This paper presents a physically-based model for animating cloth objects, derived from elastically deformable models and adapted to account for the non-elastic properties of woven fabrics. The model represents a cloth object as a deformable surface composed of a network of masses and springs, with the movement evaluated using numerical integration of the fundamental law of dynamics. The authors identify a "super-elastic" effect where high stresses lead to unrealistic local deformations, which cannot be mitigated by increasing spring stiffness without significantly increasing computational cost. They propose a new method inspired by dynamic inverse procedures to adapt the model to the stiff properties of textiles, effectively controlling the deformation rates of springs. This approach helps avoid the "super-elastic" effect and maintains realistic behavior, as demonstrated through various tests, including hanging sheets and flags in strong winds. The results show that the new model is more efficient and produces more realistic animations compared to traditional elastic models.This paper presents a physically-based model for animating cloth objects, derived from elastically deformable models and adapted to account for the non-elastic properties of woven fabrics. The model represents a cloth object as a deformable surface composed of a network of masses and springs, with the movement evaluated using numerical integration of the fundamental law of dynamics. The authors identify a "super-elastic" effect where high stresses lead to unrealistic local deformations, which cannot be mitigated by increasing spring stiffness without significantly increasing computational cost. They propose a new method inspired by dynamic inverse procedures to adapt the model to the stiff properties of textiles, effectively controlling the deformation rates of springs. This approach helps avoid the "super-elastic" effect and maintains realistic behavior, as demonstrated through various tests, including hanging sheets and flags in strong winds. The results show that the new model is more efficient and produces more realistic animations compared to traditional elastic models.