5
$\begingroup$

I'm building a world with medieval-era technology, in which there is a basic type of magic, where magic users can, rather than just waving their hands and doing anything they please, can instead control physical energies (like kinetic, heat, electrical, chemical etc.).

In this world, a lot of potential energy can be stored in a reasonably small object (I won't go into numbers, for the sake of the question let's just say "a sufficiently large amount") that can essentially be a battery to power magic if one can control it. Then I realized, it could also possibly be used to power machinery... like robots.

Now, assuming the robots can be programmed to control how to use this energy for things like movement etc. a simple constructed body (humanoid looking stone/wood/metal construct built with hinges instead of joints for example) would be able to be powered for a limited amount of time. Then, if the "battery" is replaced it can go on indefinitely as long as the body is properly maintained.

As the robots can manipulate the stored energy in their "batteries" for their own operating power, let's assume that they get their sensory inputs by detecting energy from light/sound etc. so that they can see and hear things.

However, it is the programming that I am struggling to find a reasonable explanation for. The closest thing I can think of from other fantasy worlds that operate without modern technology and programming techniques are steampunk-type robots, but I'm not sure if they are just given commands via hand-waving magic.

Is there any way to give commands to this type of robot without using hand-waving magic? Or could there be something that would act as a reasonable substitute for the internal machinery of a robot allowing it to "think"? I did think of stuffing the brain of a human/animal inside the robot and having the energy from the battery power it, but if there is a less grotesque solution that would be much better.


As an explanation for why a world with this magic is still medieval era tech, let's assume that the monarchy controls all of the wizarding population, and only allows them to build what the King requires, like an unquestioning robot army that will protect him and his family.


EDIT: I've just realized that I entirely forgot to mention that the robots don't need to nearly be as sophisticated as humans. It was pointed out that I'm essentially describing a Golem, but I'm hoping for a creation with a little more ability for free thought and reasoning, like to recognize friend from foe/ intruder from guest rather than the more literal commands of attack anything that approaches.

Also, if you can find a way to use this magic to advance the tech a little to a point that you might need for this creation, that is reasonable (you do have an army of wizards at your disposal after all).

$\endgroup$
8
  • 6
    $\begingroup$ Is there a difference between your described robot and a golem? Golems are typically only explainable by magic. $\endgroup$
    – Frostfyre
    Apr 22, 2015 at 12:43
  • 1
    $\begingroup$ @Frostfyre Yeah, I wanted something that is more able to take initiative in executing commands. I feel a golem would only take literal commands, whereas I'm looking for a creation that can have some semblance of free thought, but without the capacity to question the orders... initially. $\endgroup$ Apr 22, 2015 at 13:26
  • $\begingroup$ Sounds like you want the dwemer automatons from Skyrim. $\endgroup$
    – James
    Apr 22, 2015 at 13:48
  • 1
    $\begingroup$ I think you may have to go with handwavium or your brain idea. Medieval tech just wasn't able to manage modern computer hardware...the storage space and processing power of the human brain is tough to match. $\endgroup$
    – James
    Apr 22, 2015 at 14:03
  • 1
    $\begingroup$ I know that this seems like an obvious answer, but, do it the same way you would without magic. $\endgroup$ Jun 8, 2016 at 17:46

7 Answers 7

3
$\begingroup$

Is there any way to give commands to this type of robot without using hand-waving magic?

Realistically, no. You've probably seen videos of computers or robots that have some capacity to interpret what people say. What they don't tell you is how much raw data is required to build the data model that allows the computers and robots to do that. For interpreting text (that's already in a nice easy-for-a-computer-to-use format) you'd want to have a dataset of thousands of texts. These aren't "See Spot Run" texts, these should be on average short novels. You'd need a very large group of people dedicated to transcribing data to produce a dataset that could begin to be used to teach these robots how to understand you.

Even with a dataset large enough, medieval technology would still fail at this - the math and physics known to them are not sufficient. The algorithms for interpreting text (let alone understanding spoken language) require quite a bit of math that would be well beyond the reach of a medieval wizard. To make it even worse, you don't just have to teach the computer how to understand the words that are spoken, you have to teach it how to hear. That requires a knowledge of how sound works and the types of sound waves that speech produces.

Even with an army of wizards at your disposal, I don't think you'd be able to overcome these problems. Each step of advancing knowledge of physics, math, and algorithms takes a lot of time because of how many possible dead ends there are (with some dead ends not being apparent for quite a while).

Given that we're not going to be able to do this without hand-waving, why not introduce handwavium crystals? These crystals resonate with magic, and under certain circumstances (perhaps magic used directly on them in a particular way) they can reproduce a series of magics used in their presence. Once placed inside the robots, they would work like an artificial brain - a wizard would use magic to direct the robot's body to do a certain task. Then the crystal is activated using the battery and will reproduce the same actions.

A new robot would be very dumb - it could only exactly mimic the actions that the wizard just caused it to do. Over time, the crystal would learn more subtlety and be able to handle situations with more versatility. This makes older robots (or at least their crystals) much more valuable.

$\endgroup$
7
$\begingroup$

Summon and bind an entity into the machine that's under a geis to follow instructions. It's the occult equivalent of AI. Your magic user simply tells it what to do, and/or can use wards and runes to program it. The Summoned Entity has to obey.
Like any heavy machinery you have to treat it with caution.

Edit:
Depending on what kind of entity your magic user summons you could have varying levels of intelligence. A SE with low level intelligence would be able to handle basic instructions ("Robot, fetch water", "Robot, sweep floor"), but you wouldn't want to trust it with complicated or creative tasks ("Robot, assemble IKEA bookshelf"). A SE with high intelligence would be able to handle greater responsibility and think outside of the box, but you'd potentially have to be more careful programming in safeguards to prevent rebellion, the magic equivalent of Sky-Net.

Otherwise, you could design a magic logic circuit using runes, glyphs, gems, metals, whatever. You could program it using the magic equivalent of punch cards, but I see it being very limited. At least at first. Possibly good for simple/repetitive tasks.

$\endgroup$
4
$\begingroup$

Computation is ubiquitous. Programming will be a natural thing to do if basic building blocks for computation is easily available as in your world. Look into finite state machines to get an idea of how computers could work mechanically.

$\endgroup$
2
$\begingroup$

I feel your objection to using a "golem" may be rooted in the standard D&D monster.

The traditional Golem from some Jewish lore has a trait that may be exactly what you are looking for.

A shem is created by writing various Hebrew runes on a paper, forming one of the Names of God. This was placed in the forehead or the mouth of the Golem, which activated it.

In recent fiction (Feet of Clay, a Terry Pratchett disworld novel), a Golem was created with a rather long document placed in it, giving it direction.

This can easily be used to fit your need.

The programming for your magic robot is written on a similar document, which is placed within the robot. The specificity required for this programming is entirely up to the amount your story needs.

$\endgroup$
1
$\begingroup$

Given the low technology and the type of magic, the robots will become more of an external extension of the user. By manipulating water flow (since they will not understand circuits) robots will be hydraulic machines that do hard labor that they might not be strong enough to do.

$\endgroup$
0
$\begingroup$

I'm not sure why you deride putting the spirit/brain of an animal into the robot as "handwavium". Given the limits of a magical world's tech, that would be the best way to do it! Whatever spirit motivates a dog, or bear, "knows" how to run a body already, so magically attaching it so that those same thoughts activate the golem's equivalent parts correctly would be as straightforward as magically making an articulated physical body the same way.

You might need a different, specialist magician to bind it for you, perhaps.

$\endgroup$
1
  • $\begingroup$ I'm not saying that doing a spirit binding is handwavium, I'm saying I was looking for a solution that isn't handwavium, and also isn't spirit binding, as I'm hoping there is a different way that would make more sense within my story. $\endgroup$ Apr 23, 2015 at 9:02
0
$\begingroup$

I have also magic robots in my world. I faced the same issue as you and this is how I solved it.

I call it an animatrix. It is a sort of artificial intelligence/ soul made with magic and bound to an object. In most cases this is a cube. When activated an animatrix will light up and project itself over the object it is bound to. The reason these objects are mostly cubes is because they fit well into other machinery and the projection of an animatrix is a cube of sorts as well. The projection starts off with a cube made up from a bunch of smaller cubes. Each cube has a purpose and can be 'opened', meaning the animatrix will project that cube instead of the whole. This cube can exist of smaller cubes and so on.

The animatrix is generally controlled by though, but this needs an interface though (something to put your hands on in general). The real 'programming' is found within the logical cubes and the spell circles that make animatrices. They can be used for multiple purposes. locking a gate (picklocking these gates looks a bit like hacking then), managing machinery, robots, ... .

I hope this was helpful to you.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .