This is incorrect. Anyone who works in science and engineering would tell you iteration is often much more difficult than conceptual invention. It was much easier to invent the rocket than it was to build one that could get to the moon.
Chinese semiconductor industry
- Thread starter Hendrik_2000
- Start date
- Status
Its a different type of difficulty. Inventing the rocket is hard because you had to come up with the concept of a rocket. Which is difficult because you had to have thought about the concept from scratch without any guidelines or previous work. Iterration is also harhard, but you still have previous work and guardrails.
There’s a reason why actual scientists and engineers often dislike people who only spend time theorizing ideas and think that should qualify them for special acknowledgements. There’s a way average people like to fetishize grand novelty, but grand novelty isn’t the basis of good science and engineering.
Well, there you go. US bias hits again. It took decades to get from the concept to a working suborbital rocket but about decade to go from that to the moon. Konstantin Tsiolkovsky had the concept in Tsarist Russia but it took advances in Nazi Germany (V-2) and Soviet Union (R-7) to put someone into space. Konstantin Tsiolkovsky already had the concepts for a) chemical rockets b) propellants, namely liquid oxygen and liquid hydrogen c) multiple rocket staging to escape Earth's gravity. he even computed the mass ratios for staging at 10:1.
But you are correct that iteration can be damned hard. The F-1 engine used in the Saturn V moon rocket is one example. It had a single large chamber which was much cheaper to manufacture and more reliable. Soviet rockets either used multiple-chambers or used many smaller engines. Even the RD-170 is like that. You know why? Because of something called combustion instability. The Soviets never managed to solve it properly until really late in the game if ever. Destructive resonance in those combustion regimes caused vibrations which led to the explosion of the engine into tiny pieces. But they did come up with tricks the US didn't. Like channel wall nozzle design which is now a de facto standard in not only Russian but also US and Chinese rocket engines.
But you are correct that iteration can be damned hard. The F-1 engine used in the Saturn V moon rocket is one example. It had a single large chamber which was much cheaper to manufacture and more reliable. Soviet rockets either used multiple-chambers or used many smaller engines. Even the RD-170 is like that. You know why? Because of something called combustion instability. The Soviets never managed to solve it properly until really late in the game if ever. Destructive resonance in those combustion regimes caused vibrations which led to the explosion of the engine into tiny pieces. But they did come up with tricks the US didn't. Like channel wall nozzle design which is now a de facto standard in not only Russian but also US and Chinese rocket engines.
What’s wrong with using an example from the US? It’s not like Russia instantly went to the moon either. Would you preferred if I had gone from the V2 to Russia’s suborbital rockets instead? Because it’s the exact same point.
I'm not disagreeing that iteration is hard. I think China and Japan and SK are great or even better than US at iteration. I'm just saying that the US or Europe in general is able to come up with more original inventions than SK or Japan.
I acknowledge that going from 7 to 5Nm is hard even though it is an iteration. However, it is also hard or even harder to come up with something that replaces transistors. Where do you even begin?
Iteration in most cases is a problem with defined boundaries. Inventions are genrrally problems with undefined boundaries. That's why they are a different type of difficult if that makes sense.
Well, there is a reason why we still use R-7 rocket variants today to launch astronauts (plus satellites) into space while the Saturn V rocket has gone into the dustbin of history.
With regard to transistors, they are changing as we go to smaller geometries because of leakage. If walls of the transistor are too thin the electrons pass straight through them. That is why with 14nm most companies went with FinFET transistors. With smaller geometries there is talk of changing the transistor design again.
Both “inventions” and “iterations” follow the same boundaries of physics. “Iterations” are their own form of inventions. It was much easier making a solid state transistor than it was to shrink it down. It was much easier inventing a laser than making it compact, or increasing its output.
- Status