There are a variety of switches and buttons, wire, NOR gates, and a small handful of mechanical devices, the most important of which are pistons that allow you to move blocks around.

Since you call me out, I need to point out that you are wrong. I do understand what it takes to do that, and that he needs to handle every bit and memory and all. You're passing a judgment based on a shallow interpretation of what others wrote.

And I said I am very very (doubly) impressed. When it comes to the freaking next project, branch off, do something else, don't constrain yourself in the Minecraft world. Instead, this kid should use his immense talent/patience/creativity that he/she has and applies to something more substantial. That's what I said. I don't understand why I got downvoted for saying that. You see it as a criticism; But what I gave was a constructive criticism. If you don't see it, pay more attention to the words "very very impressed" and especially the word "next".

No macros or compilers per se. There are map editors that allow you to make copies of existing groups of blocks, like replicating seven-segment displays.

There are not really any repositories of existing designs, so the logic was probably all done 'by hand'.

He probably either used Creative Mode (unlimited stuff), peaceful difficulty (no enemies) + a superflat world (don't have to mine things out of your way) or mcedit, which lets you edit map files out-of-game and place whole sections of stuff based on schematics.

Most likely he used an out-of-minecraft editor to put together logic gates that can be copy/pasted, and then larger structures. The map format for minecraft is pretty easy to edit, and there are CAD like programs that'll build out an environment you want.

There is a Lua-based scripting mod but I doubt it is commonly used for things like this. As another commenter mentioned the use of copy-and-paste mods is much more likely.

It limits nothing just because it's a game. In World War II a guy was held in solitary confinement (a small wooden box) for 2 years. In that time to keep himself sane he designed and built 100% in his mind ... a complete working clock despite never having built anything like a clock or watch.

After the war, he built that clock and it worked.

According to NASA the Minecraft world is pretty big:

http://apod.nasa.gov/apod/ap120312.html

[NB It's just to the left of Neptune]

Actually, the complexity of a modern CPU is mainly in making things run fast. Building an in-order, no-cache, single-issue ARM capable of running uClinux doesn't seem all that difficult. The cycle time would be horrible though. You might not live long enough to watch it boot.

I'll settle for OS/9.

If it would be simpler you would get something much more complex than this calculator and once again ask the same question.

brings a whole new understanding to "debugging"

true story. when I was 16 ( way back in the early 90s), my parents enrolled me in niit.com (presently a billion USD company that was at the time a tiny 3-room outfit ). niit taught cobol. I hated everything about cobol. I thought it was a shit language and the people who taught me cobol were shit. We were supposed to build an inventory control module in cobol. I didn't know what an inventory was, so they gave me an econ book about manufacturing & inventory. I thought it was absolutely dumb to keep track of nuts & bolts & accounting & money & suchlike. So I wrote a cobol program to calculate the fourier coefficients of complex exponentials, since that's what I was studying at school. So my cobol program would painstakingly calculate the first five fourier coefficients of sawtooth waves & square waves & then using the partial sum, it would reconstruct the expansion by printing out the series on a dot matrix printer. All of this was running on some processor called intel 80286 xt and it would take 10 full minutes to just compile the code. One day I was standing by the dot matrix when the instructor walked in. He thought I was some precocious kid who had coded up a whizbang inventory module for acme corportation, so he expected nice tables with rows & columns of data on optimal number of nuts & bolts. But when he saw the printer slowly rolling out sawtooth waves and square waves of various frequencies, he completely lost it & yelled at me for wasting the precious resources of the dot matrix printer to do frivolous nonsense. I got a F on cobol.

> I get that Minecraft is fun, but wouldn't it have been more practical, in the long run, to have built something equally cool in a real CAD/HDL/etc. software?

I find it all the more impressive because he used an environment that was not meant to be used like this.

That's like running the marathon with a handicap and still winning.

And if he can do it in minecraft I think that he'll take like a fish to water once he gets his hands on other, more powerful tools. That's mostly a matter of access, 16 year olds are more likely to have minecraft on the machines they have access to than CAD/HDL software and the hardware to go with it.

I don't think this is quite true. I see minecraft as a good stepping stone into more advanced things.

When I was his age, I was using the Lego Mindstorms heavily. At first I started with the built in GUI language, which was extremely limiting. After that I moved to progressively more advanced languages and IDEs until I was using Not-Quite-C, a version of C compiled for the Mindstorms.

These things tend to work as stepping stones. Had the kid started on something like CAD, he may have gotten stuck at an impasse that was too difficult for him, lost motivation and went back to playing Call of Duty.

Perhaps his next project will be in a "real" application, instead of Minecraft, because he has the motivation and knowledge to move forward.

CAD is expensive, or confusing, or both. It's not much fun. The kit needed to play with HDL is expensive.

Really, this guy just needs a bit of sponsorship and guidance. Some tech company should send him "stuff"[1] and they'd earn some nice publicity.

[1] Development kits or test equipment or robots or whatever. And good books for it all. And maybe a gentle syllabus.

You are thinking about products and projects, the kid is thinking about having fun in a game. Totally different perspective.

It’s totally fun to just have fun without necessarily learning any important skills or developing a product that helps you make money.

Don't look down on a project just because it is done with low-level tools. Building a scientific calculator from logic gates may not put you on the track for a career in web programming, but it's still an impressive display of what can be made into a very useful set of skills.