> Barring a true breakthrough in something like quantum computing or room-temperature superconductors

Won't that also help grep and then move the asymptote down more?