A Moore's Law of Space

[pmcray]

One definition of Moore's Law is that the amount of computing power per constant currency unit doubles every Moore's Law period, where a Moore's Law period is typically taken to be between 18 and 24 months.

Now consider the cost in constant currency units per unit mass in Low Earth Orbit. Now does that figure in 2009 compare with the figure in 1957 or 1983? Has it declined over time? Has the rate of decline been constant? Can any kind of Moore's Law relationship be derived for launch costs? I presume the answer is no on the grounds that if one could, it would well known. Even if the Moore's Law period were 15 or 20 years, all you would have to do is wait long enough for the price to fall to a point where a rich enthusiast could fund a return to the Moon, an expedition to Mars. We might not get there quickly, but we would there get eventually.

Of course, there is no particular reasons to expect a Moore's Law to operate in launch costs. Moore's Law is not a law of nature, it is a law of marketing, and it is driven by the enormous demand for computer-based devices (i.e. devices that use semiconductor technology). There is no end, it seems, to the uses that such devices can be put to, while the uses of LEO are more limited. Furthermore, it happens that semiconductor technology exists in a physics domain where it has been possible to tweak continually production processes and get proportionally greater returns (a linear reduction in the wavelength of light used in the lithographic process will lead to a quadratic increase in the number of components than can be fitted in a given area - to first approximation). Rocketry (and most other technologies) just don't happen to be in that kind of domain (and, perhaps, as people always say, computers won't be for much longer). The development of most technologies is bursty with occasional breakthroughs interspersed with periods of incremental improvements. This seems to the case for rocketry. Which gets me back to favourite point: we should forget about space for the time being and concentrate relentlessly on the technologies that will create the breakthroughs that will allow cheap access to space: NMT and AGI.

Comments

[pmcray.livejournal.com]

I did some calculations once for the RAzor about how the cost of the energy needed to get into orbit. I think space enthusiasts should be focussing their efforts on getting the cost per kWh of electricity down by an order of magnitude (through, for instance, coating as much as possible with solar cells - these wouldn't have to be produced by true Drexlerian MNT, but you'd probably want some kind of smart and cheap molecular manufacturing process). That way mass drivers might be economics. (I don't know how practicable it is to recover the energy from descending payloads on a space elevator.)