How much more efficient would it be to use a launcher big enough to go directly to the surface of the Moon like Apollo did, compared with using two launches with a lesser rocket, say one with the lander and the second with the crewed command module? By "economy" here I don't mean dollars and cents, but more principally in terms of delta-V and such fundamental measures. If the Apollo program had continued, would it have been beneficial to use Saturn V to build a Gateway in Lunar halo orbit? Was this suggested in the 1960s-70s?
SLS Block I is too small to launch Orion as the command module with a proper Lunar lander attached to it. That's a motivation for the Lunar Gateway concept. How can the relative (physical) costs per Lunar landing of 1) going directly, 2) docking in LEO and 3) building and docking with the Lunar halo Gateway, be estimated and compared?