I'm looking at the OpenGL stuff in Haskell and one thing is worrying me. Doing OpenGL in the past I used to pass an array of vertices to the GPU and then a list of indices into that array together with a method of joining them up, e.g. TriangleStrip. In Haskell though, it seems like I'm just supposed to send a list of vertices and the method, meaning that lots of vertices get duplicated. Is Haskell doing something clever under the hood here or is it really that wasteful? Or can I persuade it to use these index lists?

TIA,

Adrian.