Actually, the Martian atmosphere makes it extremely problematic from an entry, descent, and landing (EDL) standpoint; the atmosphere is thick enough to pose a real problem with aeroheating, requiring thermal protection systems (TPS) to insulate a payload, but is too thin for subsonic aeroflight or deceleration for a large (>1 metric ton) payload. The payloads that have been delivered to Mars to date use enormous and highly reinforced parachutes deployed at high supersonic speeds but these are highly stressed because of the highly dynamic pressure that comes with the speed and force of deployment. (It is counterintuitive to many that the forces are higher even though the atmosphere is ~1% of the density of Earth, but because of how thin the atmosphere decelerators have to be deployed at very high body speeds.) In the ‘Eighties and early ‘Nineties a lot of work was done on ‘bluff body’ biconic descent capsule designs but they stop becoming effective for lift at supersonic speeds, and more recently work is done on inflatable conical decelerators (e.g. the Low Density Supersonic Decelerator program) but even they have limits of how much payload they can land, and for crewed mission a descent capsule is going to be >6 metric ton even if you separate the bulk of supplies and equipment from the crewed vehicle. Supersonic retropropulsion is effectively the only way to land a large capsule but that requires carrying a large mass of propellant al the way to Mars, substantially increasing logistical requirements and complexity.
Other issues are weeks long dust storms which essentially prohibit reliance upon ground-based solar power (necessitating nuclear fission as at least a backup and supplementary power source), the erosive texture of and toxic perchlorates in the Martian regolith, lack of any radiation shielding in the form of a thick atmosphere (against cosmic rays) or magnetosphere (against charged particles) which will require shielding (most likely by digging several meters under the regolith), and the general immature state of in-situ resource utilization (ISRU); while a proof of concept of extract oxygen from the Martian atmosphere by the Mars Oxygen In-Situ Resource Utilization Experiment (MOXIE), the ability to do so at scale within a realistic energy budget is still in question. As noted above, the only ‘liquid’ water to be found on the Martian surface is in the thick brines of recurring slope lineae in which the water is bound so tightly it doesn’t evaporate in the near-vacuum ambient pressure of the Mars atmosphere. It can be extracted by heating but that begs the question of how much energy that will take and how to deal with the large waste of those brines (again, including a lot of perchlorate salts) that will remain. There are likely subsurface frozen ‘lakes’ at the poles but the ice there will be so cold as to be as hard as granite and still likely contaminated with salts. It might actually just be more feasible to collect ice-bearing small asteroids and send them to the Martian surface rather than trying to extract or synthesize water in-situ.
There remains that challenge of astronauts experiencing freefall conditions for months and then having to adapt to even the ~1/3 gee of Martian gravity, and even how much that lower field will do in ameliorate the various physiological issues of not experiencing Earth-normal gravity; not just musculoskeletal degeneration but a whole host of problems that occur in freefall conditions and which space physiologists will also be a concern in low gravity fields.
It’s not just a matter of space being ‘frictionless’; there is additional impulse required to achieve escape from Earth’s sphere of influence (SOI), injection into orbit around Mars, and descent to the surface. This is only ~23% more in velocity (delta-V) versus landing on the Moon’s surface which doesn’t sound to bad but the reality is that there will be so much more mass of equipment, propellants and other consumables, shielding, et cetera for a crewed mission to Mars versus the Moon that it is several orders of magnitude more effort even for a very minimal mission profile, much less a conjunction-class mission with a ~560 day stay at Mars and all of the equipment that would be required for building any kind of long term base with shielding and recycling of consumables. All of that additional mass might just seem like ‘logistics’, but for a complex and technically challenging mission like this changes in scale are significant challenges in the difficult, cost, and reliability of the mission.
We laugh at this, but IMHO this is the biggest problem to solve with any interplanetary travel concepts, and it often overlooked or just handwaved away. As opposed to a technical problem that has a solution, or many, solving interpersonal conflicts often involves opaque and fuzzy approaches and several attempts in order to resolve (temporarily), which is something engineers can’t handle. Our monkey brains look for the simplest way to resolve the conflict, which usually involves violence, or thoughts of violence (possibly eventually leading to action). Our inbred savagery toward one another is the main reason distant space exploration will remain fantasy, no matter how many technical problems are minimized or solved (or how well things are depicted in science fiction). We’re just not good enough for our ambitions.
Surprisingly, actually quite thick enough for that. Vehicles like the one that delivered Curiosity to the surface were slowed first by a heat shield and then by a parachute before the final braking of retro-rockets.
The way to solve this is to have a heuristic algorithmic machine intelligence that is by any, practical definition of the words, foolproof and incapable of error, monitoring the crew cognitive responses and emotional health via interaction and observation, and with mission responsibilities range over the entire operation of the ship so that it can take control if it appears that the crew may jeopardize the mission.
Sample return is technically challenging but plausible with modest extensions of existing technologies and a large budget. The actually challenging part of this beyond the Mars EDL is returning to Earth without contamination or loss.
When we do start traveling among the planets and stars, I’m confident Dark Star will be considered the most prescient depiction of life aboard a spaceship.
Something I like to outline is the level of technical progress, and how mature a technology is.
Take for example aircraft. The Wright Flyer flew on the 17th of December 1903. On the 9th of February 1969, the first 747 flew. A gap of 65 years. Now, 56 years later, 747s have only just ceased production, and the majority made still fly. When the 747 is 65 years old, most will still be flying. Sure, they are more advanced models, but they are still basically the same aircraft. An engineer from 1969 looking over a new one would only see one huge difference - the electronics and especially computational systems. The rest of the aircraft would look very familiar. Yet the economics of commercial aviation have pushed the manufacturers hard to build better products. The big win is fuel efficiency. Lots of incremental gains over the decades. But nothing game changing.
Now look at spacecraft. If you went back to say 1975, just as Apollo was gone and the STS was lurching forward, if you had told me that spacecraft would look like SpaceX’s Starship in 2025, I would have been incredulous. You have got to be kidding - that is what I had imagined 1985 would look like. 50 years, and all you got is an Atlas that has scaled up to a bit bigger than a Saturn 5? Everything else looks the same. For all useful intents, space launch systems reached technical maturity about the same time as aircraft - about 1970. We have made significant strides in manufacturability - and SpaceX has pushed that. But the capability of the systems are really no different.
Then take that back to everything else. Boeing can’t even make a capsule that works properly. The whole Artemis programme is just pork. The capability is little better than we had 50 years ago. And we have no landers at all. Those under consideration are for the moon, where things are vastly easier. The ISS? A bigger version of what we had with Skylab. Launch cadences are much higher - but that is more a reflection of how modern electronics systems have made putting interesting things into orbit economically interesting.
Where modern technological leaps have improved mostly just come down to electronics. You don’t need massive computational systems to run a spacecraft. As limited as the Apollo Guidance Computer was, it got the job done with 1960’s technology. We can now make something the size a phone that could do a better job. Fine. Saves a few tens of kilograms. Maybe claw 100 or so kilograms out of a spacecraft upgrading electronics and navigation. It is a nice to have. But not a game changer. After that - there is scant new technology that makes a difference. Better welding systems, better composites, CNC machining. Each can make an incremental difference, but we are in the marginal gains over 50 year old capabilities. Like the 747, a modern 747-8 is a better airplane, but it isn’t that much a better airplane. Sure as heck not a moon versus Mars level of better.
We can’t do a Mars sample return yet. Just getting a couple of kilograms of sample tubes and contents back currently eludes us. It might be that a step change in the cost of mass to orbit helps. But it can only do so much. The difficulties of getting things to work, and stay working, in space don’t go away, even if you can afford to make them a bit heavier. And humans have not improved. They are just frail and difficult as they were 50 years ago.
Agree with all of this (and essentially most the rest of your post) but I’ll point out that while computational systems have gotten smaller and vastly more powerful, they’ve also become way more sensitive to the effects of high energy space radiation environments. Discrete element electronics were highly inefficient and not very flexible as often a lot of the logic was built into the circuit instead of the primitive ‘software’ (mostly just bit settings read in from magnetic tape), and even ‘Seventies-era integrated circuits, while subject to single event upsets (SEU) from charged particles but were still pretty resistant from actual damage to natural electromagnetic interference provided they weren’t trying to fly through the inner Jovian system or within the orbit of Mercury. But the modern VSLI microprocessors are so delicate that without a lot of ‘hardening’ and protection they can be easily damaged even in ‘moderate’ trapped particle environments like Earth’s Van Allen Belts and the South Atlantic Anomaly. Being compact (and computationally powerful and efficient) makes them highly desirable, not only from a reduction in mass and increase in capability but because they are also physically more resilient to dynamic environments (shock and vibration) experienced during launch or reentry versus through-hole mounted electronic boards or even more robust surface mount technology, but they also have very real reliability issues that can’t easily be overcome just through redundancy or system resiliency in software error trapping.
And while the massive complexity has allowed abstraction such that we can take pretty ‘generic’ embedded systems and put essentially all of the logic design in software and firmware, this comes with enormous complexity in actually being able to validate the software design (i.e. the requirements decomposition and interfaces) and verify that the implementation meets requirements without unexpected problems, requiring exercising the software to exhaustion in simulation testing on simulated or representative hardware (to catch unexpected latencies, conflicts, non-linearities, et cetera) rather than by purely logical analysis of the design solution. Development of highly reliable flight software and firmware has become such a complex task that it often ends up costing more than the rest of the entire development effort by itself even though it has no material substance and reuse across applications is difficult or not workable.
So, even advances in computing come with their own challenges and costs even as they offer a lot of advantages in terms of capability and flexibility.
Scientific exploration like in “The Martian” is absolutely possible with current science. It’s just that it is
Insanely expensive, and
Would be very, very, very high risk. It’s a hell of a lot further than the moon, and a hell of a lot more complex, so the odds of an accident are much higher.
I’ll argue the odds of accident are darn close to 100% w current tech given infinite money. With real-world plausible money the odds exceed 100% if you know what I informally mean by that expression.
