Splitting hairs on definitions to deny progress and greater potential for exponential progress

Alex Knapp at Forbes talks about the example of the limits of airplane speed as an example of real world limits to exponential progress. The speed of rockets continue the speed curve slightly.

What made us really fall of the speed of vehicle improvement curve was abandoning the development of nuclear pulse propulsion. Nuclear bombs have worked since 1945. Project Orion could have technically been accomplished by 1970 and interstellar versions would have worked and still can work. Humanity could be traveling at 3.3% of the speed of light with these systems.

In the comments Alex talks about the Kurzweil’s performance predictions, check out the Age of Spiritual machines (published in 2000 and written in 1999). Ray predicted a teraflop processor in a desktop machine for $1000 in 2009. Alex says you couldn’t get a teraflop for $1000 in 2009 and you still can’t now.

In 2005, The xbox360 was capable of 1 Teraflop performance and the Playstation 3 of 1.8 Teraflops.

Alex counter that the better performance was for GPU (graphical processing units) and not CPUs.

Sorry teraflops for less than $1000 count even if they are GPUs.


Rhys Taylor has excellent photos from his videos

The BBC had a piece that showed some of the external pulse propulsion tests and the history of the project. As Arthur Clark says, “Chemicals are feeble compared to atomic bombs…project Orion is only way we are getting large payloads around the solar system even now”.

If you liked this article, please give it a quick review on ycombinator or StumbleUpon. Thanks