Tuesday, April 24, 2012

some skinning done

better deformations. fewer wizard sleeves and weird yoga position.
making progress on the deformations. still needs some work for the final final version.

Saturday, April 21, 2012

completed controls with basic smooth bind

controls and hierarchy pretty much fully established. the entire rig scales, rotates, and translates uniformly from the global controller. still has a couple of issues, chiefly skin weights and lack of leg IK/FK switch.
some kind of weird yoga

Tuesday, April 17, 2012

hand curl with set-driven keys

full hand curl


single finger curl

the hand controller has a 'hand curl' attribute that controls the curl of all the fingers. each finger can also be individually curled. all of these set-driven keys are created by a Python script.

Friday, April 13, 2012

arm and controls update

here are a few more screen caps of the rig. i updated some of the controls based on feedback during class, including the hips con and way the feet and pole vectors appear. reverse foot has been implemented with set-driven keys, and the IK/FK switch is working on the left arm.

reverse foot joints

toe and ball roll flexed

arm in IK mode

arm in FK mode

updated controls w/ left arm in IK mode

x-ray view


nothing is skinned yet. joint building and hierarchy are progressing. much work to do for the weekend.

Wednesday, April 11, 2012

april rig update: controls

i got around to setting up alisha's control system today. the controls strive to address the requirements identified in earlier posts, and are color-coded for visual clarity. shapes and placement are heavily influenced by the norman rig.


here is an x-ray view that shows the bones and spine control handles:


the parenting and hierarchy are not fully established yet, and no skinning has been done. the rig still needs a 'key control' and necessary interfaces to handle IK/FK switching when required.

it's my goal to give angie a semi-working version of this rig to play around with by sometime this weekend.

Sunday, February 26, 2012

rig requirements

my project partner, angie, has specified some requirements for alisha's motion and interface based on her experience using the morpheus rig in her animation courses. here is everything she's given me so far, starting with more generic rigging notes and moving though specific details about alisha:

the model should be very user friendly for squash and stretch purposes. (i.e. maybe put a joint in the belly button so it can be manipulated for overlapping actions. person jumps up, stomach goes down, when a person falls down, the stomach flies up) also, i am not sure if this is doable, but the degree of rotation in joints should be limited. yes, animators want to break the joints to convey fluid motion if the action is fast, but what i have noticed with other rigs is that the joint will rotate in awkward potions that are sometimes not caught by the eye, so when the model is keyed in between and finally played back, the limb(s) will fly around/twist awkwardly.  also, create joint rigs in toes.

here are a few things to start off with. the rig looks more cartoony, like something by dreamworks rather than disney, because of her proportions. looks fun, snappy.

ik : if animating a wrist, the arm joints will follow it. the hand is the main conductor
fk: controlling joints separately. animator has to animate different joints separately.

shoulder
    1. should have limited rotation just like a human.
              a. in morpheus, that particular rig has no limitation and will spin around too much so that if an animator keys in their poses and does not catch what degree the arm is at, the arm will spin in circles. the control looks like a circle on morpheus.
    2. create a rig in the shoulder that will combine the movements in the shoulder blade/trap.
              a. in the shoulder, there is a rig that looks like a sideways lollipop, o--, an x and a large circle around the whole shoulder. this does not make it difficult to work with all these rigs, however when animating all of these rigs, most likely it comes out not believable/realistic.

elbows
    1. should only move in y axis. (shoulder/wrist rotates in every axis) with the liberty to break the joint.

hand
    1. ik and fk
    2. rig in every joint
    3. need main control outside the hand like in morpheus

torso
    1. should only have ik rigs because it shows more flexibility where as fk looks like the rig is folding and breaking awkwardly.
    2. put an ik rig in the belly button area for overlapping actions.

feet
    1. both ik and fk in feet. ik when to any type of animation besides a walk cycle. fk, walking, running.
              a. knee cap rig should follow the direction the feet are facing. makes sense right! in morpheus, hug problem where he starts to have a broken joint and looks like an insect.
              b. feet face outward, so does the knees, etc.
    2. toes ik

neck
   1. fk. limited range of motion.

face
   1. fk throughout
   2. need rig in middle of forehead to move top brow
   3. there should be 3 rig controls in the eyebrows, on the edges and middle. translations and rotations.
   4. there should be a main control outside of rig model right in front of eyes, to control both eye balls.
   5. rig controls in eyelids so they can open and shut.
   6. rig in both check bones. rotation/translations.
   7. rig in nostrils
   8. rig in jaw to open and close
   9. 3 rigs for upper and lower lips

there should be a main key control hovering the ground. this makes it easier to highlight every joint in the rig so animators and key their frames easier and faster.

the rig needs both ik and fk, however, it is a lot easier to use ik when doing typical animations without walk cycles. when manipulating a rig, there should be humanistic limitations so that the limbs do not twirl around drastically and with no purpose when blocking/ani.

since i have to play with morpheus for my ani projects, i came to realize that in typical poses (lung, line of action in a pose, etc) morpheus' skin folds really awkwardly, with a human it comes natural. morpheus has troubles, especially in the shoulders because of its rotations and translations. it becomes disjointed very easily.

to further make it easier, head and neck should be parented to the spin but should follow the world, body should also follow world. this makes it more natural and easy to track without the head spinning out of control during blocking faze along with limbs.

Saturday, February 25, 2012

preliminary views and joint placement

this is our model, alisha. she has a very exaggerated hip-to-shoulder ratio and a cartoony overall appearance. in keeping with this cartoony style, she'll need to be flexible, with the ability to squash and stretch to meet whatever poses are required.

here, i've made an initial indication of where the primary joints should be placed. the red and blue dots are the actual pivot points; the yellow lines serve only to guide the eye to the next joint in the chain. red dots are limb joints and blue dots are spine/neck/head joints. the two shoulder joints are children of the clavicle joint in the center and will be influenced accordingly.

alisha's hands are a more like normal ones, with five fingers and the standard amount of pivot points. here we see how the pivot points are distributed through the hand.

here is a look at alisha's feet. there are pivots at the ankle and the beginning of where the toes would be, if she had any. the toe joint will allow foot-roll when taking steps, changing direction, etc.

finally, we can see how alisha's joints are laid out from a side view. much of it is in-line, with a small amount of curvature in the lumbar area of the spine and a forward-cheating of the clavicle joint.

that's all for now!