How are more precise things created from less precise things...

How are more precise things created from less precise things? How did we calibrate more and more precise technology by using things that were less precise?

Other urls found in this thread:

lichess.org/pHJHk3PI
youtube.com/watch?v=ZMByI4s-D-Y
youtube.com/watch?v=pirlP48EajI
twitter.com/SFWRedditImages

It all breaks down to sharpening a knife with a rock. By chiseling away the bad parts you get something more precise

not that difficult.

all it takes is for the more accurate machinery to be built accurately enough that it does the job.

so a set of calipers that measure 0.001mm, can be made on a machine that itself only has 0.1mm tolerances.

there's always a degree of limitation - you cant build a laser inferometer for measuring gravitational waves with a handsaw and a tape measure. but you can build the machines that will assemble the machines that will.

But after a certain point you cant sharpen the knife further you get distortions due to errors in our hand movement and brain. How does the story not just end there? In other words, what happens after this bottle neck?

Yes, but that is what my question is getting. Suppose you built a more precise machine by randomly hitting rocks together but its not perfect. There is a limit to the precision. Wouldn't the imperfections in calibration carry on to the next machine that is built like a genetic disease? E.G: Wood carved by the knife wont be smooth due to the imperfect blade.

as an Example
consider using a Pantagraph to scale down a drawing

if your initial prototype is the enlarged version
then the scaled output will have any errors also scaled down - an increase in precision

It all comes down to measuring time more accurately. A cesium oscillator not only oscillates very quickly and stabily, but each oscillation has both a very specific time and physical size. By creating more and more accurate measurements of time we are also creating more accurate measurements of size. The more accurate your measurements of size the tighter your tolerances can be.
.
You can't make something more precisely until you can measure it against the tolerances it was intended to meet.

This a damn good analogy. Thank you. I can relate this to a similar concept: The reason why audio is amplified before it is sent to the computer is because interference of the universe contaminates the signal. But the distortions are constant so amplifying it early on while the signal is pure results in a much cleaner output.

only a simple example
I'm sure there are other analogues

but it should at least prove the point
that precision is not limited by your starting tolerances, it can be scaled down

But using this argument leads to a looping dead end. We would have to have a precise way to measure the time from a cesium atom in order to have a standard of time to compare against. How would you build that measuring machine?

You see, it leads to an impossible chicken or egg problem.

Newton's aproximation

Exploit geometric identities, fixed physical pheomena,
as reference points

>Suppose you built a more precise machine by randomly hitting rocks together but its not perfect. There is a limit to the precision.

.... but you dont build a fucking machine by "hitting rocks together"


Let me give an example.

I start with a tree, and some chunks of iron.

I stick a piece of wood on a bench, and put a spike of iron it it, which lets me rotate things. marking out, it makes a circular shape, with some wobbling. with it, I make a "round enough" pulley wheel.

using the pulleys, I can attach a second pivot point, the wheel rotates smoothly, I put a spade-bit point, attach the pulley to a pole. I now have a pole-lathe, and it lets me make an axle that is fairly accurate.

using the axle, I make a lathe headstock, which is supported. now, the pulley drives through the headstock, and its more accurate still. Now, I go from wood to iron, and I can use files and cutters and I make the headstock for another more accurate, iron lathe.

this accurate iron lathe lets me make another lathe, which this time has bearings in it, made from ball-races turned on the last one. this lets me get a very smooth rotation.

this next one is accurate enough that I can now link the drive wheel through gears, which lets me have a screwcutting lathe attachment. Now I can make accurate threads.

the screwcutting lathe threads mean I now have accurate control on a lathe bed. I can now make controlled, accurate, round parts, and the screwfeeds allow control of depth exactly.

I have gone from chunk of tree, to accurate lathe.

all machines are that iteration of progression.

I noticed a flaw in this argument just now actually. For a panto-graph to work we need a lens. But if that lens was not carved perfectly the focused image will be distorted and imprecise so this actually wouldn't work?

lichess.org/pHJHk3PI

case in point:
Crystals

I can suck my own dick.

your pantagraph is different from mine
I don't see any lens

just a bunch of mechanical linkages, and analog device that doesn't need to itself be make with much precision, just with minimal play

even with play present, you can operate the linkages unidirectionally to enforce consistency in any lashing errors

This actually doesn't work either:

E.G: To build a near perfect circle from scratch you can use a compass with an angle but without measuring the angles precisely you cannot build a perfect circle, also hand movement precision is also a contaminating factor. Symmetric mathematical attributes only describe symmetries and properties, in other words relative measurements, they don't teach you how to build perfection from scratch.

You see? It seem everywhere you go you hit a dead end.

I'm sure that needs micrometer-fine precision.

So crystals are our only reference, we cant go beyond that? Even then if the crystal isn't perfect and contains contaminants, as found in nature, it will be distorted. The way i see it, it is impossible.

Incorrect, no looping because our ability to measure time and therefore mechanical devices far exceeds our ability to produce mechanical devices.
.
In fact our ability to measure time is now so accurate that we can measure devices made with a single string of atoms and thus the science of nanotechnology was born.

>Even then if the crystal isn't perfect and contains contaminants, as found in nature, it will be distorted.

in which case you make an artificial crystal, to be more accurate.

This is a good answer, I guess in other words. complexity can be achieve from imperfection at the expense of entropy.

I guess its that process of the refinement that we keep rediscovering because we neglect it and loose it through our limited lifespans and war. This is an issue, no one in society cares about the process but only the outcome.

I like this. Are you an Engineer or Computer Scientist by any chance?

dafuq you talking about nigger?
you don't need to have a precise angle measurement to make a perfect circle

in my HS geometry class we were taught to do shit with pure analog compasses, no protractor, no angle measure on it

a stake in the ground with a fixed length rope gets you a circle, no angles measured

move the stake to the edge of that circle and trace another one overlapping
now go to where their circumferences intersect, place your stake and do a 3rd circle
O Shit! we just invented a reverence point for 60degree angles
with no initial angle measuring device, just by exploiting geometric identities
now we have a standard reference point to create other angle measurement tools

You hit yet another dead end right there. In order to make precise artificial crystal you need to purify the components and environment that it grows in which requires precise tools to create.

Interesting question.
Mostly craftsmanship

we can measure far less than that.


the LIGO gravitational inferometer, which measures gravity waves consists of a pair of two-and-a-half-mile (4-km) long L-shaped tunnels, along which laser light split into two beams that travel back and forth down the arms. The laser beams are used to monitor the distance between mirrors precisely positioned at the ends of the arms. Due to gravitational waves, the distance between the mirrors will change by an insanely tiny amount.

A change in the lengths of the arms smaller than one-ten-thousandth the diameter of a proton (10-19 meter) can be detected.

thats about a million times smaller than the length of an atom.

That compass you used was made with fairly tight tolerances. You need to have a good compass.

You cant just two blunt sticks of wood with a rope. For instance, If the tip is not a cone and the rope is not perfectly tight, the circle will not be perfect and the arc's end's wont meet. How would you build that compass? The rabbit hole goes deeper than you think my friend.

>dead end right there. In order to make precise artificial crystal you need to purify the components and environment that it grows in which requires precise tools to create.

bullshit
the 'precision' of the environment and the 'purity' of the extraction are unrelated
the crystal doesn't give a fuck how many micrometers accurate the growing chamber is milled to, only that it's keeping the outside air out.

Combine it with the earlier Pantagraph example
an imprecise compass on a large scale
can be used to generate a more precise measurement on a small scale

pick up a damn mechanical engineering history textbook, this shit's basic
not sure if trolling or just stupid

Crystal oscillation has long since been abandoned in favor of wavelengths of various types of radiation.
.
Additionally to restate something I've already said, so long as our ability to measure time greatly exceeds our ability to produce mechanical devices the "can't do this without that" argument is void.

youtube.com/watch?v=ZMByI4s-D-Y

>Minecraft

precisely

youtube.com/watch?v=pirlP48EajI

Bravo! Fuck I love science.

ohh snap

in a way. yes.

you cannot start out with a tree trunk and some nails and make a pocketwatch. but you (or more likely, your son) could end up making a pocketwatch, on the lathe that is built, starting with a tree and some nails.

same goes for milling, to make the machinery that can be used to create laser measuring tools, or all sorts of mental shit like that.

its exactly what people did.

>pick up a damn mechanical engineering history textbook, this shit's basic
No it is not, yes the concepts of basic physics is there but nowhere in mech-e books do they explain how to create more precision from scratch. They mostly just show the timeline of events and highlight the inventors not show how the process was perfected.

>not sure if trolling or just stupid
No I am not trolling and I am saddened by the fact that you would think this and call me out as stupid when I am trying to learn.But I bet you don't really care do you.

Holy shit, this is actually a good "for your thoughts" question.
Proud of you Sup Forums, this isn't some copypasta bait shit for once

>I like this. Are you an Engineer or Computer Scientist by any chance?

Historian and craftsman. I use historical methods for experimental archaeology. not quite engineering, but similar skills

So yes, I have made a pole-lathe, a water-wheel powered grindstone, and similar machinery of that technology level.

You underestimate human accuracy..
You can get tighter tolerances by hand than with most modern machinery with simple tricks..
Google scraping by hand

Interesting. I wonder how many generations it would take to reach the level of technological progress and precision achieved today if humanity started from scratch and only focused this recursive on refinement of tools.

>stabily

At this point, it comes down to whether Moore's Law will continue to be true, seeing that computers are basically the foundation of every industry at this point

God is the answer to everything my child

>God is the answer to everything my child

who is responsible for pedophiles and cancer?

God

some other religions god like Islam

Well, I already fucking told you how, in several posts before as well

sorry you were too stupid to absorb the concept
I'm done trying to help you out
maybe I cared at first but goddamn you're so obtuse so No, now I don't really care anymore.