I got great results with mvs-texturing. It's not as efficient with texture area or the number of textures, however. Does your approach scale to large meshes, at

I got great results with mvs-texturing. It's not as efficient with texture area or the number of textures, however. Does your approach scale to large meshes, at least in theory? I.e., what is the limit you deem feasible with 1TB RAM and a 3GHz dual socket, 14 cores each? Say, within 2-5 hours or so.