Consumerandcommercialelectronicdevicesin 2053 arelikelytolookandperformalmostexactlyliketheydotoday.
CorrespondentMark G. recentlyproposedaninterestinganalogybetweenthetechnologicalproductcycleofcommercialaircraftandconsumerelectronics. Aerospace technology experienced a Golden Age of rapid technological development that leveled off once fundamental technologies had matured. Investment in further advances reached a point of diminishing return: the cost of squeezing out modest gains exceeded the profit potential of the advances.
Silicondaddy: Moore’sLawabouttoberepealed, butdon’tblamephysics is a significant article. The principal expert quoted, Robert Colwell, is one of the true Kelly Johnsons of the golden age of semiconductors. I date this Golden Age period from 1975 to 2015. And like KellyJohnson (Lockheed’s P-38 to SR-71 wunderkind), Colwell was active over an entire 40-year Golden Age. He was already at Bell Labs in 1980 and he’s still at DARPA today.
I thinkthissemiconductor – aerospaceanalogyisverystrong. Look at what happened from 1930 to 1970 in aerospace engineering and manufacturing. We went from fabric covered crates flown by barnstorming Great Waldo Peppers to XB-70 Mach 3 bombers (production cancelled) and Boeing 747s. By 1970 a series of real physical limits were reached in an array of basic aerospace technologies, all nearly simultaneously. Come 1971 the US Congress deleted funding for the US supersonic transport (SST). The British-French Concorde and the Soviet Tu-144 went ahead. Both were commercial failures and developmental dead ends.
The result is that 44 years later Boeing is still building 747s (and 737s). Boeing undoubtedly now has numerous retirees whose entire careers were spent in the 737 or 747 programs. This is an outcome very few people would have predicted in 1968 when these two programs were beginning.
It’s ironic that Alvin Toffler published FutureShock the same year that phenomenon generally stopped in aerospace. Subsequent margin tweaking in commercial aerospace engineering has focused on three functional areas: safety, manufacturing cost and operational cost. On the other hand, continued attempts to increase military aircraft performance helped lead to the runaway program costs we’re still witnessing (for example, the $1 trillion F-35 Lightning program).
This will be replicated in military electronics once the limit of Moore’s Law is reached. Further increases in capability will start costing more, not less. This will lead to reduced total production runs and further cost increases from loss of scale.
Theabovesuggeststhatconsumerandcommercialelectronicdevicesin 2053 arelikelytolookandperformalmostexactlyliketheydotoday.Subsequent improvements will be aimed at reducing costs, not enhancing raw performance.
It doesn’t mean the world will stay in stasis. Look at the changes the ever rising population of 747s and 737s helped usher in between 1970 – 2013.
Thankyou, Mark, for a provocativeanalysisofexponentialtrends. As Robert Cowell noted in the above article, "Let’s at least face the fact that Moore’s Law is an exponential, and there cannot be an exponential that doesn’t end," he said. "You can’t have it."
It’s not just processor speeds that reflect exponential trends; cost declines in electronics (think digital memory) tend to follow near-exponential rates, too. Once the fundamental technology matures, those price declines flatten out.
AsMarknoted at theendofhiscommentary, theeconomicandsocialchangesfrommaturetechnologiesarisefromubiquityratherthanadditionalcapabilities.The revolution in commercial aerospace was not technological improvements in speed (such as the SST), it was the price reduction in the cost to passengers and the ubiquity of commercial aircraft and routes.
We can expect the same trajectory of change in consumer electronics: it will be ubiquity that creates change, rather than technological leaps in capabilities.