Program a strange, corrupted computer and discover its secrets

Imagine that you find a mysterious old computer, perhaps in a basement or a garage. When you boot it up, the "Tesselated Intelligence System"—or TIS-100—displays a copyright dating back to 1972 and an error message: "CORRUPTED SEGMENTS DETECTED, INITIALIZING DEBUGGER."

You're taken to a self-test diagnostic tool, where you have to create new programs and rewrite the corrupted code in the machine, with a little help from a photocopied computer manual straight out the 1980s.

This is TIS-100, which bills itself as "the assembly language puzzle game that nobody asked for." Created by Zachtronics, the company best known for Spacechem and Infiniminer (aka the precursor to Minecraft), it caters to some highly specific tastes involving nostalgia and coding.

Screen Shot 2015-06-09 at 12.50.59 PM

In order to debug the code, you're given a grid of twelve programming nodes and tasked with taking certain numerical inputs and turning them into certain numerical outputs, using the language spelled out in the manual.

Within each puzzle there's a debug button you can push, which doesn't seem to debug anything but instead displays strange, disjointed messages, either from the former owner or the creator of the machine itself. Who made the TIS-100, and why? What is it really for? The game suggests the answers will be revealed after you finish debugging, although I can't say for sure because it's super hard and I haven't beaten it.

Like a lot of coding problems, there's more than one way to solve each of the puzzles, and already plenty of people online who want to share and debate their solutions. There's also a sandbox option where you can create your own TIS-100 puzzles, and use them to frustrate other players.


But be aware: This is most definitely a game for the programming and mathematically inclined—or at least people willing to work really hard and stretch their brains in that direction. Although I can fake my way through HTML and a little CSS, I'm not much of a programmer, and the learning curve for TIS-100 felt steep to me indeed.

If you're not sure whether or not the game is for you, try reading the "manual" for the TIS-100 that comes with the game as a PDF, or watching a tutorial on Youtube. If it seems fascinating or makes you nod your head in understanding, proceed with confidence. If you feel like you've been catapulted into one of those nightmares where you suddenly have a math test in a class you've never attended, perhaps another game would be a better fit.

TIS-100 is currently available on Steam Early Access for Windows, OSX and Linux, and a final version is expected in several months.

Notable Replies

  1. What happens if you tell it to
    jmp ffff:0000

  2. Oh please, oh please, oh please... after you solve the 99th puzzle, have the game link to a Rick Astley video.

  3. I actually just bit the bullet on this one last night.. it's only a couple of bucks, and it's been a long time since I've done much asm programming. I've made it through the first seven or so puzzles in a couple hours playtime, with most of that being attempts to figure out how to get better scores on the solutions I already came up with.

    The instruction set's pretty stripped-down (I keep wishing for a xor), but it's been interesting so far. It definitely requires a programmer's mind, but it's still basic enough that it probably doesn't require previous exposure to asm or other languages to dig in.

    Conceptually, it's really similar to Spacechem - it's a different mechanism and is all text, but it still comes down to dragging things around in loops to accomplish each specific puzzle's goal.

  4. One of my frustrations with SpaceChem has been that it would be so much simpler if it would just let me write code...

    Maybe this one will feel better, but your description makes me think it will be similar. That's totally fine, I can always go to Elevator Saga or WarriorJS, but I'll probably pick this up anyway.

  5. I may have given the wrong impression, though I'm not sure. The frustration I've felt both in SpaceChem and TIS-100 comes down mainly to how strictly constrained the workspace can sometimes be, so if that's what you mean then yes, they're a bit similar. It's a different kind of challenge because it's code you're writing... but there are still strict limits on the available tools for the job, and the puzzle of it (at least, as far as I've gone so far) revolves around forming an algorithm that will take specific set of inputs and transform them in the right way for output.

    SpaceChem did have some interesting twists on what you were doing for the goal in later puzzles, I don't know yet whether this one has anything similar.

Continue the discussion

15 more replies