I’m looking forward to the swarm robotics Starship could enable. If we really get to the point where we are sending thousands of tons of payload to Mars (building a city) then we could have lots of cheap rovers instead of one big one. I designed a 3D printed plastic rover and I’m super curious how it would behave in Mars. An aluminum framed version with injection molded gears would be much more robust while staying low cost. And it can drive over very large obstacles:

https://youtu.be/gqLQ5Ht8YE0

significant share, by mass, of the parts of such cheap simple rovers would probably be 3D printable - say using something like sand, clay or regolith laser sintering - right there on Mars. Basically one of the most important things to be solved on Mars would be to substitute as much as possible with locally produced components (kind of like Russia after international sanctions in 2014 :) That in particular means that the original designs should be very robust and simple in order to be amenable to such a local production. Kind of like do we want to bring rubber wheels all the way from Earth or may be we can locally produce something which would most probably be worse than rubber and thus the original design should account for increased vibration and roughness of the resulting ride.

Yes local materials production on Mars will be important. But I think there will be a period before that is running where we could still send like ten simple rovers and that would be an interesting mission compared to how we have been doing it.

One thing of note, it’s too cold on Mars for rubber (as far as I can tell). The tires are one part of my robot that wouldn’t work. NASA developed some elastic-mode Nitinol tires that would do the trick.