AFAIK, the way to do this is to get your models vertices outputting to a matrix. (jit.gl.model @matrixoutput 1). Then jit.expr could be applied to the matrix data.

As a quick way to check this out you could substitute jit.expr in [p add_noise] of the jit.gl.model help file (enabling matrixoutput) .

Also, there is an Appendix B: THe OpenGL Matrix format that will probably be of use to you for deeper investigations…

HTH

]]>