I've never played with it like that, but I would imagine that if the part is develop-able, sheetmetal should be able to flatten it. I'd think that rolling a part vs bending a part should still be based on k-factors because if you're taking a flat piece and forming it, it has to stretch somewhere on the outside of the neutral axis to at least some degree. Perhaps rolling would stretch a little less, but the outer surface would have to shift/smear relative to the inner surface. It would be interesting to see a comparison of the exact same flat pattern formed into the exact same final shape done with the 2 methods. Maybe using radiography to see what the different forming types did to the grain structure? Maybe precise markings on the edges to see what happens to the straight lines after forming? Hmmmm.....
