It seems that you're using an outdated browser. Some things may not work as they should (or don't work at all).
We suggest you upgrade newer and better browser like: Chrome, Firefox, Internet Explorer or Opera

×
The manual is only 14 pages long and seems to just.... end... almost like there was more to it. Anyone have the Steam version of this game that can confirm if that's correct (as some sort of the puzzle is that you only get part of the manual) or did GoG forget to scan the whole manual into the PDF file?
Confirmed - the manual is only 14 pages. Like SpaceChem, TIS-100 gives only the basic tools and you need to work out how to use them to solve various problems yourself. I believe you can click on the red "blocked" nodes to get other info as you play.
The red block is mostly storyline information. The manual does tell you what some of the function keys do, like F1 brings in the list of instructions you can use, which is actually very minimal.

Looking at this, i'd almost have preferred the 6502 instruction set, a little bigger, a little more generic, and you can actually do stuff with it... But perhaps the largest limitation is the memory, which gives you maybe 16 instructions you can fit on the machine.

All in all the chips are basically 4-bit processors with a 12bit acc/bak/bus for transferring data. A little surreal what you can do with so few instructions... Then again, BrainFuck sorta shows that off too :P
Attachments:
avatar
pavig: Confirmed - the manual is only 14 pages. Like SpaceChem, TIS-100 gives only the basic tools and you need to work out how to use them to solve various problems yourself. I believe you can click on the red "blocked" nodes to get other info as you play.
THANK YOU!!!

This looks very interesting to me, think I'm gonna like it, even though I'm no programmer and am completely "tech challenged" I'm giving this one a go.
Good luck! I'm actually quite interested in what someone who isn't a programmer and as you put it "is completely tech challenged" might think about it.

Do tell us how it's going after you've given it a shot
Post edited July 22, 2015 by wTwT
For some reason this thing reminds me of transputer design. Very cool toy. The instruction set is this limited for reason - even adding one scratch register would make it considerably less challenging.
The actual design can actually be built in metal, it would be a viable but primitive computer.
Bump
avatar
rtcvb32: A little surreal what you can do with so few instructions... Then again, BrainFuck sorta shows that off too :P
You don’t need a large instruction set for a Turing complete language. The execution time, however... Brainfuck is called a Turing tarpit for a very good reason ;)
Post edited November 18, 2017 by 4-vektor
avatar
OldFatGuy: did GoG forget to scan the whole manual into the PDF file?
No, the look of the manual is a deliberate design choice by Zachtronics. It’s supposed to look like a scanned paper document, reminiscent of the old paper manuals from the 1980’s/1990’s. The look is part of the visual storytelling that fits the narrative in the game.

It immediately gave me flashbacks to the good old C=64 programming handbook ;)

And the manual is indeed complete.

You only have the registers:
ACC, BAK, NIL

The ports:
LEFT, RIGHT, UP, DOWN, ANY, LAST

Comment, label, breakpoint:
#, :, !

Arithmetic instructions:
ADD, SUB, NEG, NOP

Register and port related instructions
MOV, SWP, SAV

Jump instructions:
JMP, JEZ, JNZ, JGZ, JLZ, JRO

And the visualization module command sequences


Being a code golfer and a fan of esoteric programming languages, I can see that they are even being generous with the instruction set. I mean, who needs SUB if you have ADD and NEG, for example? ;)
avatar
4-vektor: Being a code golfer and a fan of esoteric programming languages, I can see that they are even being generous with the instruction set. I mean, who needs SUB if you have ADD and NEG, for example? ;)
With a limit of 12-16 instructions per chip, every instruction counts. to do subtraction without sub you'd have to load the instruction, neg, add, neg again... that's a 3:1 net loss. True if you add sufficiently large numbers and use wrap around that works, but I don't think it wraps around the numbers in TIS-100, as it's effectively 12bits for the math/registers.

Still, as an experiment I created a superset instruction base where you had a few instructions and then built your own instructions in order to emulate the 6502 chip in a way of microcode. Not sure if it's good enough yet though.