Will Mars-bound Starships actually be reused?

[steveleach]

Robotbeat/meekGee -

Would you be able to work together to build a spreadsheet model, with a set of adjustable parameters, where you both agree on the model but have different opinions on the values for the parameters? We could then use that to explore the possibilities.

[Robotbeat]

How are you going to get more than 100,000 people to fly to Mars every synod?

You'll need thousands of outbound ships over many (tens of) synodes, and because of the very long time span involved, you can't treat the manufacturing capacity or ship design or ship size as constants to be divided by 10 or 20...

For those time spans, fabrication capacity will always increase, either by number of ships or size of ships, and 10-year-old designs might still work, but they won't provide the kind of cost savings you're imagining through reuse.

Doubling the number of ships produced every synod just means that you keep building more factories.  More high bays. Trains more people.  Tesla did it, right?

This wil be the same, but the cars will all be going to this place that's 2 years away.

Except a lot of people already use cars, already hate having to pay for gasoline, already would like really fast and/or advanced tech cars. The addressable market for Tesla is like 2 billion people.

How many people want to and can afford to go to Mars? Tesla wouldn’t have had that growth without demand for their product.

So it seems to me that finding a way to get the ticket price extremely low has got to be part of this. And having more realistic expectations for demand. Also, think about what stuff Martians could potentially do just as well (or better) as Earth, since Martians will need good jobs and economic activity.

[waveney]

Except a lot of people already use cars, already hate having to pay for gasoline, already would like really fast and/or advanced tech cars. The addressable market for Tesla is like 2 billion people.

How many people want to and can afford to go to Mars? Tesla wouldn’t have had that growth without demand for their product.

So it seems to me that finding a way to get the ticket price extremely low has got to be part of this. And having more realistic expectations for demand. Also, think about what stuff Martians could potentially do just as well (or better) as Earth, since Martians will need good jobs and economic activity.

The first few hundred people won't be paying themselves.  The initial colonists will be:  Scientists, Engineers, those with vision (and useful skills) a diverse but highly capable crew to establish a colony, start on Mars production of many resources (Probably steel, glass and plastics, to farm (highly efficiently) to cook, to provide medical support, to provide childcare, to teach, to maintain equipment, replace equipment, build the base, tunnel, clean, maintain spacesuits, farm, prospect for resources.  There will be a thousand jobs and only a few people to do them.

Later when there like seams like they are doing better than those stuck on Earth, many will want to pay to go there.  I suspect those paying will be small in number initially, as most will still be those with useful skills.

As costs to get to Mars drops, those paying may go up, but will remain a small portion for many synods.   As Mars will be very depended on those with skills rather than money.

[laszlo]

...Later when there like seams [their life seems] like they are doing better than those stuck on Earth, many will want to pay to go there. ...

When life underground on a dry, essentially airless and freezing world is perceived as doing better than living on Earth, the people and infrastructure to travel to Mars will be gone.

[Slarty1080]

I actually don’t think Mars will be power starved.

We know how to make almost nothing on Mars.

The only things we've made on Mars are electricity from solar and nuclear (radioisotope). AND oxygen (and carbon monoxide fuel, I suppose) from MOXIE.

Propellant, and the power to produce it, is pretty much the only thing we know how to make on Mars right now with high certainty.
<Snip>

In a sense you're right. We have very little experience with producing anything useful on Mars and even the electrical generation from radioactive isotopes is unlikely to scale well. What is needed is a large scale fusion reactor.

There will need to be a good deal of development work with anything needed on Mars, including nuclear power (reduced gravity and cooling differences etc etc), but a vast range of chemicals can be synthesised from simple gaseous components available on Mars using well established chemical technology that "we" are very familiar with.

Given sufficient energy, water and the Martian atmosphere a vast amount can be produced. Nitrogen and argon are available direct. Hydrogen and oxygen from water via electrolysis, methane via the sabatier process and carbon monoxide and carbon from direct reduction of carbon dioxide.

The Fischer–Tropsch process (in use since the 1930's and still widely used today) can be used with a variety of catalysts to combine carbon monoxide and hydrogen into a wide range of aliphatic carbon compounds including alkenes such as ethylene, propylene and butylene, alkanes such as pentane, hexane, octane and heavier molecular weight solid waxes.

Ethylene is probably the most commonly synthesised organic chemical in use on Earth today and can be further converted into a wide range of other chemicals such as ethyl alcohol, acetic acid, ethylene glycol, ethylene oxide, polyethylene and many more. The chemistry won't hold things up BUT it will take an awful lot of energy up front to generate the hydrogen needed at around 40-50MW for each tonne of hydrogen generated.

[andrewmcleod]

I would point out two things:

Firstly, that it will always be easier to make things on Earth than Mars in general, with only occasional exceptions (in the near future sense where we don't have terraforming/massive apocalyptic devastation of Earth etc.).

That will be true even if the population of Mars increases to match that of the Earth, which is not plausible any time soon.

Anything you can do on Mars will ordinarily be easier on Earth where you don't need pressure vessels/radiation protection/a massive amount of infrastructure just to support life.

Secondly, dreaming of 'self-sufficiency', trade between Earth and Mars, colonization etc. is science fiction at this point. That doesn't mean it won't happen (plenty of scifi does, albeit often not the way we expect), but there are really not many reasons why it would. It would be easier to fill LEO with people if we wanted to 'colonise space' - we can already do this. Yet we haven't got any further than a dozen people in a few tin cans. Why? Because it is very expensive, and there is at present no economic reason to go to LEO.

There is at present no economic reason to go to Mars. I don't think there's any likely prospect of any economic reason appearing in the near future. That leaves only a few reasons:
1) science and 'exploration' funded by governments (or otherwise) for its own sake
2) very very long-term planning as a backup plan to the Earth; the sort which Musk is advocating but governments are typically very, very bad at
3) FOMO (fear of missing out) - the idea that if 'we' don't colonise Mars, the Chinese/Russians will do.
etc. None of which require colonization (not the same as maintaining a permanent presence).

With all this in mind, I think Starships will be reused, simply because I don't think there will be long-term colonists to Mars any time soon in the same way there are not really long-term colonists to Antarctic research bases - at least for the first 10-20 years or so. After that, its anyone's guess...

[andrewmcleod]

I should add that I think colonization of Mars is actually inevitable eventually, just that we are nowhere near that with science and technology yet. We don't colonize plenty of the Earth because it is 'inhospitable', yet there is nowhere on Earth as inhospitable as Mars.

[Zed_Noir]

<snip>
I can imagine SpaceX producing 100 Starships per year... after all, they produce that many Falcon upper stages per year right now (roughly). But much beyond that seems very unlikely. The good news is that 100 starships per year, if reused around 10 times, is ~2000 Starships per synod, maybe half of those available for Mars (and passengers), so you get your 1000 Mars Starships per synod, around 100,000 passengers per synod. That's enough for a sizable Mars city, maybe a million people.
<snip>

SpaceX could copy Tesla and build a few more ship yards for Starship. So could have about 400 to 500 Starship equivalent builds annually with a Super Heavy equal to 3 Starships. So that is a lot of Starships. Eventually building the Starship wave to Mars each synod in multiples of 100 units. However don't expect the Starships going to Mars be reuse more than 7 planetary transits before being retired on Mars.

When infrastructure is set up to receive and support more colonist. Estimate the first few synods after will have about 10K colonists migrating to Mars per synod. After that the viability of the Martian economy determines how rapid the population of the Martian colony increases or possibly decrease.

[steveleach]

Except a lot of people already use cars, already hate having to pay for gasoline, already would like really fast and/or advanced tech cars. The addressable market for Tesla is like 2 billion people.

How many people want to and can afford to go to Mars? Tesla wouldn’t have had that growth without demand for their product.

So it seems to me that finding a way to get the ticket price extremely low has got to be part of this. And having more realistic expectations for demand. Also, think about what stuff Martians could potentially do just as well (or better) as Earth, since Martians will need good jobs and economic activity.

The first few hundred people won't be paying themselves.  The initial colonists will be:  Scientists, Engineers, those with vision (and useful skills) a diverse but highly capable crew to establish a colony, start on Mars production of many resources (Probably steel, glass and plastics, to farm (highly efficiently) to cook, to provide medical support, to provide childcare, to teach, to maintain equipment, replace equipment, build the base, tunnel, clean, maintain spacesuits, farm, prospect for resources.  There will be a thousand jobs and only a few people to do them.

Later when there like seams like they are doing better than those stuck on Earth, many will want to pay to go there.  I suspect those paying will be small in number initially, as most will still be those with useful skills.

As costs to get to Mars drops, those paying may go up, but will remain a small portion for many synods.   As Mars will be very depended on those with skills rather than money.

Elon seemed to think that the standard way to emigrate to Mars would be to sell your house on Earth to buy a ticket, so he wants to get that ticket price down to the value of a middle-income house. That would open it up to hundreds of millions of people.

[RoboGoofers]

my gut feeling is that it'll take a lot of refurb to get a ship back up to snuff after a round trip. If you just cashed in your life savings and were going to Mars would you want the 20 year old version of the space toilet, or the newest model?
So they'll have to spend parts and labor to strip, inspect, rebuild and certify returned ships. The parts and labor could instead just be spent at the factory making new ships. Refurbing a synods worth of ships would be an entirely different process flow but with the same scale as new ship production.

However, they won't even know what's going to still work or not until they get a handful of ships back (we need to get a handful of people back to make sure they're not irreparably broken by the trips, too.)

what i think will happen is they'll strip out the crew amenities and make them low risk commodity haulers and then just leave them on the surface after a second trip back to Mars.

edit: another thing to consider is how SpaceX treats the lower numbered SNs: they just scrap them when they diverge too far from the current design. They could chop and rebuild them, but it's just too much effort.

[meekGee]

How are you going to get more than 100,000 people to fly to Mars every synod?

You'll need thousands of outbound ships over many (tens of) synodes, and because of the very long time span involved, you can't treat the manufacturing capacity or ship design or ship size as constants to be divided by 10 or 20...

For those time spans, fabrication capacity will always increase, either by number of ships or size of ships, and 10-year-old designs might still work, but they won't provide the kind of cost savings you're imagining through reuse.

Doubling the number of ships produced every synod just means that you keep building more factories.  More high bays. Trains more people.  Tesla did it, right?

This wil be the same, but the cars will all be going to this place that's 2 years away.

Except a lot of people already use cars, already hate having to pay for gasoline, already would like really fast and/or advanced tech cars. The addressable market for Tesla is like 2 billion people.

How many people want to and can afford to go to Mars? Tesla wouldn’t have had that growth without demand for their product.

So it seems to me that finding a way to get the ticket price extremely low has got to be part of this. And having more realistic expectations for demand. Also, think about what stuff Martians could potentially do just as well (or better) as Earth, since Martians will need good jobs and economic activity.

We're diverging.  I was just showing that doubling production every synod is not unheard of, directly answering your question.

And back to the thread title - if production keeps doubling, reusing ships saves very little, burdens the colony building effort, and deprives them of major hardware they could have gotten for free and early.

[oldAtlas_Eguy]

The estimate of the cost of an SS to manufacture is somewhere currently between$1000/kg and $500/kg ($150M to as low as $75M). That for the near future cost level may not change much as there is likely a push to reduce dry weight to increase payload capacity in the short term to a dry weight of ~125t or a cost range of $125M to as low as $62.5M. Then the costs will start going down even more while the weight stays the same causing the $/kg cost range to start dropping as well. Such the lower end starts going below $500/kg. The reason a cargo SS will be cheaper than an aircraft is that a larger proportion is cheap structure vs expensive equipment. A crewed SS will definitely be at the higher end of $1000/kg while a cargo will like be closer to the lower end around $500. Over time both will start to slowly decrease.

Be careful of analogies. Starship system is more closely resembling a large truck (cargo) or bus (crewed) vs the F9 system which is more like that of a van (cargo) or very small motorhome (crewed). Note FH falls somewhere between the two vehicle systems. So usages of a smaller system's demand may not translate very well to the other system because their usage is different.

Normally scale is its own cost per kg reduction in transportation. Such that any significant amount of stuff going to one point is cheaper with a larger vehicle the longer the distance. That is what is going on with the already cost per vehicle $/kg dry weight of Starship system (Starship system includes various versions: Tanker, Cargo, Crew [Note the Cargo and crew may have subversions for use with the 3 major destinations LEO, Lunar, Mars]) vs F9 system (F9 system includes all its parts including Dragon for specialized cargo and crew delivery).

Early Starship system design evolutions will be rapid. But reuse will be performed. If for nothing else than to produce the technology to do reuse where it will be most economical. Note is that unless you have a reusable system you will not know if it is cheaper than using an expendable version of the system. So vehicles will be returned and designs upgraded to get the costs of the reuse of the system down to see just what the economics can be for cargo and crew deliveries to certain locations like LEO, Moon, or Mars.

After this period the question of (Will Mars-bound Starships actually be reused?) will be answered in detail because of economics and need/demand.

We can model to identify the what if cases and then what the decision would be if a certain set of cases is what is believed will happen. But until detailed data both economically as well as a more stable and capable design emerges, we will not be able to come to definitive answer for all cases of Mars bound Starships over the larger current just 9m diameter versions many design iterations life before it gets completely replaced by some other sized vehicle and general operational concept.

[Robotbeat]

I actually don’t think Mars will be power starved.

We know how to make almost nothing on Mars.

The only things we've made on Mars are electricity from solar and nuclear (radioisotope). AND oxygen (and carbon monoxide fuel, I suppose) from MOXIE.

Propellant, and the power to produce it, is pretty much the only thing we know how to make on Mars right now with high certainty.
<Snip>

In a sense you're right. We have very little experience with producing anything useful on Mars and even the electrical generation from radioactive isotopes is unlikely to scale well. What is needed is a large scale fusion reactor.

There will need to be a good deal of development work with anything needed on Mars, including nuclear power (reduced gravity and cooling differences etc etc), but a vast range of chemicals can be synthesised from simple gaseous components available on Mars using well established chemical technology that "we" are very familiar with.

Given sufficient energy, water and the Martian atmosphere a vast amount can be produced. Nitrogen and argon are available direct. Hydrogen and oxygen from water via electrolysis, methane via the sabatier process and carbon monoxide and carbon from direct reduction of carbon dioxide.

The Fischer–Tropsch process (in use since the 1930's and still widely used today) can be used with a variety of catalysts to combine carbon monoxide and hydrogen into a wide range of aliphatic carbon compounds including alkenes such as ethylene, propylene and butylene, alkanes such as pentane, hexane, octane and heavier molecular weight solid waxes.

Ethylene is probably the most commonly synthesised organic chemical in use on Earth today and can be further converted into a wide range of other chemicals such as ethyl alcohol, acetic acid, ethylene glycol, ethylene oxide, polyethylene and many more. The chemistry won't hold things up BUT it will take an awful lot of energy up front to generate the hydrogen needed at around 40-50MW for each tonne of hydrogen generated.

Or just use solar.

I think space nerds continually under-estimate solar.

Perovskite plus mono-silicon cells that are being built out now at scale (and for very cheap) will produce as much power on Mars (33-34% efficiency has been demonstrated) as the kind of original solar cells on ISS (14% efficient) or many of the large scale thin film utility scale solar farms. Solar cell efficiency improvements have compensated for the increased distance from the sun compared to Earth. And in fact, even without efficiency improvements, solar at Mars’ equator has about the same sunshine watts per area averaged over a year as most of Germany.

But even more than that, mars has plenty of area. There’s really no reason, once you have just basic equipment on Mars, that you couldn’t deploy square kilometers of solar on Mars. In fact, lower gravity and much lower wind forces means you don’t necessarily need as strong of frames as in Earth, saving a lot of mass. You probably want single axis tracking to allow increased production near the equator but also integrated dust mitigation (at high angles, the dust is highly likely to just fall or be blown off, as Spirit and Opportunity showed).

Slightly longer term:

I can see vast solar farms on Mars, potentially even larger than ones on Earth. Orbital mirrors could reflect sunlight down, thereby allowing production day and night and at intensities similar to (or even higher than) Earth. If the orbital mirrors are lightweight enough, this could allow solar production on Mars for vastly less than on Earth. Because there’s no way you could get away with orbital mirrors like that on Earth as the spot size would be larger than on Mars and because you have a lot of neighbors. I think once you’re generating Gigawatts of solar on Mars, extremely lightweight solar mirrors in medium orbit start making sense for this sort of thing.

At Terawatt-scale, you just put the mirrors at Areosynchronous orbit and you’ll be able to efficiently use the whole spot size. Placing the solar farm at Olympus Mons (once you have a railroad or something built up there) would also give you immunity to the occasional long dust storm while still having coupling to the atmosphere.

Terawatt scale sounds insane, but since solar cells are 1000W/kg roughly, that’s “only” about 1 megaton of solar cells. And solar cells are as low as 7.5¢/Watt now and are reducing long term, so buying the solar cells would cost $75 billion… who knows how much less by the time this is ready.

Let Wright’s Law work for you on Mars, generating cheap power and propellant for cheaper than on Earth.

[Twark_Main]

My spreadsheet takes into account:
- increasing launch cadence of 2x per synod
- reduction of fabrication coat of 0.8x per doubling
- reuse lag time of either 1 or 2 synods

Just curious, how does this change if the learning rate (and "doublings") are calculated w/r/t the total number of rockets SpaceX has manufactured, F1 and F9 included?

It seems like most of those lowest-hanging discoveries would be found in these very early stages, hence the extremely rapid learning rate. By only counting the total number of Starships in the "doublings," there's a risk overestimating the true (Wright's Law) rate of learning.

I suspect the real learning rate is going to fall somewhere in the middle, basically just an all-time-units curve superimposed on a current-gen-units curve, with step discontinuities in between generations. Does that make sense?

Did I miss the spreadsheet? Sorry catching up

True, but the conclusion stands even without a cost learning curve.

Just assume a doubling of the outbound fleet size every synod, and then even a super fast 1-synod turn around ship (which IMO is somewhere between plain impossible and just prohibitive) is not that attractive.

If it's a 2-synod turn-around, it's basically worthless.

Is it possible to get that in numbers, instead of qualitative statements?

I'd run numbers myself but I can't find the spreadsheet.

This is before you factor in Wright's Law, so if you're suspicious of it, just let it be.

I'm not trying to advocate for a particular "side," I'm just trying to explore the shape of the analysis landscape.

[meekGee]

My spreadsheet takes into account:
- increasing launch cadence of 2x per synod
- reduction of fabrication coat of 0.8x per doubling
- reuse lag time of either 1 or 2 synods

Just curious, how does this change if the learning rate (and "doublings") are calculated w/r/t the total number of rockets SpaceX has manufactured, F1 and F9 included?

It seems like most of those lowest-hanging discoveries would be found in these very early stages, hence the extremely rapid learning rate. By only counting the total number of Starships in the "doublings," there's a risk overestimating the true (Wright's Law) rate of learning.

I suspect the real learning rate is going to fall somewhere in the middle, basically just an all-time-units curve superimposed on a current-gen-units curve, with step discontinuities in between generations. Does that make sense?

Did I miss the spreadsheet? Sorry catching up

True, but the conclusion stands even without a cost learning curve.

Just assume a doubling of the outbound fleet size every synod, and then even a super fast 1-synod turn around ship (which IMO is somewhere between plain impossible and just prohibitive) is not that attractive.

If it's a 2-synod turn-around, it's basically worthless.

Is it possible to get that in numbers, instead of qualitative statements?

I'd run numbers myself but I can't find the spreadsheet.

This is before you factor in Wright's Law, so if you're suspicious of it, just let it be.

I'm not trying to advocate for a particular "side," I'm just trying to explore the shape of the analysis landscape.

See below.

I wasn't saying you're taking sides... just that if you're skeptical about the Wright discount, the argument stands even without it so just ignore it.

The formulas are self-explanatory, though this doesn't mean I didn't make mistakes

The point should be clear even without the table - reuse buys you very little when the volume grows exponentially (as it must, if you want to reach thousands of ships).

Exponential growth is possible, as demonstrated by Tesla.

So, Instead of ship cost being 10% of nominal (due to 10 reuses), it is is 50% at best, and more like 70%-80%.

Then, on top, as noted above, these additional issues exist and are NOT captured by the table:

- The ships left on the surface have a huge value, since large storage tanks, in large quantities, are an absolute necessity for a Mars colony, and will have to hold precious fluid, often under pressure, so must be high-quality tanks.  Not only do landed one-use ships provide them, they provide them EARLY.
- Sending ships back adds a burden on the colony, in term of building the launch infrastructure, and also in terms of ISRU. It's not impossible, but in a new colony, resources are limited, and every burden counts - and again this burden occurs EARLY which makes it worse.
- 1-synod reuse adds extra difficulty on both the Mars and Earth sides: More demanding trajectories, heat shielding, and expedited processing times..

[meekGee]

My spreadsheet takes into account:
- increasing launch cadence of 2x per synod
- reduction of fabrication coat of 0.8x per doubling
- reuse lag time of either 1 or 2 synods

Just curious, how does this change if the learning rate (and "doublings") are calculated w/r/t the total number of rockets SpaceX has manufactured, F1 and F9 included?

It seems like most of those lowest-hanging discoveries would be found in these very early stages, hence the extremely rapid learning rate. By only counting the total number of Starships in the "doublings," there's a risk overestimating the true (Wright's Law) rate of learning.

I suspect the real learning rate is going to fall somewhere in the middle, basically just an all-time-units curve superimposed on a current-gen-units curve, with step discontinuities in between generations. Does that make sense?

Did I miss the spreadsheet? Sorry catching up

Regarding F9, it's operating in a different economic reality, where indeed if you fly a booster 10x times, it cuts the cost of the booster by 10 (and then add refurb costs, but those are very low)

It does not have a 2 or 4 year reuse lag, nor is SpaceX engaged in a growing effort to make thousands of them - so reuse is basically a no-brainer.

[TheRadicalModerate]

...Later when there like seams [their life seems] like they are doing better than those stuck on Earth, many will want to pay to go there. ...

When life underground on a dry, essentially airless and freezing world is perceived as doing better than living on Earth, the people and infrastructure to travel to Mars will be gone.

I could see life being roughly the same as Earth with enough creature comforts, and a lot of opportunity thrown in to boot.  But this is why a smart colony operator will throw excessive amounts of hardware at the problem, to attract the <1σ-crazy colonists sooner.

This is what keeps the cargo-to-colonist ratio very high for a long time.  I'm assuming that, once you get above about 20 people on a Starship, it makes more sense to put passengers and cargo on separate ships, both customized for their purpose.

I expect that passenger ships will have a decently high chance of returning, given that somebody's always going to be rotating off of their Mars assignment and back to something on Earth.  But the cargo ships are only going to be returned if the resources devoted to returning them aren't better invested in making the base/colony as nice as possible, as soon as possible.

Despite Elon saying that he's not that interested in operating a colony, we tend to have an assumption in a lot of these discussions that the interests of the colony operator and SpaceX are aligned.  That'll be true to a certain extent; after all, SpaceX only makes money on Mars if the colony operator(s) are successful and need a bunch of flights.  But what does that mean for shared resources like power, heat rejection, and water?  Who decides how they're allocated?  Will SpaceX bring its own dedicated resources to return ships?  Does dedicating those resources change the economics substantially?

Another possible factor on cargo ship return:  How many payloads will be actually unloaded before deployment, vs. how many will simply be built into Starships?  I've played around a lot with self-contained prop plants, but the same logic is likely to apply to an awful lot of equipment.  Rather than unloading it, unboxing it, and setting it up, Starship's large volume makes it a lot easier to engineer stuff just to be hauled--in the Starship--into place and hooked up to common resources to achieve its task.

[Robotbeat]

My spreadsheet takes into account:
- increasing launch cadence of 2x per synod
- reduction of fabrication coat of 0.8x per doubling
- reuse lag time of either 1 or 2 synods

Just curious, how does this change if the learning rate (and "doublings") are calculated w/r/t the total number of rockets SpaceX has manufactured, F1 and F9 included?

It seems like most of those lowest-hanging discoveries would be found in these very early stages, hence the extremely rapid learning rate. By only counting the total number of Starships in the "doublings," there's a risk overestimating the true (Wright's Law) rate of learning.

I suspect the real learning rate is going to fall somewhere in the middle, basically just an all-time-units curve superimposed on a current-gen-units curve, with step discontinuities in between generations. Does that make sense?

Did I miss the spreadsheet? Sorry catching up

True, but the conclusion stands even without a cost learning curve.

Just assume a doubling of the outbound fleet size every synod, and then even a super fast 1-synod turn around ship (which IMO is somewhere between plain impossible and just prohibitive) is not that attractive.

If it's a 2-synod turn-around, it's basically worthless.

Is it possible to get that in numbers, instead of qualitative statements?

I'd run numbers myself but I can't find the spreadsheet.

This is before you factor in Wright's Law, so if you're suspicious of it, just let it be.

I'm not trying to advocate for a particular "side," I'm just trying to explore the shape of the analysis landscape.

See below.

I wasn't saying you're taking sides... just that if you're skeptical about the Wright discount, the argument stands even without it so just ignore it.

The formulas are self-explanatory, though this doesn't mean I didn't make mistakes

The point should be clear even without the table - reuse buys you very little when the volume grows exponentially (as it must, if you want to reach thousands of ships).

Exponential growth is possible, as demonstrated by Tesla.

So, Instead of ship cost being 10% of nominal (due to 10 reuses), it is is 50% at best, and more like 70%-80%.

Then, on top, as noted above, these additional issues exist and are NOT captured by the table:

- The ships left on the surface have a huge value, since large storage tanks, in large quantities, are an absolute necessity for a Mars colony, and will have to hold precious fluid, often under pressure, so must be high-quality tanks.  Not only do landed one-use ships provide them, they provide them EARLY.
- Sending ships back adds a burden on the colony, in term of building the launch infrastructure, and also in terms of ISRU. It's not impossible, but in a new colony, resources are limited, and every burden counts - and again this burden occurs EARLY which makes it worse.
- 1-synod reuse adds extra difficulty on both the Mars and Earth sides: More demanding trajectories, heat shielding, and expedited processing times..

I agree we’ll hopefully get doubling each synod for a while, but I think it’ll asymptote dramatically after reaching 100 or so ships. In fact, the curve might be 1 starship, then 10, then 30, then 60, 100, and then constant at 100 new Mars Starships per year.

BUT you can’t just have exponential growth indefinitely without the demand coming from somewhere. Tesla has huge demand for cars for obvious reasons. Falcon & Starship have Starlink. But Starlink doesn’t need 1000 new starships per year!

I just don’t see where this is coming from. I don’t think the demand will be there WITHOUT making them reusable.

[meekGee]

My spreadsheet takes into account:
- increasing launch cadence of 2x per synod
- reduction of fabrication coat of 0.8x per doubling
- reuse lag time of either 1 or 2 synods

Just curious, how does this change if the learning rate (and "doublings") are calculated w/r/t the total number of rockets SpaceX has manufactured, F1 and F9 included?

It seems like most of those lowest-hanging discoveries would be found in these very early stages, hence the extremely rapid learning rate. By only counting the total number of Starships in the "doublings," there's a risk overestimating the true (Wright's Law) rate of learning.

I suspect the real learning rate is going to fall somewhere in the middle, basically just an all-time-units curve superimposed on a current-gen-units curve, with step discontinuities in between generations. Does that make sense?

Did I miss the spreadsheet? Sorry catching up

True, but the conclusion stands even without a cost learning curve.

Just assume a doubling of the outbound fleet size every synod, and then even a super fast 1-synod turn around ship (which IMO is somewhere between plain impossible and just prohibitive) is not that attractive.

If it's a 2-synod turn-around, it's basically worthless.

Is it possible to get that in numbers, instead of qualitative statements?

I'd run numbers myself but I can't find the spreadsheet.

This is before you factor in Wright's Law, so if you're suspicious of it, just let it be.

I'm not trying to advocate for a particular "side," I'm just trying to explore the shape of the analysis landscape.

See below.

I wasn't saying you're taking sides... just that if you're skeptical about the Wright discount, the argument stands even without it so just ignore it.

The formulas are self-explanatory, though this doesn't mean I didn't make mistakes

The point should be clear even without the table - reuse buys you very little when the volume grows exponentially (as it must, if you want to reach thousands of ships).

Exponential growth is possible, as demonstrated by Tesla.

So, Instead of ship cost being 10% of nominal (due to 10 reuses), it is is 50% at best, and more like 70%-80%.

Then, on top, as noted above, these additional issues exist and are NOT captured by the table:

- The ships left on the surface have a huge value, since large storage tanks, in large quantities, are an absolute necessity for a Mars colony, and will have to hold precious fluid, often under pressure, so must be high-quality tanks.  Not only do landed one-use ships provide them, they provide them EARLY.
- Sending ships back adds a burden on the colony, in term of building the launch infrastructure, and also in terms of ISRU. It's not impossible, but in a new colony, resources are limited, and every burden counts - and again this burden occurs EARLY which makes it worse.
- 1-synod reuse adds extra difficulty on both the Mars and Earth sides: More demanding trajectories, heat shielding, and expedited processing times..

I agree we’ll hopefully get doubling each synod for a while, but I think it’ll asymptote dramatically after reaching 100 or so ships. In fact, the curve might be 1 starship, then 10, then 30, then 60, 100, and then constant at 100 new Mars Starships per year.

BUT you can’t just have exponential growth indefinitely without the demand coming from somewhere. Tesla has huge demand for cars for obvious reasons. Falcon & Starship have Starlink. But Starlink doesn’t need 1000 new starships per year!

I just don’t see where this is coming from. I don’t think the demand will be there WITHOUT making them reusable.

Where have I said indefinitely?

I said 10 synods, which is what it takes to go from 1 to 1000, or 2 to 2000 or what have you.
Some time within those 20 years, ships will evolve.  They'll become cheaper (per colonist) and probably larger.
This tilts the scales even further towards non-reuse, since now 10 reuses give you even less, in light of the new technology.

Remember that 1000 ships is maybe 10,000 people (since there's a large cargo:people ratio) so it's not a giant number.
The evolution towards newer and larger ships will allow the exponent to continue, not by ship count, but by colonist count.

And sure, at some point things will level off.  But I don't see the argument for reuse still.
What may happen at a far-future date is dedicated inter-planetary vehicles, using advanced high-ISP propulsion, and planet-local ships for ascent and descent.  These ships will be highly reusable, but that's not what we're talking about in this thread.

[Robotbeat]

You can do 10,000 settlers with just 10 Starships per year. Cargo can be stacked in containers with Starships pushing 10 container loads at a time. Or just pushed on its trajectory in free space.

You’re basically just assuming the money for exponentially more ships comes from nowhere. But you don’t get exponential growth without massive capital expenditures. It’s far better to invest in your colony than in building massive, massive factories on Earth greater than the combined output of all commercial airliners in the world just for like 10,000 passengers per synod.