h a l f b a k e r yFewer ducks than estimates indicate.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
The challenge would be the mixing of materials. Most of it could be plastic, that's no problem, but you'd need conductive and motor elements and the hardest part of course, the heating elements.
Not saying I know how to do it, but that's why it could be an X Prize contest or something.
RepRap project
https://en.wikipedia.org/wiki/RepRap
[bs0u0155, Apr 07 2025]
How dates were picked before we automated these machines causing them to burn tons of coal for every date picked.
https://youtu.be/wB...si=RmuJpf8LPw4q5Ve1
[doctorremulac3, Apr 08 2025]
Poc's steampunk date picker.
https://www.dropbox...ep&st=v6ax8ej7&dl=0
With AI's addition to the idea of having the dates be massive. [doctorremulac3, Apr 08 2025]
masochistic computing companion
masochistic_20computing_20companion
[Voice, Apr 08 2025]
Please log in.
If you're not logged in,
you can see what this page
looks like, but you will
not be able to add anything.
Annotation:
|
| |
Heating elements are a piece of cake.
For example, Fisher & Paykel have been "writing" their dishwasher heating elements with conductive ink for a long time.
Because some of it needs to be metal, probably easier to make it ALL out of metal (I can't think of any parts that NEED to be plastic). Getting different mechanical properties from a single metal supply would be the tricky bit, but there is a lot of research into nano-structured materials with weird capabilities). |
|
| |
Exactly, it's got possibilities and roadblocks but obviously the payoff isn't just making 3D printers, it's making everything. |
|
| |
The problem is fundamental. To melt a material, you need something hotter than that temperature. If it's made of that material, it will melt. |
|
| |
So what’s the solution? Remember, the idea here is for a contest. Are there ways to form the necessary heating parts besides heat? |
|
| |
You can use magnetic fields to melt metal, but I don't know what temp the coils themselves get up to.
I'm thinking "3D printing" on it's own won't be enough; a fully replicating unit will need at least a bit of machining & other manufacturing capabilities; & almost certainly some way to assemble sub-components. |
|
| |
Can the device include a black-box unit that has a midget Human hidden inside? |
|
| |
Hush; I don't want them to know I'm here! |
|
| |
Gonna change this to "contest" on the off chance that the boners are judging this as a actual proposed design. |
|
| |
My bone is because you've added nothing but the word "contest" to something that has been widely discussed for decades. |
|
| |
So if something's been discussed you can't have a contest to achieve it? Might wanna let the X Prize folks know that. |
|
| |
But thanks for manning up and explaining your bone at least. |
|
| |
And after I was in a good mood and bunned all your ideas. Oh well. |
|
| |
The RepRap project <link> is clear precedent here. 3D Printing owes a lot to Adrian Bowyer. Opening up the tech has changed the world. |
|
| |
Nowadays, I can draw up my nephew's idea for a suspension component for his RC car, hit print and have the object in carbon fiber reinforced glory in an hour for single digit $ on a machine that's ~$1000. |
|
| |
Prusa is a major player and they 3D print many parts of their printers, but I think this highlights a fairly fundamental problem. A 3D printer will generate dimensional inaccuracies, making a printer with this will generate a machine with more, each generation gets worse. So the important parts are made with other techniques. Prusa is printing large amounts of printer parts because they can, not because they should. Injection molding would spit out parts in seconds not hours with far superior dimensional accuracy in better plastics. |
|
| |
My point is, do you want a self-printed printer given that you know it will be inferior? |
|
| |
You might argue that we have high precision machines now, yet we started with sticks and rocks. So it's clearly possible to make a more precise/accurate machine with a less precise/accurate machine. However, I think this might be fundamentally reliant on rotation. |
|
| |
Reading up on the invention/ development of the machine lathe is fascinating stuff |
|
| |
//My point is, do you want a self-printed printer given that you know it will be inferior?// |
|
| |
Well that's the interesting question isn't it? For the particular system, be it 3d printer or crop picking robots, at what point does the "Quantity has a quality all its own." axiom kick in? |
|
| |
So sticking with the farm analogy, if I've got robots that can make all the tools necessary to manage an agricultural venture and they just give me a menu: "Bring X amount of iron powder, X amount of plastic beads, X amount of lithium or whatever..." then after the growing season I sit back and watch the trucks arrive and get filled with my crops, am I particularly concerned that I'd sit there and watch a robot take 4 hours just to pick the dates off a particular date tree? |
|
| |
At what point is simply being autonomous good enough? |
|
| |
Dunno. That's why it's an interesting discussion. I think anyway. |
|
| |
//At what point is simply being autonomous good enough? // |
|
| |
Do you have a set of rules? I think I can demonstrate a self replicating, autonomous, 3D object that can run entirely on water and corn. It weighs less than 30g, and squeaks. The major problem seems to be that it is fixated with printing endless mice. |
|
| |
I'd say only that it can't be a mouse, either robotic or biologically produced. |
|
| |
Assuming it would be run with AI it might just say "Analyzing... answer, to get self replicating autonomous corn eating flying machines that will transform corn to fertilizer, simply throw corn onto your front lawn. Birds will eat the corn and reproduce." |
|
| |
"But I want robots to do it." |
|
| |
"Well... I dunno, because, they should be made of plastic and make beeping noises." |
|
| |
"Would you settle for having devices that attached to the various birds that made the desired beeping noise?" |
|
| |
//At what point is simply being autonomous good enough?// |
|
| |
Energy usage, [Doc]. Everything else derives from that. Its simple thermodynamics. Moss growing on a boulder, Human labour, robotics, huge industrial power plants, its all just capturing free energy from the environment and using it to develop and increase complex structures. |
|
| |
So an autonomous device that burned tons of fuel to accomplish the picking of one date is not going to compete with a trained monkey at the same task. |
|
| |
Who said they’d be powered by burning tons of fuel? |
|
| |
Robots well be doing everything in a few hundred years, including building the nuclear and solar power plants, mining the raw assembly materials etc. |
|
| |
Doubtful we’ll still be picking crops by hand in 1,000 years. Short of nuclear Armageddon, the possibility that smart automation and technology is suddenly going to stop evolving for some reason is probably a stretch. |
|
| |
As far as your calculation that to pick one date you’d need tons of coal, when and how will that happen? How much coal does it take to pick one date now? What technological advancement creates a situation where automation of a current date picking system increases the energy requirements by several billion percent or more? |
|
| |
That being said, you did bring up an interesting point. How much fuel does it take to pick a date currently? I'd guess maybe a drop or two of kerosene. Amortize the cost/amount of fuel needed to make the tree shaker and gathering truck then to transport it, maybe... a quarter cup? Dunno, just a guess but somebody could figure that out probably by working backwards by how much the equipment costs, how much the farm spends to get the date to market etc. Pretty sure it's not tons though. |
|
| |
ADDENDUM: Just made a rough guess of how much diesel fuel just to drive that date 500 miles from the farm to the store by truck. I came up with .00016 gallons. Checking my number against AI's estimate now. |
|
| |
Wow! AI said "0.0168 gal/lb × 0.015 lbs = 0.000252 gallons of fuel per date." Basically the same number I came up with but it added the fuel to grow it as well. Oops, okay, I was wildly off. To transport them ONLY 0.000031 gallons per date. Oh well. Anyway, having to feed and house the humans would add to that. Remove them and automate you're using a lot less fuel. |
|
| |
Re: link. Whoa B, that’s awesome. Better tell them about the megatons of coal they’ll burn for each printer built though. (Kidding) |
|
| |
I was imagining an enormous steampunk date picking device a hundred feet tall with 50 foot long unbalanced cast-iron rocker arms fuelled by a team of 20 firemen shovelling anthracite into a red-hot firetube boiler... |
|
| |
Anyway 0.000252 gallons of diesel fuel contains about 37kJ of energy. One date contains 98kJ. So as long as your picking machine uses less than 60kJ to pick one date, you are fine. (that's about 2g of coal) |
|
| |
More than that and you would be better off to cut out the middleman and just drink the diesel fuel and eat the coal. |
|
| |
Interesting solution, I do like the steampunk thing. Put your exact description into AI and it added the concept of having the dates being 300 pounds and about 5 feet tall. (link) |
|
| |
//A 3D printer will generate dimensional inaccuracies, making a printer with this will generate a machine with more, each generation gets worse// |
|
| |
Evolution says the ones that produce inferior copies won't reproduce long, so soon they'll all be able to reproduce. |
|
| |
//Doubtful we’ll still be picking crops by hand in 1,000 years// |
|
| |
I doubt we'll be doing it for most crops in 30 |
|
| |
AI control of accuracy would remedy any de-evolution of the quality. Simple numbers, the metrics of each unit would be fed into the upstream manufacturing process. "This part was too big, compensate." etc. |
|
| |
//I doubt we'll be doing it for most crops in 30// |
|
| |
Without the nuclear winter scenario, yea. Automation's evolution is nothing new, I just think the evolution curve is about two swing up wildly within our lifetimes. |
|
| |
The big problem will be how will power made people who love controlling over others get their purpose in life? Whipping robots? I don't think that'll do it for them. The drive to control other humans is gonna have to look for another "reason" than picking dates or cotton. |
|
| |
// The big problem will be how will power made people who love controlling over others get their purpose in life? Whipping robots? // |
|
| |
The LLMs disagree, but I say, why not? |
|
| |
Oops, power MAD not power made. |
|
| |
//people who love controlling over others get their purpose in life? Whipping robots?// |
|
| |
This suggests a lucrative market for a robot built to be excellent at suffering. That's dark, and probably why we will be overthrown. I for one, welcome our new robot overlords. |
|
| |
Between the race among the world's top militaries to create all powerful AIs, the race among three letter agencies to automate propaganda, and the race to make better UAVs and UGVs, there's at least one flavor of MAD to add to the current list. Nuclear, Biological, Chemical, and Digital |
|
| |
When there accumulates enough negatives, they will all be flipped on their heads and become positives. |
|
| |
It's just the way it works. Over and over again as far as I've been able to determine. You history buffs out there will know better than I do what I'm talking about as my own position has been constructed without knowing history... just the talking monkeys I've been subjected to. |
|
| |
Notice how fast the planetary political pendulum swings now. It's all part of the process. |
|
| |
We all determine where it stops... briefly. |
|
| |