that's interesting...

[jay]

Early results...
Remote-rovers, terrestrial-controlled (w/delays) observation rate = 0.12
Remote-rovers, Mars-controlled (w/o delays) obs rate = 1 (normalized)
Spacesuited human, obs rate = 5
Shirtsleeve - free human geologist, obs rate = 27

There are too few datapoints to say definitively, but these results imply that a local spacesuited human has something like 40x the science productivity of a Earth-controlled rover...

Comments

[brian1789.livejournal.com]

Two caveats: This study assumed a larger, much-more-capable 2020-class rover, capable of self-navigation, machine vision and onboard planning. The Pathfinder mission was capped at $150M development costs in 1992 dollars... that would be ~220M now, and it didn't include another $120M for the launcher (a Delta II). Or operations costs after launch... plus a bigger, smarter rover would cost more to develop. To do Pathfinder today would cost about $400M, including the launcher and ground operations. A more-capable rover mission would be somewhere around $550-700M total cost.

And the comparison is against one human in the field... actual expeditions would probably send a crew of five, of which three might go out at once.