Factorio Forums

Wake up, sheeple, the biters are onto all of us.

I recently started playing Factorio and a few things quickly became apparent. Let's state some facts.

1) Assumption: the world of Factorio is an infinite plane.
Confirmation: it says right at world gen screen that if you do not specify the dimensions, then the generated world is potentially infinite.

2) Assumption: there is a day/night cycle in the world.
Confirmation: play at any map and let time passes a bit, the pattern soon become obvious. If you want more precision, install a ui mod that shows the daytime. If you want precision without any mod, build a solar panel nex to a electric pole not connected to anything, then observe the power levels on it.

3) Assumption: the world generates gravity.
Confirmation: it says in the backstory that the player was on a ship that fell.


Now, how exactly can all of these facts be true at the same time? Let's try a few hipotheses:

A) Simple Star System hipothesis: the world orbits a star, much like the way Earth orbits the Sun.
Counter: the world cannot be tidally locked to the star, otherwise there would be no day cycle, but if there is axial rotation, then the star would soon collide with some point of the surface.

B) Binary Star System hipothesis: the world orbits a binary system, composed of a (small) blackhole and a normal star. When the blackhole passes in front of the star, it appears to be night.
Conter: if the world orbits a star system, then there can be no other star system in the universe, othewise those would eventually collide with the surface; moreover, the game says that the player is on an alien world, which implies they came from a different one, contradiction. If, instead of orbiting, the world is stationary in relation to the star system, then a positive attraction exists with its center of mass, which means they would eventually collide. Another counter is that if the world orbits a binary system, then the further you travel from the origin coordinates, the less light reaches the surface, which would mean distant solar panels would generate less energy.

C) Moving Star Grid hipothesis: on a plane parallel to the surface exists a star grid, travelling at constant speed in one direction (say, east->west). The distance between lines of stars on this axis is considerably larger than on the transversal axis (north-south), so as the stars move in relation to the ground, it creates the appearance of a day cycle.
Counter: Again, gravity laws mean the world and the stars would attract each other, eventually colliding. Also, if the player hops on a train and travel great distances in the direction the stars are moving, they would observe doppler effect on day lengh, which does not happen.

D) Infinite Stack hipothesis: the world/star grid planes described above are in fact part of an infinite stack of such interleaving planes. There is gravitational equilibrium everywhere (while still allowing each world to exert gravitational force on surface objects), which would explain why the star grid do not collide with the surface.
Counter: still does not explain the absence of doppler effect on day lengh.

E) Simulation hipothesis: the world exists as a simulation inside an electrical device. Thus, it can be constructed in such a way that it works outside any plausible astrophysics explanation.
Counter: because reasons.


Well.... "When you have eliminated all which is impossible, then whatever remains, however improbable, must be the truth".

F) Illumibiter hipothesis: Flying biters exist. Their sole task is flying in circles and emitting light from their skin (hence their name - illumibiters), an ability that helps them locate their prey. They eventually developed the ability to stop emitting light in a cyclical pattern as to induce any prey into eventually enter panic attacks as "night" falls.
Counter: none. All three initial statements hold true without incurring any absurdities.


Phew, I can finally take the tongue out of my cheek.

Assumption #1 is not true, the world dimensions are limited. It is false advertising on the map generation page.

Though to be fair, if you die of old age before you walk off the edge of the world, it seems infinite to you.

For many many centuries, the earth was flat, and had day/night cycles, and no one ever fell off the edge, so it must have been infinitely large.
So either
a: you are smarter then all the humans who lived on earth before the earth became round
or
b: you've made a mistake somewhere in you logic, and its just a matter of time before someone smarter then you sticks your nose it it.

guess which is much more likely!

The world is like quantum dynamics. It only exists once we observe it. Till then, it's only 'potentially' unlimited.

(Thank you for this. Gave me a good laugh.)

Let's say like so:
- It's a planet - more or less like the earth
- so it's round
- the reason why it looks flat is the same reason as on earth: Because it's so big

And there is currently no way to prove this thesis, cause there is no computer, which has enough RAM to test it. Or is there any? Who wants to drive ~40,000 kilometers? (40,000,000 tiles!) How much coal is needed for that? How long does it take to drive around the surface to the start? How long does a save need then?

ssilk wrote: And there is currently no way to prove this thesis, cause there is no computer, which has enough RAM to test it. Or is there any? Who wants to drive ~40,000 kilometers? (40,000,000 tiles!) How much coal is needed for that? How long does it take to drive around the surface to the start? How long does a save need then?

That's actually pretty easy to disprove with current PCs, if the planet is just as small as Earth (or comparable). Just use linear map (or scan the planar map only along one axis). If the world is round, you should be able to circulate it before running out of computing resources. I have pretty dated PC and had maps that extended beyond 6k*6k tiles, which is the equivalent of 40M*1. If the alien planet follows basic rules of "our" physics, it does not even need to be of comparable size to Earth. Planets that are e.g. 100 times bigger than earth simply can't have the features that are portrayed in Factorio, because of gravity crushing everything, and you are nearing nuclear fusion capable environment (Jupiter is ~10, Sun is ~100 times the earth diameter).

Ok, assue we create a world with 20 tiles width, then 40 Billion tiles length will result in:
20 tiles * 40,000,000 = 800,000,000 tilesquadrat.

Lets now assume one tile needs only 2 bytes:
800,000,000 tilesquadrat * 2 / 1024 / 1024 = 1,526 Megabytes or 1.5 Gigabyte.

I think 2 bytes is a bit too less, but assume that this is manageable with todays computers.

To travel the 80,000,000 tiles by car you need about 800 hours (30-35 days). Hm. This sound a bit too much for me, but I didn't found an error in my calculation. And you need a lot of coal. I think even a completely filled inventory of a car and character is not enough.

You have a lot of coal and other resources on map 40M tiles long, even with the scarcest settings possible, and you can of course build factory to supply you en route But I wasn't the one suggesting it's good idea to explore the map manually using car expedition... I just asserted the argument on computing power. There are commands to explore the map for you, and you are then looking into 35 minutes task rather than 35 days one And from the explored map you could easily show that such a world would be physically unfeasible (using the "physics do apply" assumptions above).

ssilk wrote:To travel the 80,000,000 tiles by car you need about 800 hours (30-35 days). Hm. This sound a bit too much for me, but I didn't found an error in my calculation. And you need a lot of coal. I think even a completely filled inventory of a car and character is not enough.

But what if you use a modified version of FARL, that not only auto builds tracks on the front, but also auto deconstructs tracks on the back. You start the journey with two wagons full of solid fuel (sans one stack reserved for the tracks) and the FARL also picks any wood standing in the way. You could even carry a stack of miners in your pocket to be safe.

It doesn't solve the memory problem, but maybe that's only a problem for today's computers.

RedDagger wrote:

ssilk wrote:To travel the 80,000,000 tiles by car you need about 800 hours (30-35 days). Hm. This sound a bit too much for me, but I didn't found an error in my calculation. And you need a lot of coal. I think even a completely filled inventory of a car and character is not enough.

But what if you use a modified version of FARL, that not only auto builds tracks on the front, but also auto deconstructs tracks on the back. You start the journey with two wagons full of solid fuel (sans one stack reserved for the tracks) and the FARL also picks any wood standing in the way. You could even carry a stack of miners in your pocket to be safe.

It doesn't solve the memory problem, but maybe that's only a problem for today's computers.

Why would you need modified version of FARL?
It already does exactly whats described. Only problem would be potential biters and supply of concrete for bridges (or no water in map gen).

You assume the daylight comes from a sun but in fact, we can show that this is not the case by looking at the shadows of objects on the ground. If the light comes from a point source, then it must be infinitely far away because the angle of the shadows that it produces are always the same no matter where you are on the surface. Also, the source cannot move, since the shadows never change their shape or orientation.
The solution: There is a second infinite plane a fixed distance above the suface, that uniformly emits light at a single angle. If the light was emitted radially from each point on the surface, then there would be no shadows. Since there are shadows at a fixed and consistent angle, then the case must be that the surface only emits parallel rays of light.

Oh, Malcolm, we are still deliberating whether the world is really an infinite plane or just a very large sphere.

Giving more thought to this, making a circumnavigation (circumtravelling maybe, given its not really on a boat) is probably overkill. We can infer the radius with a much more manageable experiment: travel 100km north from the origin coordinates, turn 90 degrees (east, for instance), travel another 100km, then travel back to the origin coordinates by a geodesic. We can use the lengh of this geodesic to calculate the radius. It should be easier on both time and memory constrains.

Moreover, I think it is safe to assume any test could be conducted on peaceful, since it is unlikely that the world's shape is correlated to the aggressiveness of the locals.

MalcolmCooks wrote:You assume the daylight comes from a sun but in fact, we can show that this is not the case by looking at the shadows of objects on the ground. If the light comes from a point source, then it must be infinitely far away because the angle of the shadows that it produces are always the same no matter where you are on the surface. Also, the source cannot move, since the shadows never change their shape or orientation.
The solution: There is a second infinite plane a fixed distance above the suface, that uniformly emits light at a single angle. If the light was emitted radially from each point on the surface, then there would be no shadows. Since there are shadows at a fixed and consistent angle, then the case must be that the surface only emits parallel rays of light.

It could be so, that the planet faces always to the sun. Like Mercury for example. But at the same time he is rotating around a bigger planet (a gas-planet like Jupiter).
So the "night" is when the planet is in the shadow of the gas-planet.

That would explain the non-moving shadows.

ssilk wrote:It could be so, that the planet faces always to the sun. Like Mercury for example. [...] That would explain the non-moving shadows.

https://en.wikipedia.org/wiki/Mercury_(planet) wrote:For many years it was thought that Mercury was synchronously tidally locked with the Sun, rotating once for each orbit and always keeping the same face directed towards the Sun, in the same way that the same side of the Moon always faces Earth. Radar observations in 1965 proved that the planet has a 3:2 spin–orbit resonance, rotating three times for every two revolutions around the Sun.

Mercury has some interesting sky constellations due to its 3:2 spin–orbit resonance, but there is no place on it that always faces the sun.
On some locations on Mercury, you can see the sun go up high in the sky, then reverse (or even set), and then resume its path.
A pretty good illustration: http://www.messenger-education.org/Inte ... ercury.php

https://en.wikipedia.org/wiki/Mercury_(planet) wrote:For the same reason, there are two points on Mercury's equator, 180 degrees apart in longitude, at either of which, around perihelion in alternate Mercurian years (once a Mercurian day), the Sun passes overhead, then reverses its apparent motion and passes overhead again, then reverses a second time and passes overhead a third time, taking a total of about 16 Earth-days for this entire process.

Well, that may be the case with Mercury, specifically. It doesn't necessarily mean what ssilk described cannot exist. The problem I see is that day as represented in the game would only occur on a portion of the surface.

You guys missed the obvious.

Once the game is finished (you know, satellites and spaceplatforms and such) it is revealed, that everything happened was just a computer simulation to research of a cheap (because you just need one person in a one way capsule) colonization method for distant planets.

You are assuming that the planet exists in our reality. For all we know, the Player's race developed some sort of way to jump between dimensions. Such technology would explain how player can store hundreds of various items in the inventory, the inventory is a pocket dimension with even more different rules than the one where the game takes place.

So the 'planet' could be as finite as universe. Have fun driving to the edge of the universe in your car...

I vote for the dyson sphere hypothesis, or ring world if your limit the map along one dimension

It could also be a two layered dyson world. Your layer has all the stuff, while a higher layer provides regular day and night.

Oh, yes. A Dyson Ring? https://en.wikipedia.org/wiki/Dyson_sphere - sadly not in the English version, only German: https://de.wikipedia.org/wiki/Dyson-Sph ... e#Der_Ring

A Ringworld? https://en.wikipedia.org/wiki/Ringworld