Skip to content

← ArchivePaper1998

Large Steps in Cloth Simulation

David Baraff, Andrew Witkin

SIGGRAPHAcademic56 descendantsCFX

Implicit integration makes cloth simulation stable at large time steps, the foundation of production cloth (and of Maya nCloth's lineage).

Abstract

The bottleneck in most cloth simulation systems is that time steps must be small to avoid numerical instability. This paper describes a cloth simulation system that can stably take large time steps by coupling a new technique for enforcing constraints on individual cloth particles with an implicit integration method. The simulator models cloth as a triangular mesh, with internal forces derived from a simple continuum formulation supporting anisotropic stretch or compression and a unified treatment of damping forces. The implicit method generates a large unbanded sparse linear system at each step, solved with a modified conjugate gradient method that simultaneously enforces particle constraints exactly, yielding a system significantly faster than previous cloth simulators.

How to read this

Category
Method: a cloth simulation algorithm
Contributions
  • An implicit integration method letting cloth simulation take large, stable time steps
  • A constraint technique enforced exactly on individual particles, coupled into the implicit solve
  • A continuum-style triangle-mesh force model with anisotropic stretch/compression and unified damping, solved by a modified conjugate gradient method
Context
Builds on the deformable-models lineage (Terzopoulos et al.'s Elastically Deformable Models) and became a foundation for production cloth solvers.Builds on: Elastically Deformable Models
Correctness
The key claim is stability at large time steps via implicit integration; the trade-off, common to implicit methods, is numerical damping and per-step cost from solving a large sparse system, which a reader should weigh against the step-size gains.
Clarity
Mathematically dense but well organized; a first pass conveys why implicit stepping helps, real understanding needs a second and likely third pass.
How to read it
First pass for the stability argument and the force model overview; budget a careful second/third pass on the implicit formulation and the constrained conjugate-gradient solver if implementing.

Builds on

Built upon by

Related work

Keywords

This page summarises the entry and links to its original source. The archive never hosts or redistributes the publication itself.Show it in the full archive list →