I think there's a long graveyard of conceptually better architectures which died because people underestimated costs, from Transmeta, to all sorts of SIMD and M

I think there's a long graveyard of conceptually better architectures which died because people underestimated costs, from Transmeta, to all sorts of SIMD and MIMD systems, to VLIW. I agree it's not too expensive to prototype, but it's super-expensive to do *right*.