Wednesday, 30 November 2016

Four: Building the Bones


  With the UV maps done (I mistakenly called this "UV week, it's more like "Rig-and-Skin week") I've been working on the structure for Four. Coincidentally, which will be shown in another post, today's Maya workshop demonstrated using IK splines for tongues and for curves.


A big change form Four is the toes. I was given the suggestion of adding a sixth 'digit' that ran underneath the middle toe. This digit would act as a focal point for the foot's pirouette movements. This makes twisting the foot slightly easier than before.


I have installed the tongue on a spline with a series of bones connected to it. There is one slight technicality however in that I'll need some sort of control for these bones to give Four's tongue some flexibility. I also considered a bend deformer that will allow the tongue to curl.If certain parameters are tied to a control I can hopefully move the curler to specific points on the tongue.

  This was an expectation as the tongue is arguably the most flexible muscle in the body.Fortunately Four's tongue isn't as bizarre as Colo Colo's which makes production a little easier.



  I might leave the hands free of reverse joints, but I am considering an IK-FK switch to make any switching of functions as feet and grasping hands a feature to ease the animation process. I imagine I'll be skinning in a few days, if not next week if I can get a system for  All in all progress is strong and Four's visualisation is working out quite nicely..



  I am currently questioning how to approach the spine. Currently there is an IK handle installed that can be used to twist the bend in the spine in certain directions and keeps the structure rigid as the IK flexes and compresses to give that "arch" that will come whenever Four is sprinting.



  The function of the spine will take some thought. Like with the neck it will be more realistic to spread out any twist or bend through the spine. But due to the location of the arms, I will either require a setup that only bends the spine, or I make clever use of the rotate coordinates when animating.

Regardless, the current ability to curl or straighten the spine as the legs move inward and outward seems quite useful.


Tuesday, 29 November 2016

Four: Starting the Rig


  The UV mapping process was swifter than I initially expected, in part due to the layering I have considered. Everything is just about on there, the eyes could have their own texture map but on the other hand there is still some space on the above UV map not filled that could possibly fit the square UV map of the eye. Except for the core body and the head, the body has a line of symmetry down the middle of it which could be used to give a touch of asymmetry to some of the more easily definable areas of characterisation such as the face or the back. This reduced some of the time spent UV mapping as half of the map was created when mirrored. This was especially fortuitous for the UVs of the claws and teeth, to which there are a lot of models that could otherwise be easily repeated.


With the UVs done I moved on to creating the rig. There might be some tweaking needed but I did not want to repeat the mistake of Colo Colo when it came to 'solutions' that broke the various joint orientations. So I was more careful this time around, making sure the joints were oriented properly and each node was added using a locator.


  It might be tricky to see within the model but there are five bones in the tongue. The tongue is an extremely flexible muscle so I figured this is where the most amount of control is required to allow it to bend, flex and curl. I had been contemplating how to deform it as well, as it appears to me there's going to be a good deal of stretching (nothing extreme, Four doesn't have a snake's tongue), I looked at dog and lizard tongues for ideas on how to construct it.






  I spent some time twisting my wrists and ankles about in order to get an understanding of where the joint's centre was. From what I can estimate,the point of rotation is a little below the bone growths that define the ends of the wrist and the ankles.


  I'm still considering some reverse joints on the feet. The hands, conversely, I could install a reverse joint set maybe with the ability to turn its control over the hands on and off for those times when Four will be using her digits to grab or manipulate objects. 



  Geometry-wise I think she might be about done. Though I might or might not at some point go back to the fins to see how I could make them blend. Or better yet refer back to my influences to see how fish fins, or possibly the fleshy 'sails' on certain breeds of dinosaur attach to the body.

Monday, 28 November 2016

Four: On the Way Through UV Mapping


  This past day, and last night, I began work on the UV maps. As of posting the outside of the head, the front legs and the back legs (except for the toes) are about ready. I still need to refine the UVs for the teeth, the claws and perhaps arrange the fins. Ideally on a single plate but there could be exceptions as the bone requires a different texture setup to the rest of the body..


I figured I might be able to reduce the complexity of the model by creating a mirror-line. Using the clone stamp tool in Mudbox I can easily blend the texture maps so while they don't quite fit right now, the divides should be invisible with some clever tool use.



Friday, 25 November 2016

Four: Geometry Virtually Complete and UV maps


  Development of the foundation model is coming to a close. The feet are roundabout the best I can make them so I moved on to the other parts of the body to get started on UV mapping. There is still a lot to be done with the UV maps but that and rigging will be a task to conduct over the next week.


  Like with Colo Colo, because Four's core body is a tube and will likely be quite prone to squash and stretch I thought it appropriate to keep the relevant UVs straight so that areas that look like they could be stretched considerably can have generous areas for texturing.


  I've flattened the feet a little more to coincide with the sort of decline common between the ball of the heel and foot. Which I feel has helped sell the toes as not being quite so vertical as before. This helped make the membranes feel less thick but not separate them and the tops of the toes and fingers with a massive chasm.


  Finer details for musculature can be made with Mudbox. I'm still unsure if this is the above is the right sort of chest for this kind of creature. It'll have pectorals, certainly. Such muscles are used to drive the shoulders. I could perhaps blend the shoulders a little with the pectorals to suggest connection.


  Before UV mapping I made one last minor tweak to the gums in the form of rearranging the sockets to that the teeth closer to the tip of the snout can more easily cross between each other when munching down on any food.




Thursday, 24 November 2016

Major Project: A little More Definition


  Today was a continuation of the refinement of Four's body. The front hand I think is looking a lot better while the rear foot still needs a little more work.



  So I wasn't spending all my time on just the feet I tried to define the musculature, which I feel will need to have a pass inside Mudbox.to create the finer shapes. A mudbox pass could also help define the base geometry so the normals wrap better, as work I have done on previous Mudbox models the program can calculate the positions of key vertices at the blower levels.of complexity after sculpting on higher-density meshes.








Wednesday, 23 November 2016

Maya Workshops: Photogrammetry


  Today's workshop was in the use of photogrammetry in order to mix computer-generated models with photographs. This technique can be used to create realistic-looking objects using photo-projection. This scene however is only a static image and the camera can't be moved to explore the scene. But it can be used to move objects though a static frame.

  I think this technique is good for showing how the setup of lighting and shadows can make something convincingly real.

Tuesday, 22 November 2016

Major Project: Working the Feet


  With the teeth in an improved state I moved on to sort out the hands and improve the webbing and the claw roots..


  The roots seem fine but the membranes between the I think might need further improvement. It does have a little more thickness to it though which gives it more of a presence. I probably should consider that on my hands at least there's trace of a membrane that could, millions of years ago, have been webbing. As while I could look at otter paws again, their paws, the membranes included save for some species, are covered or obscured with fur. I have a feeling the foundation for getting it right is there, however.



  I tried a little more sculpting on the neck and forearm but I think I will definitely need to look at that tutorial to understand what could be done about the thighs.


Monday, 21 November 2016

Major Project: Refining the Teeth


  With Four's head coming along I focused my attention on her teeth today. Using the techniques Alan suggested I have what feels like a more realistic jawline.


  I think there is still room for improvement, there's quite a bit of geometry cleanup required for one, but this feels like a significant improvement over the previous more simplistic model. The rearmost teeth could do with some raising perhaps, but so far I feel the process is much stronger.


  Four is an insectivore. In part of the story she eats smaller insects and her body is thought to contain a compound likened to spider-silk. So her teeth need to be optimised to pierce or crush hard carapace (on recollection giving her a nutcracker-like set might sell that she can eat insects. But as an upper-level predator she needs to be adaptive and teeth designed to crush hard carapace won't work so well when trying to pierce skin. So like a number of carnivores, she'd have more than one type of tooth in her mouth, each with different functions.


  That may require retraced steps but the crafting of the gum sockets was probably the most taxing part of today's build. That might not need too much changing if I'm going to be swapping teeth out.


  As well as that I've tried using Maya gemoetry to sculpt some of her muscle. I've only done her collar and front leg (which doubles as an arm) so far and might need to do a little work on the shoulder to sell the structure a little more, but it's progress. I'll look into refreshing my memory on building quadruped limbs and body at some point, the thumbnails for the tutorials (where the subject is a horse) suggest they could include tips for sculpting particular muscles.