Inside Sabertooth
Learn how Sabertooth uses 3ds Max to create 3D interactive projects, including HBO Go’s Game of Thrones interactive experience
  • 1/3
You are here: Forum Home / Autodesk® Softimage® / Softimage 7.5 / Before building characters
  RSS 2.0 ATOM  

Before building characters
Rate this thread
 
34298
 
Permlink of this thread  
avatar
  • Total Posts: 1
  • Joined: 03 January 2009 07:11 PM

This may sound like a n00b question, but when building a character from a model, let’s say for animating, do you have build the character first, or the skeleton (chains)?

Anything helps. =D



Replies: 0
avatar
  • mantom
  • Posted: 11 September 2009 02:25 PM

You need to establish design of the character (proportions, bend locations, ...).  Once you define all that, modeling and rigging can go on parallel tracks of development. 

Modeling should begin early as possible, but for animation purposes you only need enough geometry to give basic size/shape of the character so something can be plopped onto the skeleton for the animators to see while they animate.  Often that can be achieved with simple primitives stretched and deformed to fit the bones and joints, then later replaced when the official geometry becomes available.  You do it this way for performance as well as to make your pipeline more parallel rather than sequential.  You don’t want animators twiddling thumbs waiting for models before they can begin work.  Chances are whatever model becomes available will need some tweaking for animation such as adjusting locations of vertices and edges to accomodate deformations.  Therefore, it’s very useful for modelers to have a moving skeleton to see if their model will function correctly rather than waiting until they’ve done a complete revision only to have it kicked back weeks later after it gets rigged.  Now they’ve got 2 things to work on - the model they were already working on, plus the model that was kicked back.

The skeleton needs to be constructed to the proportions of the design, and include controls for all the important parts (arms, legs, torso, ...).  You can add controls for the secondary motion stuff later (eg: clothing) when the official geometry becomes available and the primary motion is near completion.  Like with models, it’s not too far fetched to start with a simple bare bones rig for roughing out a character, then moving to a more complete rig later for the polished animation process.  Although majority of studios use one rig, not two.

One handy technique is to envelope your geometry to nulls as deformers, then constrain the nulls to the skeleton.  This provides an abstraction layer so the rig can be swapped if necessary without having to redo all the envelope weighting.  Likewise, it allows several characters to use the same skeleton provided they need the same controls.  Proportions can be adjusted dynamically with pose clips or whatever to account for size differences.

Matt



Replies: 0