colour shows through most clearly.
 

Stepping into the Facial Studio object’s sub-object mode and you’ll discover no less than sixteen different, the first of which provides controls for the overall head generation, with slider-based controls for gender, race, age, fatness and so on, as well as how much the character is to be styled as a caricature or anime. There are also spinners that can adjust the overall shape of the face towards seven different shapes, from pear to egg-shaped, square to round. You can save your head at any point, and there’s a handy Hold/Fetch button available at all times.
 

Several of these sub-object levels control how a particular aspect of the head looks: for instance, within Ear mode you have controls for length (both upper and lower), rotation, and relative position on the head object, as well as detailed controls for how the tops fold over, how pointy they are, even how the tragus appears (the little mound of cartilage between the ear canal and cheek that can be pierced). Similar incredible levels of detail can be applied to the teeth, tongue, chin, forehead, neck, jaw, eyebrows, mouth, cheeks and eyes. There’s also a Deformations sub-object mode where the head model can be tapered, scaled and made to bulge. Indeed, there are over 500 different modelling controls, so Di-O-Matic’s claim that its product is capable of creating any style of realistic or caricatured head is not as far fetched as you might initially think. Indeed, there’s even controls for making your characters ape, cat or wolflike. Furthermore, these controls are all based on MAXScript files, so extending and tweaking the modelling tools further is perfectly possible.
 

This leaves just one further sub-object mode, whose purpose is the same as its name: Animation. Having modelled and saved your head, this is the area where you would build up your facial animation, and despite the controls being contained within just three rollouts; its abilities are quite remarkable. The first of these rollouts contains what are dubbed the Animation Deformers. These begin with controls for the chin and how defined and angular the area that links it with the neck is. There’s a control for expanding and contracting the nostrils, changing the pupil size and altering the vertical position of the eyelids as well as their contraction. Finally, there’s three controls for the mouth that control in a quite straightforward manner to what extent the mouth is open, its width, and the vertical position of the mouth’s corners. Adjacent to this area is a button marked Build Lip Sync, which does just this, constructing the mouth shapes for subsequent lipsyncing. The plug-in actually builds automatically fifteen phonemes (the key shapes that your mouth makes whilst speaking) which then appear in the final rollout, labelled Generators.
 

In addition to the phonemes for lipsyncing, this section features six further generators that enable you to automatically produce the emotions of sadness, anger, joy, fear, disgust and surprise in your head. Further morph targets can be developed using these animation controls and can be added to a custom section, where they are used like all the aforementioned ones, using a simple spinner that can be set between 0 and 1 in some cases, -1 to 1 in others. Finally, this animation sub-object mode also contains a rollout dubbed Pseudo-Muscles. This appears to require some knowledge of anatomy as each control is given its anatomical name – risorius, quadratus, zygomaticus, nasalis and so on – but all that is required is a few minutes of experimentation playing with each of the sixteen facial muscle controls. These names then become far less cryptic and you’ll know which spinners to change to make your head’s temples pulse, top lip curl and so on.
 

Though the copy of the software that was provided for review was a late beta version, there were no problems whatsoever with the plug-in. The only thing that the product was lacking was its documentation and the library of fifty maps that will are bundled with the shipping product. Experimentation with textures revealed that the mapping coordinates produced by the software are clean and shouldn’t require any further manipulation before painting (a map that’s twice as wide as it is high works best for the face map), either using a third-party 3D paint program or just using the UVW Unwrap modifier. Finally, the software adds a new map type to max, which acts like the existing Mix map, but with up to 1,000 layers, with blending and mixing controlled by operations like multiply, darken and so on.
 

Overall then, it’s hard to fault Facial Studio, which actually delivers on all of its ambitious promises. Of course, it can never be all things to all people, but Facial Studio’s unrestrictive approach to head modelling coupled with the quantity and quality of animatable parameters makes for an extremely powerful and versatile tool.