# Can you “glitch” physics?

I recently came up with an interesting idea for a power: glitch power!

I originally thought the idea was great until my brother brought up a valid point: what is there to glitch? Normally glitches are coding errors, but this power is supposed to work in the real world, so there is no code. At best this power would just become an illusion power with a cool aesthetic. The best solution I could conjure was that he would be glitching physics..

Let me be more clear:

The power itself is an uncontrollable power which can act as a catalyst. At random moments the character can cause events which seemingly defy all logic. This, I thought, would work OK with the real world.

At this I was presented an interesting question: Are the laws of physics similar enough to game code that someone could exploit their flaws and/or break them without exploding the universe (shall we say "crashing the game")

• In the title (and tags) you ask about "physics", then in the question body you ask about "logrc" (which I take to mean logic). Those two are not the same at all! Which one are you interested in? – a CVn Sep 2 '17 at 11:06
• Also keep in mind that "the laws of physics" (as the term is commonly used) are really just descriptions or models of how the world (and the universe) behaves. The world will behave however it is going to, regardless of what our models of it say; usually when that happens, massive amounts of headscratching ensues until someone comes up with a way that explains not only the new behavior, but also doesn't fail to explain any previously observed behavior of the world. Every once in a short while, that boils down to "there was an error in the experiment" or "the results were misinterpreted". – a CVn Sep 2 '17 at 11:07
• While I know it's not the real definition, I only mean logic to be the expected result of an action(ram someone with your shoulder, expect them to move backward; with the glitch power they fly fifty feet in the wrong direction). Also, if the universe behaves how it wants to, that could be what the power is exploiting. I mentioned a similar thing in the final paragraph. – BlippThePanda Sep 2 '17 at 11:20
• If there's a glitch in the model, the model is wrong, not physics. It just means we need a better model. Going by your metaphor, the "laws of physics", such as they are, are the result of hundreds of people trying different things in order to hack the system response of the game. "Ok, so if I do X, the game does Y, consistently". If, at some point, it doesn't do Y, that means the player assessment of the game is wrong; not that there is a bug in the code; it's all closed source from our end, anyway, and despite claims to the contrary, we do not have the included documentation. – nzaman Sep 2 '17 at 13:03
• Quarantine by Greg Egan might be good to read for ideas. – Schwern Sep 3 '17 at 1:53

## 20 Answers

You asked this question: "Are the laws of physics similar enough to game code that someone could exploit their flaws and/or break them without exploding the universe (shall we say "crashing the game")"

The answer is absolutely and categorically no. The laws of physics are not at all like game code. By the laws of physics this means the nature of physical reality. There is nothing to glitch. End of story.

However, if there was a more fundamental level to nature that was response for making physical reality real and the laws of physics what they are, then, purely hypothetically, that might be glitchable. Except, that really doesn't make sense.

Metaphorically this would be the equivalent of marionettes, that is string puppets, being able to control their own puppeteers and make the other parts of the puppet play they are in that is reality dance to their tune. Again, this doesn't make sense.

Is physics capable of being glitched? Nix. Nyet. Nein. Non. No way. No.

• Until somebody does come up with a glitch, that is. ;) – NVZ Sep 2 '17 at 19:48
• I know it isn't, but quantum tunneling feels glitchy to me. – tox123 Sep 3 '17 at 1:04
• @NVZ There's always someone or something that wants to throw a spanner in the works. Right now there is probably a paper in Nature or Physical Review about a genuine glitch in physics and my answer is down the drain. – a4android Sep 3 '17 at 5:03
• To be fair, a lot of our technology is basically exploiting glitches in our previous understanding of physics. Show an electrical engineer a circuit diagram with two diodes back to back, in the 50s, and they'll tell you it's nonsense. But now the transistor is the foundation of modern electronics. – Draconis Sep 3 '17 at 18:11
• What if it turned out the universe was a simulation? Wouldn't it then stand to reason that "glitching" it might indeed be possible? – Ajedi32 Sep 4 '17 at 0:00

The closest you could get is collapsing the quantum superposition of states into the least likely one. Or something like a non quantum size tunnel effect.

Normally if you run into a wall you smash your head onto it. By having quantum effects scaled up, it could happen that you tunnel through the wall. This would make all convicts in jails, or bank robbers outside the bank, really happy!

Another example could be by entropy inversion: some scent dispersed in a room which spontaneously fills its bottle, or air molecules which coherently hitting a body and make it move up a stair. Such events are extremely unlikely according to our present knowledge of physics.

They are all not glitches, in the sense that is not a malfunction of the law, but it is just an odd consequence of the law.

• If these instances are not breaking the laws of physics they do not answer my question. My question is about relating physics to game code, not explaining the power. However, I will give you props for the interesting concept. – BlippThePanda Sep 2 '17 at 11:08
• This turns out to be physically impossible, the laws of probability are solid. There are some combinations of perfume in air that are about to go back into the bottle, however almost all combinations drift around. The uncertainty comes from you not knowing where each molecule is. – Donald Hobson Sep 2 '17 at 16:53
• @DonaldHobson, since the "full bottle" is a possible configuration of the molecules distribution, it's not impossible, but only highly improbable or it can happen once every several universe lifes. Practically speaking it's the same, but yet there is a subtle difference. – L.Dutch - Reinstate Monica Sep 3 '17 at 4:02
• So, what this answer need to work is an infinite improbability drive. – Arturo Torres Sánchez Sep 3 '17 at 4:18
• Exactly what I thought. In the end a glitch is nothing more than unexpected behavior. Messing with reality at the quantum level sure would lead to something like a "glitch" at the non-quantum level. – r41n Sep 5 '17 at 14:46

It all depends on what you mean by "glitch".

If you mean an error of programming then the answer is, as others have rightly said, a resounding "NO".

Reason for this is that "this specific programming" is reality, so any (possible??) "error" in it is a "law of Nature" and not a mistake (yes sir: this is not a bug, it's a feature!).

If you look at it in this way then the specific programming of any (even the buggiest) game is what actually defines that world, so "glitch" assumes a different meaning: an unexpected result of some action (i.e.: experiment) obeying to "laws" not yet discovered.

In this meaning history of Science is full of "strange glitches", from the famous Michelson-Morley experiment to quantum effects that allow your PC to work.

What would be different in finding a wormhole connecting two different points in space (unlikely, but not ruled out by current scientific knowledge) and your "using a plate to penetrate a wall"?

Only difference is in our expectations: we know our reality still has undiscovered aspects while you pretend to know what's "right" in a game.

• Program can't understand that it is wrong and that's the case. – rus9384 Sep 9 '17 at 0:14
• @rus9384:sorry, I fail to understand your comment; can you elaborate, please? – ZioByte Sep 9 '17 at 8:50
• Well, if you want a program to do action A, and it does action B instead, the problem is in program, but how can program correct itself if it has no such function? – rus9384 Sep 10 '17 at 14:20
• @rus9384: I might have not been clear enough. What I'm trying to say is: if we take the "programming" as "gold standard" then there never is something like "a glitch"; what you call "a glitch" is merely some law of nature that has not yet been discovered. Being able to walk through a wall in a game may seem "obviously wrong", but I can assure you that Lorentz contraction seemed very wrong at the time and Einstein never accepted quantum mechanics. In a certain setup "a glitch" can (and perhaps should) be interpreted as a hint to explore to extend knowledge in some new direction. – ZioByte Sep 10 '17 at 14:32
• This is all relative. Whether program has a glitch is determined by it's users/creators, but not by a program itself. This is almost equal to your words, but generalized to the whole programming. – rus9384 Sep 10 '17 at 14:36

# Possible? Yes. Probable? No.

What are "the laws of nature"?

In short the laws of nature are...

Descriptions of how reality has behaved so far

Let us take a simple law of nature: the theory of gravity. What does this law tell us? Well in short it says:

"Everyone that has taken a long walk on a short tall cliff has gone SPLAT at the foot of the cliff"

That is all it says really, along with some nicely detailed formula on how to calculate at what velocity everyone up until that point made that big SPLAT at the foot of the cliff.

We as humans make one critical but unproven assumption about the laws of nature:

The laws of nature will not change in the future

This is an assumption we make. Can we know with certainty that the Law of Nature that is Theory of Gravity will not change in the future? No, we cannot. Gravity might very well change in a year from now... tomorrow... or in ten seconds after you finish reading this sentence. There is no way we can know that for certain, until we get to those ten seconds into the future.

However — hand on your heart — you are not very willing to challenge that law, now are you? You are not willing to bet your life that you can take that long walk off of the short, tall cliff and remain suspended in mid-air because gravity just then decided to stop working.

# What governs the laws of nature?

The funny part is that the laws of nature are governed by other laws of nature. and those laws of nature are in turn governed by yet more laws of nature. And finally we reach a point where we have to say "Ok, we do not know the laws of nature that governs that particular law. So this last law we know, we will consider immutable.

...for now".

The "for now" is due to that fact that sometimes we do discover new laws of nature. This in turn means that some of the old laws of nature are amended (never changed, only amended). So in the case of gravity, this was amended as follows:

"Everyone that has taken a long walk on a short tall cliff has gone SPLAT at the foot of the cliff... unless they had been using parachutes."

So will it be possible to "glitch" the laws of nature in the future? We have — after all — "glitched" a number of laws already, such as splitting the atom when previously the law was that atoms are not possible to split.

Well, maybe. But it is not very probable. If you find a way, congratulations: Nobel Prize for you.

• This was a long way of saying, "glitches are laws we don't understand yet," but it was well said. Cheers. – JBH Sep 2 '17 at 16:23
• "We as humans make one critical but unproven assumption about the laws of nature: The laws of nature will not change in the future" Actually, I believe that a part of the definition of "laws of nature" is that they are invariant. So it's not an assumption as much as a prerequisite. – a CVn Sep 2 '17 at 19:32
• Actually, as far as I remember, there was some research that stated that the universal constants have changed by a significant [for those constants] amount since the start of the universe. The assumption is they continue to change and at some point our universe will get out of the V Goldilocks zone where our kind of life is possible – Gnudiff Sep 2 '17 at 21:35
• @MichaelKjörling No it is just a hopeful assumption. A fairly solid one because by all estimates the laws of nature appear to be holding steady. But we cannot see into the future. We can only extrapolate from the laws of nature and present conditions. – MichaelK Sep 2 '17 at 21:38

What you mean by "laws of physics"? Nature does not know "laws of physics", "laws of nature", or whatever you may call it.

"Laws" as you mentioned, are just phrases formulated by scientists, who observe certain numbers of phenomena and notice a pattern happening. The phrase itself is just an approximation of how nature behaves. There's always an exception in human-defined "law".

We call it a anomaly.

something that deviates from what is standard, normal, or expected (from "human law")

It's not an exception to mother nature. It's how it's always been. Nothing excepted because it's always been like that.

Is water expands when it's chilled from 4 degrees Celcius a anomaly? To mother nature, no. It's always been like that. It is, however, an anomaly to what scientists have observed from all other liquid, which contract when they are chilled. Might we found other liquid that behaves anomalistic? Maybe. Maybe we'll write a new law that covers the behavior of this anomalistic liquid.

So, there are no glitches or anomalies, or anything else that suggests you can break "laws". It's always been like that, and you just discovered how it works differently than the law(s) you know.

• In spaceflight, I believe it's called a "funny". As in, "that's funny, it's never done that before...". Which typically are the words surrounding great scientific discoveries; more so than "eureka". – a CVn Sep 2 '17 at 19:33
• @MichaelKjörling: Yup. Some specific examples: The discovery of cosmic microwave background radiation. At first it just seemed like noise in their receiver, which wasn't even meant to be listening to the sky. They were trying to receive signals bounced off balloon satellites, and they thought it was a problem with their instruments. Electricity, X-Rays and radiation were also mysteries when first discovered. – Peter Cordes Sep 2 '17 at 20:56
• @MichaelKjörling In spaceflight, those words also surround a death/horrible accident/big explosion – tox123 Sep 3 '17 at 19:55
• @tox123 Yes. IIRC it was Gene Kranz in an interview who specifically referred to the Apollo 13 oxygen tank explosion and subsequent weirdness (before they figured out what had happened) as a "funny". – a CVn Sep 3 '17 at 19:58

You seem to not actually understand what a 'glitch' is.

A 'glitch' (more commonly known as a 'bug') does not defy the code that runs the game, it exists because the code running the game is somehow causing that behaviour.

For example, the infamous Pac Man kill screen is caused by a block of code that does not properly account for 'arithmetic overflow'.*

This is true of almost all bugs: the bug exists because the code dictates that the bug will occur. The main exception to this is hardware-level bugs, where the issue is a hardware fault.

If you were to try to make the analogy between physics and program code, for there to be 'glitches' in physics, they would have to be there because the laws of physics themselves contain the glitches, not because your character is capable of introducing them.

Anyone with a copy of pacman has the ability to initiate the kill screen by completing level 255 (though not everyone is aware of the bug and not everyone has the skill required to do so).

Ultimately though, the laws of physics are not governed by software code. Admitedly nobody can say they aren't because nobody actually fully understands how physics work, and all humanity can say is that they appear to be governed by mathematical equations. (Or, more strictly, physics is full of behaviours and relationships that can be modelled by the human concept of mathematics.)

Whether you can get away with saying that the physics of your world are governed by some sort of 'code' or not is a different matter. If you declared your world to be a computer simulation it would be a lot harder to question it though.

All that being said to directly answer your question:

Are the laws of physics similar enough to game code that someone could exploit their flaws and/or break them without exploding the universe?

The answer to that is that it's extrememly likely that they are not similar at all.

There is no reasonable/unquestionable way of doing this based on real-world observations of physics. You can explain it by saying that the world is a computer program written by some greater being (who naturally would make mistakes that introduce bugs into the program) and that everyone is an AI, but you cannot explain it without giving the world a different set of physics to those that govern our world.

* Computers work with fixed-size numbers that can only represent a limited range of values. When a fixed-size integer attempts to reach a value larger than what it can represent it 'overflows' and wraps around to the beginning. In the case of an 8-bit integer (a byte) where the minimum is 0 and the maximum is 255, attempting to calculate 255 + 1 results in 0 because the byte overflows.

• "all humanity can say is that [the laws of physics] appear to be governed by mathematical equations" Um, no. We use mathematical equations to describe and model the world. The world, however, didn't suddenly start behaving differently when Einstein thought of the special relativity colinear velocity addition formula as it differs from that in Galilean relativity (see the linked Wikipedia article), for example; we just got a better model. Later work is even more detailed, but the world still behaves the same. – a CVn Sep 2 '17 at 20:18
• @MichaelKjörling I never said that the equations governing physics were dictated by humans. The behaviour of physics follows some equation, all that happened there was that someone got the equation wrong. We can never truly know what the equations are, we can just make a best guess that seems to fit the observed behaviour. Remember that maths is just a way of describing theoretical relationships. When I say 'there are equations governing physics' what I mean is that 'physics is full of relationships that can be modelled by equations'. – Pharap Sep 2 '17 at 20:24
• "physics is full of relationships that can be modelled by equations" Now that one, I can accept. You just might want to clarify that in the answer itself as well. – a CVn Sep 2 '17 at 20:26
• That's certainly good enough for me. Thank you for making the answer just that little bit better. – a CVn Sep 2 '17 at 20:32
• Congratulations for discussing what glitches would be in physics and how unlike game code it is. A well argued case. I enjoyed reading it. Plus one for well done. – a4android Sep 4 '17 at 6:30

You write in your question

Normally glitches are coding errors, but this power is supposed to work in the real world, so there is no code.

As @Christmas Snow has pointed out, however, this is does not need to be true. See Wikipedia's article on the simulation hypothesis. In this case, as has been discussed (see especially @ZioByte), what we think of as real-world physics would be whatever is coded up in the universe-game's physics engine. Hence, a glitch in the sense of "unexpected or non-physical behavior due to a logical error in the code" is very arguably impossible.

The phrase "logical error" was used just now. A logical error reflects a failure by the programmer to impliment an algorithm in the way that he or she intended. I want to emphasize that logical errors are not the only type of error that a program may encounter. See, for example, the stdexcept C++ header.

An overflow error is a typical runtime error that will not generally cause a crash. Modern programs normally use 64-bit chunks of memory to store integers, which means that they can represent unsigned integers as large as $2^{64}-1$ or 18,446,744,073,709,551,615. If you try to calculate $(2^{64}-1)+1$, then the result depends on which computer you are using. The easiest thing for your computer's ALU to do is to give a result of 0. So, say that the universe-simulation tracks the number of rain drops in a storm using an unsigned integer. If it gets just a little bit too rainy, and if the programmer forgot to check for overflows in this particular scenario, then suddenly the number of raindrops is zero.

The raindrop example is not that great for a number of reasons, the most obvious being that physics as we now understand it does not directly tell us how raindrops behave: our best fundamental descriptions of the real world come through quantum field theories, such as quantum electrodynamics. Basically, physics tells us how stuff works on the subatomic level, and the interactions of a huge number of these subatomic particles (or fields...) is supposed to give rise to the behavior of the macroscopic objects that we observe in everyday life.

Of course, if the computer responsible for simulating the universe does not have infinite computing power, it may well take a few shortcuts. The most advanced modern simulations of stellar evolution do not track every single particle in a star. They track a few macroscopic quantities like pressure and temperature. These quantities change according to differential equations derived from the laws of thermodynamics, which are in turn derived from the most fundamental laws of physics. Assumptions and approximations are made at every step along the way, but a good simulation will consistently turn out stars that grow and die much as we observe. So, it is not ridiculous to think that macroscopic quantities, such as raindrop count, might be directly accounted for in the universe-simulation.

This opens the door to talking about the single biggest difference between the real world, as we currently understand it, and every existing game's physics engine: real spacetime is continuous, but game spacetime is not. Perfectly simulating continuous space and time would require infinite computing power.

Consider that many physics engines use Euler's method to solve the standard Newtonian equations of motion for the trajectory of an object, where time is the independent variable. As the step size approaches zero, a better approximation of the continuous-time universe is achieved, but the computing power required to complete the simulation is inversely proportional to the step size. This forces a discretization of time in the simulated universe. Combine this feature of simulations with an algorithm that uses an adaptive step size, and you can get all sorts of unpredictable weird-looking effects showing up here and there, if the programmer was not extremely careful.

Think about a satellite in orbit around Earth. After each time step, the size of the orbit will increase very slightly, due to time discretization. Maybe the programmer anticipated this effect and wrote in a piece of code to occasionally bump the satellite back into the correct orbit. If the adaptive step size algorithm and the orbital correction algorithm are not talking to each other, then you might imagine that the step size could be greatly increased at the same time that the bump occurs. The effect of the bump would be magnified far beyond what was intended, sending the satellite plummeting down to Earth.

As discussed in the answer written by @L.Dutch, such an effect might not be considered a true glitch, since time in this simulated universe really is discrete, even if the programmer wanted to make it look macroscopically as though time were continuous. If I were programming the universe-simulation, however, I would log this satellite-crashing behavior in a bug database and try to modify the code such that the behavior would be eliminated in the next release. Whether or not you consider the behavior a glitch is a valid metaphysical or linguistic question, but the answer would not change my behavior as the programmer.

For an alternative metaphysical/linguistic interpretation of "glitch", see the answer by @Pharap.

• Space (lengths/positions) may in fact be quantized, not continuous, at length scales below the Planck length. The Planck length is believed to be the shortest meaningful length, the limiting distance below which the very notions of space and length cease to exist. Related: Planck time is the time for a photon in a vacuum to travel a Planck length, but IDK if anyone thinks time is really quantized. If not, a perfect simulation would still require infinite computing power. – Peter Cordes Sep 3 '17 at 21:48
• If I was going to write a simulation of humans living in a universe, I'd make it "lazy": only simulate in enough detail to make observations by conscious observers consistent with the physical laws. Far-away suns can be simulated very roughly using just enough parameters for their stellar evolution to be consistent with a higher-accuracy simulation. (This assumes the simulation mostly cares about the conscious observers, rather than the whole universe.) – Peter Cordes Sep 3 '17 at 22:01
• I recall an old SF short story where the world really did get more complex as human science advanced (and figured out the limitations of the simulation they were living in). Maybe with resetting the simulation every time, since I remember something about the characters ending up in what was implied to be the Garden of Eden at the end. – Peter Cordes Sep 3 '17 at 22:03

Absolutely, physics can glitch, and it does regularly. But it's not of any use.

The reason there are "glitches" in games, is that there are different sets of rules that make up a game, and they conflict.

1. The rules, as defined by the running code, as interpreted by the running CPU - load a memory location into a register, add 1, store it in another location, and so forth.
2. the rules, as defined by some arbitrary "canonical" code, or by any code released by some (arbitrary) "canonical authority", as interpreted by the CPU
3. The rules, as defined by , as interpreted by a correctly functioning CPU.
4. our (social) understanding of what "the game is supposed to do". When you press the jump button, Mario should go up a certain distance, then fall down until either there is something solid under him (when he stops), OR we have fallen off the screen (when he dies). We find this out by by playing the game and seeing what stimuli provoke what response.
5. The rules as set down in the instruction manual.
6. how the author intended "the game" to behave. I can make a guess at what Shigeru Miyamoto might have intended for Mario, I do not claim the same for this universe. Note that this includes cheat codes intentionally left in.
7. how the author intended us to understand "the game" - similar to to the previous, but excludes cheat codes, even if intentionally left in.
8. Our moral code applied to our understanding of the game. "in-game murder is imoral", "bots are cheating".
9. The way things are perceived for an in-world character - what does Mario experience/remember when he dies/restart? Does he know about the load/save menus?

Reality and games always play by the very first set of rules. Note that, by this definition, everything is allowed - patches/mods/cheat systems redefines the rules, the player is bound by the (new) rules. And, if your CPU crashes, that's part of the rules too.

We think we are playing by the "what we think the game is supposed to do" set of rules. When the rules we play be (#1 by definition) and the rules we consider the "true rules" conflict (Mario walks on empty air), we call this a glitch.

These two both have counterparts in the real world, and sometimes they do, indeed, contradict each other.

The first (equivalent of running code on a running cpu) is how the universe behaves around us.

We observe how the universe behaves, and formulate "the laws of physics". Newton's law of universal gravitation states that "a particle attracts every other particle in the universe using a force that is directly proportional to the product of their masses and inversely proportional to the square of the distance between their centers."

However, the universe does not care about our rules, and blindly went on following "the code". The result was a glitch - photons, which do not have mass, continued to be affected by gravity, in contradiction to "the rule". (I am not a physicist, and the details are probably wrong. If so, please treat it as allegory, metaphor, alliteration etc - "If we spirits have offended....")

At this point, we have two options (both in-game and real-world):

1. We can decide that our made-up rules are right, and reality/the code is wrong. This is a glitch, Mario/light is breaking "the rules", it's "magic", Mario/light should behave the way it's supposed to, you're not supposed to look behind the curtain/figure out what the rules really are.
2. We can decide that reality/the code is right, and our made up rules are wrong. We can use this to our advantage, if we understand how things work.

In-game, we tend to follow the first option (usually attributing it to the creator of the game, though that's not necessarily true).

In "real life", most people choose the second option.

Note: I started this posting with 2 options, and realized as I got further in that there were more and more alternatives. Some of these have real-world equivalents, some do not, and some are the domain of religion.

If the "Universe" we live in is a simulation (a serious theory proposed as being almost certain - see https://en.wikipedia.org/wiki/Simulation_hypothesis#Ancestor_simulation ) then most/all of the above rulesets should exist. This also implies a de-facto deity.

As an example, see https://www.youtube.com/watch?v=Ixu8tn__91E for someone who managed to use a glitch to modify a savegame, to add an memory editor. The rules, as defined by the running code, allowed changing of the rules.

This sounds a lot like Douglas Adams' HHGTG http://hitchhikers.wikia.com/wiki/Flying. Fall but get so distracted that you forget about falling.

(The joke being that you forgetting about physics turns into physics forgetting about you. Think about it too much and "physics will notice you". A lot like Wile E Coyote not falling until he looks down.)

This is not how nature works: it "pays attention" to everything in parallel all the time, not like a simulation on a CPU.

This is why physics simulations take so much CPU time: you have to calculate what happens to each atom, or rigid body, soap bubble, or whatever separately, but real physics just happens in parallel on everything consistently all the time.

Without adding something like magic (you might find magic in Terry Pratchett's Discord novels interesting, as uncontrolled magical discharge causes similar effects), or state something generally held to be true is not (say for example that quantum theory is completely wrong, then this might be possible), then no.

There are recent documentaries on youtube presenting the hypothesis that the universe is a computer simulation so detailed and so intelligent that it has created conscious artificial intelligence which includes us. Sort of matrix, but way more advanced. You can play on that theme with sufficiently convincing argument where this consciousness or a group of such manages to access this computer code and exploit glitches. Are biblical miracles the result of alien programmers attempting to interact with such consciousness? If our world is just computer information, and you can create information out of nothing, can you create "matter" in our universe, and free energy? If the "glitch" does exist and we find it, we can make what seems impossible to our ancestors, as we have become a "computer virus" created by that same computer.

Search youtube for "is the universe a simulation" and "the simulation hypothesis".

• However, since the simulation hypothesis cannot be disproved, it is unlikely to ever be accepted in scientific circles. Science is all about proof and disproof; if you can't disprove a hypothesis, it's not useful. – a CVn Sep 2 '17 at 19:35
• The simulation hypothesis has been an idea in philosophy for longer than YouTube has existed. :P "glitching" reality might be very strong evidence in favour of the simulation hypothesis, but more likely it would just mean that we'd discovered some new physics that would eventually explain a weird observation in a consistent way. (e.g. X-rays weren't understood when they were discovered, but weren't thought to be a glitch in reality.) – Peter Cordes Sep 2 '17 at 20:41

Well, as others have said, the actual laws of physics cannot be glitched ever. Our models of the laws might fail, but if it's allowed by the Universe then it's a part of the laws. So no, I don't think the actual laws of physics would support that.

However, I still think it's an interesting concept! Especially when one considers that our Universe might actually be simulated (I mean probably not, but it's an interesting thought experiment). What sorts of things might glitch out in such a Universe? A programming error might suddenly cause the character's position information to be overwritten allowing them to teleport instantaneously elsewhere. A known glitch in many programs that use GPS data is that rather than reporting Null for an unknown location, they report (0 degrees, 0 degrees), resulting in many people and objects appearing to be located at Null Island briefly until their actual position is located.

The exact state of a quantum system cannot be known until it is observed because it is in a superposition of states. If one were to wave their hands vigorously, this almost seems like it's because the Universe doesn't bother to simulate it until it absolutely has to. Why waste computational cycles needlessly? That's not the actual reason as far as we know, but it does sound interesting.

Current velocity, momentum, energy, and mass might "glitch out" allowing them to travel very quickly, or punch someone while their fist has the mass of a small star, or other oddities. Perhaps the Universe even "forgets" about them for a period of time rather than instantaneously overwriting one of their properties.

In summary, while I don't think it's exactly compatible with the real Universe, I think it would be pretty interesting to have a character who is able to ignore the laws of physics temporarily. Maybe they learn to control it? Maybe they eventually find out there are laws that govern how the glitch works? In that case, is it possible that their "glitch" isn't a glitch at all but is really written into the laws of the Universe for some reason? Whenever we have a scientific model in the real world that fails, the reason is always that the model is wrong, not the Universe. If it was written that the laws of physics in the character's Universe actually support them being able to do what they do, then I do think that would fit more closely with our real Universe.

# What is a glitch?

In the general case, a "glitch" is a surprising (and often temporary) interaction of systems under unexpected circumstances. (This definition might be a bit broader than usually, as the term "glitch" is normally only associated with a program or hardware malfunction, but it should translate well unto other fields than computer technology.)

# So, what might constitute a "real-life glitch"?

This depends a lot on our perspective and understanding of reality, but basically: Two or more "laws of nature" can under specific circumstances interact in a way that is contrary to all reasonable expectations.

### Just a thought experiment:

Imagine you have a wall with 2 openings in front of another continuous wall, and you kick footballs against the wall. Some of them will hit the wall, and bounce back. Others will hit one of the openings and hit the wall behind, hitting the wall in 2 regions corresponding to the 2 openings.

The systems in this case are "movement of matter" and "collision with a static object", with expected interactions between the footballs and the walls. No surprises this far.

But now, let's miniaturize our setup: Instead of footballs, we're now shooting electron at a double slit. Suddenly, we get a completely different hit pattern on the wall behind the double slit: Instead of 2 initial hit regions, we get an interference pattern! (But only when we don't measure which slit each electron travels through!)

This is/was surely surprising, as a simple understanding of the systems involved would let one expect the same outcome as in the football example.

Of course, nowadays scientists have a lot of fancy theories to explain it all, but how is this not a glitch according to my definition above? The devil is in the detail: Are those circumstances "unexpected" for the "laws of nature" involved or is it just our lack of understanding as we have no absolute definition of those laws, just theories based upon observations?

Of course, deviations between expectations and reality have historically often been the trigger for new discoveries and better understanding of our world. But the point is: We don't actually know whether our understanding of our world, our so-called physics, are the actual laws governing our world. So it isn't strange to find edge cases in our understanding of the world where our expectations simply fail.

### Back to reality

So, glitches are "defects" in the interactions among the actual laws of nature where those surprising outcomes might or might not be covered by our existing theories (as we derive our theories from observations, and those observations would include that glitchy behavior in cases where we got the "unexpected circumstances" right).

So, in the case that a glitch is covered by our scientific theories, your power wouldn't actually do anything really unexpected for us. Of course, there might be subtle aftereffects (see butterfly effect), but generally speaking, that power could as well be non-existent.

However, in cases where a glitch is not (yet) covered by our scientific theories, your power might result in unexplained events. And while these will not break reality per se (as they work alongside the real laws of nature), they should be really surprising for our scientist.

However, in those cases you are left to your fantasy on the question "what might those unexplained events be?", as everything unexplained by our current theories could still be a valid consequence of the actual laws of our world.

Looking at other answers, I am confused about the question. Do you want to glitch physics in the game, or in reality, or glitch physics in the game in such a way that it could be done in reality? Do you want to know how to do it in a game? Do you want to have unusual responses to the controls, so the resulting actions are still law-abiding, just unexpected, or do you want the actions to respond to the controls appropriately, just in an exaggerated or diminished fashion?

Every game has 'constants' in it, that are used in algorithms. These constants, for instance, determine the level of play, or the speed of characters, or the level of intelligence. Inserting some random number function (that is itself triggered by a randomized 'do/not do' routine) to randomly alter them at some random point during game play would certainly look like a glitch, but would not change the overall coding of the game. For instance, something might fall slower or faster, fly farther, hit harder, have more or less strength, based on the value of a variable constant. These changes could be temporary, or lasting.

Also, every game has selection structures, where a value of choice determines what consequent code is executed (jump, run, duck, fall, etc, in response to what controller button is pressed). Putting in a random number function for the variable choice that is triggered at some random time would make the game appear to be glitchy and unpredictable, but the coding would actually be intact. Again, these changes could be temporary, or lasting.

You could also do this with the variable that determines what comes next. Usually, this variable is incremented, but if it randomly was randomized, it would certainly look like a glitch. (For instance, sent to a different room than expected, or go in a different direction than expected).

Since everything still follows rules (which can simulate physics), the integrity of the game is not altered, just the implementation.

In a way, it introduces the concept of unpredictability to the game. Things are going peachy-keen, you think you have it mastered, and then...

However, someone who plays the game long enough would be able to discover what glitches and what doesn't, so it would have coherence to them. Eventually this information would hit the internet blogs. It would just become another 'rule of the game'. Thus, the game designer would simply make the probability of the glitches in the random function even less probable, and thus more sporadic. Perhaps some players would never experience them.

In essence, this is what was built in to the 'poker-playing' algorithm that out-played world champion human poker players - the ability to bluff. 'The feat represents a leap forward in developing artificial intelligence that can learn with incomplete information' link - the ability to respond in a way that was not expected or predicted.

But if you want to know if such things are possible in the real world, I would posit that only the first one - being able to change the value of generally-accepted constants would even come close. For instance, a 'bubble' in which the gravitational constant were changed locally, so things fell slower, faster, or not at all. Or that the universal speed limit c was changed in a local bubble, so things could go faster than our expectation of c in that bubble. That is, you are not changing the rules (the math, the equations) that apply to gravity, just the value of the constant. e = mc^2 still applies, just a different c value.

Physics today can't rule this out, we just have no idea what the mechanisms for doing so would be, or even look like, let alone how to implement or manipulate them, to modify the constants. But just because we can't do it today, and have never experienced an event where it was done, with our limited knowledge, doesn't mean it can't be done. My guess? There is absolutely no reason why these constants could not be different in a black hole, or perhaps a worm hole, or even another universe, for instance. We just don't know, and therefore can't answer that question definitively.

TL;DR: Whether something is a glitch depends on the intention of the universe's designer, which may or may not be known by the characters in your universe.

Wikipedia describes a computer glitch as: "the failure of a system... to complete its functions or to perform them properly." Implicit in this definition is that fact that the software behaves in a way that was not intended by its creator – in a way that is considered improper. If your characters are ignorant of the intentions of the universe's creator (or your universe has no creator), then there is no way to define a glitch because the intent of the creator is unknown. Any glitch would be seen as part of the rules of the universe.

If you were PacMan, Level 256 would simply be a physical limit that you can't overcome. You might even see it as a feature intended by the creator of the universe. It is PacMan's end of time. Often glitches in games are later seen as features. For example, straferunning in Doom is based on exploitation of a bug in the game, but was later considered by players to be an intrinsic part of how the game is played ( http://doom.wikia.com/wiki/Straferunning ). This Wikia page claims (without a source, mind you), that straferunning was "unanticipated in the design of the original levels."

Depending on how you imagine our universe's creator, many aspects of our physical world can appear to be glitches. For instance, who would have thought that if you concentrate a certain component of an element left over from a supernova (uranium), you can make a chain reaction that will produce another rare substance. If you put this rare substance at the center of a sphere and time some explosions just right, you can destroy a whole city. If we suppose that our universe's creator is kind and merciful, it sure sounds like the Manhattan Project was exploiting some unintended behavior of physics – a glitch.

For something to "glitch", i.e. to act in a wrong way, the right way has to be clearly defined. In physics, the universe is postulated to always act in the right way, and artificial laws are invented to describe it. When a contradiction arises, the laws are always considered to be at fault and modified, amended or discarded as appropriate. Of course, if every feature is considered to be "by design", it's simply impossible to have a glitch.

If we decide on a "perfect" set of physical laws, then it's indeed possible to have all sorts of violations. Radioactive decay doesn't respect the mass conservation law, objects which receive enough impulse to travel faster than the speed of light under Newtonian laws don't go as fast as they should be going, etc. If we choose to ignore the 20-th century physics, we might as well call these things "glitches".

To glitch physics doesn't depend on the glitch but on the universe.

The movie "The Matrix" is the prime example. It looks like the real world but once you know it's actually a program simulation, you can change the code to do things normal people couldn't

Are you familiar with the Infinite Improbability Drive in the Hitchhiker's guide to the Galaxy? Your idea of using glitches in reality reminded me a lot to that. Basically, an event might be very very improbable, but your superhero could have the (uncontrollable?) "gift" to make it happen.

Physically possible? No way. But very few superheroes would exist if they had to respect all laws of physics. In fact, that is the reason why superheroes do not exist in reality.

If you have at all studied chemistry, physics, astronomy, or other 'physical' sciences, you will note the proliferation of 'constants'. Almost every equation has one or more. These constants of proportion turn 'A is proportional to B' into 'A = B x (constant of proportionality).

These constants were defined at the time the universe was created (the Big Bang) and really, they could have been anything. (That they turned out to ALL be exactly what was necessary to produce an earth capable of life, instead of some other kind of universe, is quaint, and begs the issue of 'why?' but I digress)

Take c , for instance. A constant that defines the speed limit of the universe. But why exactly is the value of c what it is? Why not a few thousand metres per second faster or slower? Bear in mind, any such change in the value of c could be quite disruptive to the universe as we know it. Things just wouldn't work the way they do with 'our' particular value. But, there is no inherent reason in physics it to be exactly that value. It just is, so it are taken for granted that it must be. But there doesn't seem to be anything enforcing these constants to be or even remain what they are, except the will of the physicist to insist they must be a constant with a constant value. Otherwise, well, they just can't predict anything, and that would make them useless. (It would also destroy their argument against there being a god,that the world must be what it is, for all time, but again I digress.)

So, to 'glitch' physics, all you need to do is to change the value of a constant or two.

But that is a very big 'All you have to do is..."

In some of the more obscure explanations of what astronomers and cosmologists see when they look at the heavens, some posit that c is a constant across a region, but not necessarily constant between regions. Some areas of the universe might have a different value for c. Really, really far out on the extreme of thinking, extremely implausible, but not entirely impossible.

So, what happens if the value of the gravitational constant, G, is altered? The equations, the relationships, the theories, the calculations, the math, all stay the same, but the answers are different.

So this 'glitch' would simply be a change in the value of some constant.

But it would be result in the butterfly effect. One small change in one constant would so alter the universe that it would be hard to predict exactly what would result.

• Regarding the discussion on the speed of light, there are a lot of questions on our sister site Physics which deal with that. For example, What is so special about speed of light in vacuum? and The reference frame of c and Do we know why there is a speed limit in our universe?. I personally am particularly fond of JDługosz' answer to the latter. – a CVn Sep 2 '17 at 19:41
• @Pharap your edits are interesting. Thank you for noticing them. I haven't quite gotten used to the formatting standards of this site, and I am just learning how to format text. But I am still confused over how to format a variable, and how to format a constant. This site seems to have different editing preferences to physics. – Justin Thyme Sep 3 '17 at 14:04
• @JustinThyme - if you read the general FAQ for Stack Exchange, there is a superb section there on formatting that will tell you all you need to know: meta.stackexchange.com/questions/7931/… – Rory Alsop Sep 3 '17 at 15:43
• Nice idea about 'glitching" physics by changing constants. The change might need to be transient to be 'glitchy'. But super-powers need to be like that. The main drawback could be the unpredictably of the outcome. Plus one for a good answer that deserved more love. – a4android Sep 4 '17 at 6:34
• @a4android unpredictability is a drawback in a glitch? Bugs behave courteously? I have another answer posted herein where I explain how modifications to a constant in a program can be done using a randomization function. It would be more of a disrupting power than an advantageous power. I suspect the villain might use it rather than the hero. – Justin Thyme Sep 5 '17 at 1:50

The universe doesn't glitch, but in the context of very scant probabilities, bizarre things are possible.

This would be an interesting phenomenon in the context of a Rick-and-Morty style multiverse, where your main character happens to be the luckiest idiot in the multi-verse. If your character is escaping death in incredibly lucky ways, the downside is that all the nearly-infinite versions of your main character would have died horribly, and your main character might be basically the only one (or one of a very few) that haven't managed to kill themselves.

It wouldn't be so much a power as sheer dumb luck.

It would be interesting to see a character wrestle with how his decisions have resulted in the deaths of countless millions or billions of his counterparts in other universes in the multiverse - once he finally finds out the "reason" he's been so lucky. It might also cause him to reconsider continuing his risky behavior - since it could still get him killed at any moment.