New Zachtronics Puzzler TIS-100 Out Of Early Access

What even is a computer?

SpaceChem and Infinifactory creator Zach Barth has released his latest thing-making puzzle game, which sits somewhere between fiddling with chemistry and building automated factories. TIS-100 [official site] is an assembly programming puzzler, having you literally learn and write code to fix up corrupted code in the mysterious eponymous ’80s computer. Yes, you do need to learn and write the TIS-100’s assembly code. Computers are puzzles!

After a seven-week stretch in Steam Early Access, TIS-100 properly launched yesterday.

The final major update added the TIS-NET DIRECTORY, a series of 25 player-made puzzles which reveal more of the strange system and brings the game to 45 head-scratchers in all. Puzzles have leaderboards so you can compare your coding efficiency with your pals, and the game has sandbox modes so you can make whatever you want. Yes, you can make games within the game.

The fictional TIS-100 has its own manual to pore over, though someone has helpfully made a text version so you can see what you’ll be getting into if you play it. I had a quick glimpse, flashed back to programming lessons, and closed it very quickly. But you might dig that.

The official launch has also brought the game to more stores. TIS-100 is out for Windows, Mac, and Linux on GOG, Steam, and the Humble Store for about £4.50.


  1. Awesomeclaw says:

    I picked this up straight away when I saw it had come out of early access. Not had a lot of time to play around with it so far but it seems pretty interesting.

    I do a fair amount of low level and assembly programming as part of my work so I’ve been able to pick up the concepts pretty quickly, I’d be interested to know how people with less experiments find this game.

    • Xerophyte says:

      My initial reaction to TIS-100 was “well, that sounds a lot like work” and as I usually code in something as high-level and abstract as C++ I don’t really come closer to the metal than than the odd vector intrinsic. Then the other nerds at work started competing for best program a couple of weeks ago and I felt I had to out-optimize them, as one does.

      It’s been more fun than I expected, the extreme (by modern standards) limitations on the programs in the game make it an interesting puzzle more than work. It definitely feels like it has to be incredibly difficult to get anywhere if you don’t have a programming background, though.

      • Awesomeclaw says:

        The way that it involves moving values around nodes really makes it feel like an assembly programming conversion of the spacechem/infinifactory gameplay model. It’s interesting that Zach has been able to take the same basic concepts and apply them so well in three completely different environments.

      • Geebs says:

        Yeah, I only got into hobby coding a few years ago when smartphones started to be a thing (so, Obj-C and OpenGL) and I figured they were simple enough for me to be able to get my head around. Programming environments and new languages have advanced so much since the BASIC days and that huge gap that used to exist between “hello world” and making a program that actually does something has finally been mitigated.

        The point of this rant is: in this day and age, programming assembler for fun has very selective appeal and I’m not sure I believe that even reasonably nerdy types want to inflict that on themselves. If you really want to drive yourself insane trying to deal with multi-dimensional arrays mapped into two-dimensional interleaved memory buffers, no decent debuggers, wildly inaccurate documentation and a huge variety of “undefined behaviours” which mimic the expected behaviours except when they don’t – well, OpenGL is right there waiting for you!

        • phlebas says:

          Are we talking about holes in the docs or explicitly undefined behaviour? If the latter, I’ll be disappointed if the puzzles require it – undefined behaviour is generally so for a reason, and subject to unpredictable change.
          (yup, compiler engineer here. How’d you guess?)

  2. Eleven says:

    There’s a point I tend to reach with Zachtronics games where the amount of planning and lateral thinking required to complete the next level is comparable to my day job, and I start guiltily wondering if I should doing something productive instead.

    I love Zach Barth’s games, and I’m totally up for playing this one, but with each passing game I reach that guilty moment a little earlier…

    • Dare_Wreck says:

      On the other hand, I’m a programmer and find his games absolutely dreadful – I hate taking my work home with me. Looking over that text manual, I am so absolutely uncompelled to give this a try.

  3. User100 says:

    Probably the best game EVER, to play at work!

  4. Gotem says:

    This makes me wonder if my HC11 board still works… first I would have to figure out where I put it.

  5. Gnoupi says:

    I’m interested by this, because while I liked much Spacechem, I was always a bit frustrated to have to deal with “spatial constraints” (and I imagine infinifactory is the same, with another dimension). Not having to make object go through a path to encounter the instructions might make it better for me.

    • phlebas says:

      Agreed – I quite enjoyed Spacechem but the constraints felt rather artificial, as though I was trying to find the right way around a convoluted system rather than engineer a solution myself. Maybe I just didn’t stick with it long enough to find the flow, but I’m still looking for something I’d like as much as Manufactoria.

    • sadwatertunnell says:

      I actually found infinifactory to be the least ‘spatial constraint-ey’ of the three games. With both spacechem and tis-100 i had a lot more experiences of having to rebuild a large portion of my solutions because i couldn’t manage to fit them into the space given. In infinifactory there is almost always enough room to build anything in any way you want, and you are never really asked to fit objects through a small gap / other strict spatial manipulation.

  6. bonuswavepilot says:

    A great game, though slightly weaker because Zach didn’t accept the level I made… :)

    Apparently it was ‘too math-heavy’… If you’re game to try it yourself, can be found hither:
    link to
    (I managed a score of 3757/11/90)

    @Gnoupi – I found it easier than Spacechem (well, I still have a couple of levels to go, but I think I can see how I need to do them). But then, I’m a coder, but don’t necessarily have great spacial skills, so maybe that’s it.

    Incidentally, Alice (that sounds like a song) – I hope you read the PDF of the manual, rather than that helpful person’s txt version, because they put a bit of effort into making it very much reminiscent of an old-school programming manual, complete with having been photocopied badly so that you can see staples, and text faintly coming through in places from the page behind, etc.

    It would definitely be easier for folks with a programming background, but I don’t think it is beyond a non-coder who just sees it as an interesting puzzle. It would certainly take longer that way though – I found a few times in reading wikis & fora that I had independently discovered most of the optimisations folks had come up with just out of necessity.