The mechanism
An orbital computing proposal must pay for spacecraft, payloads, launch, ground infrastructure, operations, replacements and lost work. Solar energy availability can change operating inputs but does not erase those costs. Suncatcher’s launch-price analysis is explicitly conditional and incomplete as a total economic study; its projected future launch input must not be presented as a current quote. A fair terrestrial comparison uses the same accepted workload, quality and service targets. Space-generated data reduction and ground-user interactive services can produce very different communication costs. A useful scenario identifies the assumptions that allow its budget to close and any unresolved constraint that prevents a useful service.
- Choose a concrete workload and matching terrestrial service baseline.
- Close energy, thermal and link budgets before assigning sustained useful throughput.
- Include complete initial and recurring costs with a consistent lifetime and accounting boundary.
- Apply availability and accepted-result criteria without double-counting utilization already in measured throughput.
- Vary uncertain launch, mass, lifetime and service assumptions and identify where the preferred option changes.
Science and assumptions
Launch-only annualization
Launch-only annualized cost per payload kilowatt is launch price per kilogram times mass per kilowatt divided by useful years.
Symbols and units
- C_launch
- Launch-only cost annualized per payload kilowatt
currency/(kW·year)
- p_launch
- Launch price per delivered kilogram under the selected scope
currency/kg
- m
- Launched mass per payload kilowatt
kg/kW
- L
- Useful operating life for the annualization
years
plaunch is currency/kg, m kg per payload kW and L useful years; result is currency/(payload kW·year).
Assumptions Only launch amortization; excludes spacecraft, ground, operations, financing, failures and other lifecycle costs.
Engineering example
Synthetic $500/kg, 20 kg per payload kW and five-year life yields $2,000/(payload kW·year) for launch alone. Doubling mass doubles that component; no token rate or total ownership cost follows from it.
Distinctions and tradeoffs
Comparing launch-only annualized cost to a terrestrial electricity bill is not a full system-cost comparison.
An attractive launch assumption can be outweighed by ground-data traffic, short lifetime, low availability or expensive thermal support.
AI workload fit determines useful results; no universal dollar-per-kW or MW-to-token factor can compare all terrestrial and orbital systems.
Specifications and interfaces
- Economic boundary
- Complete lifecycle per accepted result
Launch-only cost is a component, not total cost of ownership.
- Scenario status
- Conditional assumptions and sensitivity
Future prices and fleet-scale availability require explicit uncertainty.
Interfaces
- Physical and operational budgets Input
- Mass, power, thermal, links, lifetime and service quality
- Conditional service economics Output
- Cost per accepted result and unresolved bottlenecks
Check your understanding
Which denominator makes an orbital and terrestrial comparison meaningful?
- Only nameplate electrical power
- Only peak accelerator FLOPS
- Equivalent accepted workload results under matched conditions
Answer and reasoning
Equivalent accepted workload results under matched conditions
Useful service depends on quality, latency and availability, so identical power or peak arithmetic need not deliver identical outcomes.
Evidence and scope
The example is synthetic. The source’s mid-2030s launch-price projection is forecast evidence, not a current market price or a completed feasibility demonstration.
No current calculator directly models this lesson’s main mechanism. Use the library for related bounds while retaining their stated omissions.