For the whole linear vs inverse square thing, like it would be correct if they said 1/r was linear falloff and that 1/r^2 was proportional to the inverse square. But just saying linear and inverse square is inconsistent. It should be linear and quadratic falloff, or inverse and inverse square proportionality. This is a stupid thing I'm nitpicking. Ok bye like and subscribe
bahahaha Man that is an old piece of code you found there. You are totally correct, and to my relief, the repo tells me someone else wrote that bit of code. And i will bring it up so that the error will bother them to some degree. (Because it already bothers me, as someone that supposedly took an orbital mechanics class at some point. They must share in the bothering.)
@@jeffrey_yu Shocked I never noticed this comment, really cool to have a mobius dev respond here! Thank you for spreading the bothering, and for making a great game! :)
@@user-bw4jm1bv1i yeah! It's a neat quirk: The Interloper has to be able to detect fluids so it can destroy itself when it enters the sun (the sun counts as a fluid in the game), but this makes it also detect when it enters the water on Giant's Deep (another fluid) so it gets stuck inside of it. Other planets just phase through each other when they should collide. Because of this, the Interloper can also hit Hollow's Lantern and get melted there since burning up in the lava is the same as burning up in the sun (this only happens if you extend the time loop and prevent the sun from expanding)
To be honest, that's _incredibly on-brand_ for the Nomai. Figuring out loopholes in universal laws is kind of their _thing,_ in a way. (not a spoiler since i didn't say _what_ loopholes, more a general mission statement for the species)
@@xen-42In an interview they stated that they made that decision because it works better with their small scale solar system. I presume the gravity fall off as you move away from a body feels better for the player.
I also love the choice to use a slowed down video instead of a screenshot, seeing the head dev flailing in and out of frame like that is hilarious when put together with the zoom and music.
It feels like the more stuff you try to simulate in real time, the less stuff ends up being realistic. Things are just too far away and too big in the first place for them to be experienceable in a 1 to 1 to life manner in any space game. But games like Outer Wilds and Kerbal Space Program that give you the feeling of exploring a solar system (and not just a bunch of menus that take you to other planets) are amazing, and I'm willing to sacrifice some physics for that!
I totally agree about the fun of playing space games being a result of some realism being taken away. And it's funny to imagine if a 1:1 realistic space game were actually to be made and become popular somehow. I think it would be quite a shock to a lot of people who aren't already space-geeks at all. I'd love to see their reactions to being immersed firsthand in a game and suddenly vividly realizing just how ridiculously slow, tedious, and boring space travel would actually be. How much more visually empty the solar system appears without doing things to tweak the scale and speeds. How much inconvenience/precision there is in managing trajectories, departures/destinations/arrivals, realistic resource management, and changes of plans. It would be quite the vividly jarring experience for someone just learning about the true scale of space/physics to play an interactive simulation of the truly massive, isolating, and desolate scale of things in the solar system and beyond. Maybe there'd be a few less people obsessed and caught up in the hype built up around the idea that soon we'll be happily colonizing Mars...
Wow I assumed that the system was on rails. That's awesome that it is a proper physics simulation and seeing you mess with the parameters was fantastic. This game has so many interesting design choices.
And here I thought the planets were just a precariously balanced orbital equivalent of stacking rocks. Turns out they have no affect on each other, and don't even share the same laws of physics from one body to another. I'm impressed. Great games truly are smoke and mirrors. Great video.
Damn, i was SURE the gravity only affected the players and the planets orbit was properly pre-defined movement. the fact that they actually are following gravity from the sun (and sometimes each other) EVERY FRAME is absurd to me, this game is crazy
I vaguely remember that they said in the NoClip documentary that linear gravity was stronger than inverse square when closer to the planet and weaker when farther away. The idea is that when you're on a planet, it should be the main force of gravity you feel, but when you're far away you shouldn't feel its pull too much. The sun has inverse square gravity so you always feel its pull, if just slightly. I'm not sure if this is accurate, it's just what I remember.
Makes sense, I should go back and rewatch that! Also tracks with the Attlerock having inverse square gravity, to have you always slightly feel its pull on TH.
From the video it looked like you primarily checked the orbits but you could have tested how it feels to fly and walk around with the different gravity settings. I think that's where their design decisions will make sense.
I think I remember them saying that too, but it's the wrong way round. 1/R^2 decays faster, so it can be stronger when you're on the planet and weaker further away. Idk why they got confused. It's possible they wanted you to feel the gravity of the planets more while in space, to make the solar system feel more interconnected.
@@blblblblblbl7505 maybe it was make sure the artificially increased weight of the attlerock and the sun when you're nearby them isn't so strong that it's impossible to escape/land on the sun station? i might have misunderstood this video though
i think a lot of it is about the speeds that the player is traveling at. because the game doesn't have a time warp feature (like kerbal does, for example), you're gonna be scuttling around the solar system at speeds that are within a few orders of magnitude of each other. with newtonian gravity, you barely feel the effects of the planet until you get right up next to it and suddenly you get slingshotted around at ludicrous speed. you actually ended up noticing a really important detail with the constant orbital velocity in inverse-linear gravity, since orbital velocity can also be thought of as the speed at which inertial forces start to outweigh gravitational ones. with inverse-square gravity, inertial forces would start to always win out the moment you get away from the planet at all.
@@xen-42 As far as I know that would not cause the end of the solar system. Gravitatiional pull for the planets around would remain the same as the mass remains the same, only being pulled into the black hole as it reaches the event horizon
1. In order to have accurate physics AND celestial distances, they would have to start somehow factoring in relativity to the stupid speeds you would need to travel. 2. Also, the planets themselves would need to be massive as well, which means traversing a planet on foot should take actual years. By multiplying the strength of Gravity by one `distance`, you can hypothetically shrink everything to the sqrt of its original size, and your equations will give you velocities and momentums that make it look like you are in ‘big world’ moving at ‘big speeds,’ as long as your camera is zoomed way in. To the player, they experience moving around the local environment as a normal person that exists at correct proportions to their environment would, and the outside environment appears to move relative normal to itself. - If you were an omnipotent, Bird’s Eye camera, you would see a model of a solar system that moves similar to our own-in terms of how the planets are positioned relative to each other, and how that changes over time (but sped up, ofc.) However, you would see the protagonist zipping all over the solar system at impossible speeds, sprinting like an F-35 when on foot and zapping across space like a tachyon when traveling by spaceship. -The simulation would be very confusing as a physicists, until you inspected the code and revealed the trickery. You would then see that the problem is our programmers faked the part that made the orbiting motion look real by playing with the `distance` variable. In a bird’s eye cam-that is, what is presumably meant to be a proper physics sim-this would be silly and a bug. In game design however, it’s a brilliant and even somewhat elegant solution (particularly since tech is filled with engineers that “hate math.”) By altering the physics slightly-which already is a more involved technical task; providing physics, animations, etc. is arguably the largest problem people solve by using a game engine-the developers were able to take what is likely the impossible task of organizing all the code, camera work, changes, transitions, etc. that would have to go into making your character travel realistic distances to realistically-sized planets, yet fit into your screen and not take 40 billion years to complete without looking incomprehensible. 3. Linear and inverse are not mutually exclusive. G = m*M/r; the relationship between changes in r with changes in G is still linear.
The moment I fell in love with this game is the details of gravity that I experienced firsthand, when I land on the Interloper comet, I was reading the text from the shuttle as I see my spaceship indicator just moving, turns out this comet was traveling pass the Giant’s Deep, the gravitational force of this planet is so strong, it pull my ship into their orbit, while I have to fly off and match the velocity, move painfully slow closer to the ship, well at least I just sit there in silence and appreciate the detail of this game like Gabbro appreciate the time and nature.
Yeah the physics are so great, its crazy to me that some people really complain about the controls and stuff even though it allows for cool events like that to happen
Remember that this also affects the Coriolis effect, since the Coriolis force experienced would be Fc = m * a′ (derivative of a) and it would be REALLY annoying to pilot a ship while stiring in a weird curve when you intent to be just falling in a line without modifying the "path" you're intending to do. What I'm trying to say it's that if you stir the ship while moving in orbit, it would make you curve left or right and change your trayectory instead of just stirring as you're advancing forward, I'm not even sure how the "matching velocity" would work in this scenario?. Edit: I just worked it out, in an universe where gravity decays at 1/r, it would be: Angular Velocity: Av = √GM/R^2 θ being the angle between the velocity vector and the axis of rotation. Coriolis Force: Fc = −2*Av*m*v*sin(θ) (this one doesn't change, but A does) the normal Av would be √GM/R^3 so it DOES change a lot the things that spin, I wonder if you could try shooting a vessel inside the stranger to see if the trayectory it takes feels "weirder" than usual, or the falling of the things in Brittle Hollow, idk
This was oddly terrifying to watch, much like OW in general. Was constantly expecting every celestial body to suddenly get sucked into each other because of some typo in the equation or something.
trying to land on the sun station made me so mad because it became obvious that gravity from the sun worked differently for me, my ship, and the sun station
I could not understand why the ship absolutely would not just settle down on the station with matched velocity, or why in a sun-station-height almost-stable orbit in the ship when you unbuckle you're flung against the ship wall like you're under huge acceleration. I ended up wondering if there was some kind of turbulent solar wind (but only very close to the sun), and I also didn't know why you aren't flung against the wall if you jump in the sun station if that's basically what happens in your ship (it shouldn't actually happen in either case)... tbh though I still don't understand what makes it so hard unless it's the sun station's inverse linear gravity overpowering the sun's and messing up your orbit? but then shouldn't it be easier to land and not harder? Yeah I don't get it.
I can pretty much match the station's orbit, I just can't get out of the ship and into the station without taking on a bunch of momentum that I can't understand or mitigate
@@silphv i think it helps to match velocities a little bit past the sun station and tilt the ship so it's looking just below the horizon before getting out of your seat. that way you should be able to walk with the sun's gravity and when you leave your ship the sun station should be below you. but god it's so annoying
Awesome, I saw the stuff about inverse square/linear in the codebase and never bothered questioning its inconsistency until now. Was also not expecting to see my mod appear in a video I'm watching lmao
I think what he meant by linear is that they take the reciprocal of a scalar instead of an exponential. And it would simulate a stronger gravity than inverse square since as an example; 1/60 is 0.016 and 1/60^2 is 0.00027 or put another way 1/60 = (1/100 + 6/1,000) 1/60^2 = (2/10,000 + 7/100,000)
I was literally just wondering if anyone had found a way to use the planets' gravity to knock one of them out of orbit, and this video perfectly answered that question. Everytime I find out a new way the developers "cheated" I come away more impressed by their design decisions.
This has proved (to me) that newtons laws are not required to function. At least in outer wilds, this is great, I wonder what other things the devs tried when trying to get thing to orbit nice.
Some places, like the white hole station, just ignore gravity entirely and put the player into zero gravity. Else you'd always be sliding towards the sun since its not moving.
The accuracy is so shockingly close to realism that it feels like the dev really wanted it to have an already 'settled' feel. It's like it's naturally progressing, just like our own exploration is intended.
I would be very interested to see a mod that makes all the gravity the realistic type, but adjusts the numbers so everything remains playable. I feel like I could get a better intuition for orbital mechanics that way.
Even if it isn't entirely accurate, I do like the term linear gravity. If you just say inverse gravity, it sounds more like gravity working the opposite way where things would get pushed apart. Thank you friend.
Cool derivation on the orbital speed! So if the sun wasn’t inverse squared, manually inserting yourself into the sun station’s orbit wouldn’t work right.
I don't personally know anything about maths or astrophysics but I always had a sneaking suspicion that if the physics were really realistic, the player would start spinning like a top at bizarre tilt angles with incredible speed due to any slight imperfection in their movement. It would become immediately impossible to deal with as a human(hearthian) at the wheel.
One thing I was wondering about (SPOILERS for base game and dlc ofc) When strangers and Nomai arrived to the Solar System, they did not comment on the absurdly short distance between planets, nor the small size of them. Thus, I wonder: is the whole OW universe this small? I mean, this is a "toy" solar system with respect to real ones, and the nomai (which travelled the whole cosmos at this point) should know that
It has 1/r gravity? That explains why whenever I tried to slingshot around the black hole to get around there easier I always ended up on really weird paths that I didn't expect despite my thousands of hours of ksp
Very interesting! How stable were the orbits over the 22 min window with everything on inverse square? Did the planets influence each other too much in this scaled down universe? How well do gimmicks like the 0G cave work with inverse square gravity? That probably would have to have been much smaller with proper gravity. Oh, I guess the orbital velocity is one of the major reasons for the sun to be on inverse square gravity from a game design standpoint, as you have planets passing each other which is good for exploration.
With inverse square everything stays stable just moves a bit slower since the planets don't influence each other. When I set it to do that, the solar system falls apart immediately. The 0G cave is entirely separate from the gravity of the planet. There's like, four types of forces going on there: long distance gravity when you're in the air (this is 1/r or 1/r^2), then theres the surface gravity (its just constant), then the interior gravity (it starts going decreasing past a certain depth) and then it immediately cuts off to 0G inside the cave. Yeah inverse square makes sense for the sun to have the planets pass each other in a more realistic looking way. Harder to notice the effects of 1/r when it only affects the moon of a planet, since none of the planets have more than one moon so you can't really compare the effects.
Technically, not _every_ body is simulated directly. A few are scripted, like the you-know-what around the sun (its _debris_ is orbitally simulated though), as well as the DLC area... and I thought that the Hourglass Twins were scripted _with respect to each other_ (they'll always be an exact distance apart), while only their _combined midpoint_ was subject to orbital physics? Also fun experiment in vanilla: Fly far, fly fast, and look back via map view once you're at least 25,000km out. Deep Space Kraken has come to play!
Is the you-know-what the sun station, because what makes you say it's scripted? Looking at it I see no way in which it differs from any of the other orbiting bodies. The Twins are a bit weird in that they orbit each other (with forces simulated) and then the force from the sun felt by their combined midpoint is applied to them. The Stranger and Dreamworld are unique yeah in that neither experiences gravity from the sun and the Stranger is scripted to start accelerating away. The floating point error deep space kraken stuff is fun I agree!
Someone able to develop a mod, understand and implement the math of astrophysics and explain it in a simple yet entertaining way? Count me in! Side note, you seem to have a slightly monotone voice when you record. I can't say it's a fault but that's the only flaw I was able to find in your content. Love your style anyway! ^^
Good idea, I tried something similar in a mod I made for a Lore Explorer video (see his Scale Showcase Mod video) but for some reason when he went to record the eye stopped appearing so it didnt make it into the video. Its definitely something I could consider revisiting.
"You can see why they chose to not have the game play like this" Hm... no, actually I can't. What am I supposed to notice? Are the planets too far apart or something?
hmm yeah lol looking back at that part of the video i dont really let it play out enough, but everything runs into each other and falls into the sun or gets kicked out of the solar system, you can at least see that the twins have completely separated from each other too
@@xen-42 Ah yeah, that's what happened in most of my n-body simulations too! I just didn't see it happen in the video so I assumed you meant something else. Thanks for the reply!
Yeah, but those tidal forces from Giants Deep only affect the player (and the scout and the ship) and don't affect the Interloper, else it would dramatically affect it's orbit!
great video but it felt really confusing to know when exactly the simulation was considering inverse law or inverse square law, maybe adding a visual cue or just specifying exactly what planets/stars/comets followed which law in the original game and in the simulation. I thought at first all the cosmic objects had inverse gravity in the original game, but I kinda got lost at the part you showed attlerock, like, did it follow 1/r² originally? Also, there's a easier way of showing why v = sqrt(GM) in inverse law case, simply taking GMm/r = mv²/r (centripetal force)
Everything has inverse gravity except the sun and the attlerock which have inverse square. Fair point about the derivation potentially being easier, I originally did the math to account for elliptical orbits as well hence the complicated math but simplified it down to circular for the video
how does gravity work on the Stranger? I think the "gravity" there is MUCH too strong for the rotational speeds they're using and it suddenly cuts in when you get close to a surface you're supposed to be allowed to land/stand on
Yeah you're absolutely right, they fake it and add in some extra gravity in a cylindrical volume that pushes u outwards. The amount of artificial gravity the spin gives is barely enough to stop you just flying everywhere.
Nope, plus when you enter into its cloaking field there's also a zero-gravity volume that prevents you from being affected by the sun's gravity anymore too. Else you'd just get pulled out of it and sucked into the sun. If you leave your ship floating inside the field, when the Stranger starts to leave midway through the loop your ship will fall out of the zero-gravity field and get sucked into the sun.
Hold the fuck up. Elliptical orbits are not possible* with F=1/r, only with F=1/r^2 and F=1/r^3which is why you cannot orbit any of the planets. Now what I am stuck on is the hourglass twins. These can be reduced to a 2D Lagrangian problem, masses connected by a spring, but what fucks me up is that they seem to remain at their equilibrium separation even after the perturbation begins and one of them starts gaining mass. Can you answer this? Are those two on rails as a singular object at the center of mass? Is there some trick where the sand is weightless? Thanks
Do you have example code of how to get the gravity area to work? I created a rigid body player and a planet with a gravity field, but the player doesn’t move toward the planet. I set the player gravity scale to 1 and the gravity field to priority 1
How did you calculate the escape velocity for gravity which is inverse with the distance? ln(infinity) = infinity, so Fdx with x from r to infinity is infinity. Also, there' was a typo in your escape velocity for normal gravity, it's v = sqrt(2GM/r) (but that's just a nitpick)
Oh nah what I showed was meant to be circular orbit velocity, should've made that more clear in the video. You make an interesting point though, I hadn't realized that the integral just doesn't converge. So mathematically there is no such thing as an escape velocity when orbiting any of the planets in Outer Wilds, although mechanically the devs just cheat that and have a strict cutoff radius where you stop feeling the effects of gravity. If you still want the derivation for the non-escape orbital velocity I uploaded it over here: github.com/xen-42/xen-42/blob/main/Linear_gravity.pdf
@@xen-42 Thanks, that was exactly what I was looking for. Taking the limit when going from (9) to (10) was non-trivial. Had to use L'Hôpital's rule and calculate myself that lim (x->a) (ln(x/a)(x²/x²-a²)) was indeed 1/2. Was fun, though :) Didn't expect the outcome to be so simple and elegant and resemble the escape velocity of normal gravity, even though we calculated something else.
Nah, each planet is only effected by the planet/sun that it orbits. Some are affected by water which is a weird one, so the Interloper can get stuck inside Giants Deep if it were to collide with its orbit because its physics controllers detect the water (it uses this to detect entering the sun, since water and the sun are both handled by a "fluid detector" in the game's code it ends up also detecting water as a side effect). Unity (at least the version Outer Wilds was made in) doesn't have built in physics support that would allow the player to push planets unless the devs were to recreate all the planet collision models (it only works for convex meshes), else it'd be easy to just change a variable and allow the player to deorbit any planet.
Smh my head, outer wilds can't even simulate real gravity, truly ruined the immersion, 0/10 game, I wish I had never played this game So tht I can play it again
"Why didn't they just do This?" Changes this. Then procedes to show WHY "this" wouldn't have worked. This was a bit of the most useless video I've seen. I know I'll watch ANY thing on outer wilds... this video proves why that needs to stop.
@xen-42 I'm sorry you took this as a hate comment. I'd like to rescind any hate you felt from me, it wasn't intended. It was negative, sure. But the things you complained about, like the physics, are some of the best things in the game to me. So i commented what i did. I apologize if you took that as hate though, it wasn't. I've seen multiple of your videos and thumbs up and commented positively before. I appreciate the time you took to create your video. Thanks for sharing outer wilds content. I hope you can re read my comment hear my apathy and confusion now, instead of hate.
@@GrandNoble I see. I think you've misunderstood me though, because I'm not complaining about the physics, I'm just saying that the gravity is inaccurate to real life physics. Anyway, shame to hear you didn't like the video. Cheers.
For the whole linear vs inverse square thing, like it would be correct if they said 1/r was linear falloff and that 1/r^2 was proportional to the inverse square. But just saying linear and inverse square is inconsistent. It should be linear and quadratic falloff, or inverse and inverse square proportionality. This is a stupid thing I'm nitpicking. Ok bye like and subscribe
bahahaha Man that is an old piece of code you found there. You are totally correct, and to my relief, the repo tells me someone else wrote that bit of code. And i will bring it up so that the error will bother them to some degree. (Because it already bothers me, as someone that supposedly took an orbital mechanics class at some point. They must share in the bothering.)
@@jeffrey_yu Shocked I never noticed this comment, really cool to have a mobius dev respond here! Thank you for spreading the bothering, and for making a great game! :)
i was going to say this but then i saw you already said it
At 4:19 does the Interloper hit Giant's Deep?
@@user-bw4jm1bv1i yeah! It's a neat quirk: The Interloper has to be able to detect fluids so it can destroy itself when it enters the sun (the sun counts as a fluid in the game), but this makes it also detect when it enters the water on Giant's Deep (another fluid) so it gets stuck inside of it. Other planets just phase through each other when they should collide. Because of this, the Interloper can also hit Hollow's Lantern and get melted there since burning up in the lava is the same as burning up in the sun (this only happens if you extend the time loop and prevent the sun from expanding)
“They made a few tweaks to the laws of physics” is my new favorite quote.
'science compels us to blow up the sun'
To be honest, that's _incredibly on-brand_ for the Nomai. Figuring out loopholes in universal laws is kind of their _thing,_ in a way.
(not a spoiler since i didn't say _what_ loopholes, more a general mission statement for the species)
So you’re telling me they haven’t actually been using newtons law this whole time????? I’m gonna need some time to process this
newtons rolling in his grave
@@xen-42wait newton died?? Rip bro we're losing all the good ones this year
@@xen-42In an interview they stated that they made that decision because it works better with their small scale solar system. I presume the gravity fall off as you move away from a body feels better for the player.
1:24 I just find the zoom in on the equation with the ominous Echoes of the Eye music very funny
I also love the choice to use a slowed down video instead of a screenshot, seeing the head dev flailing in and out of frame like that is hilarious when put together with the zoom and music.
It feels like the more stuff you try to simulate in real time, the less stuff ends up being realistic. Things are just too far away and too big in the first place for them to be experienceable in a 1 to 1 to life manner in any space game. But games like Outer Wilds and Kerbal Space Program that give you the feeling of exploring a solar system (and not just a bunch of menus that take you to other planets) are amazing, and I'm willing to sacrifice some physics for that!
yeah for sure, outer wilds in an ultra realistic life sized solar system would be nowhere near as good
I totally agree about the fun of playing space games being a result of some realism being taken away. And it's funny to imagine if a 1:1 realistic space game were actually to be made and become popular somehow. I think it would be quite a shock to a lot of people who aren't already space-geeks at all. I'd love to see their reactions to being immersed firsthand in a game and suddenly vividly realizing just how ridiculously slow, tedious, and boring space travel would actually be. How much more visually empty the solar system appears without doing things to tweak the scale and speeds. How much inconvenience/precision there is in managing trajectories, departures/destinations/arrivals, realistic resource management, and changes of plans. It would be quite the vividly jarring experience for someone just learning about the true scale of space/physics to play an interactive simulation of the truly massive, isolating, and desolate scale of things in the solar system and beyond. Maybe there'd be a few less people obsessed and caught up in the hype built up around the idea that soon we'll be happily colonizing Mars...
Wow I assumed that the system was on rails. That's awesome that it is a proper physics simulation and seeing you mess with the parameters was fantastic. This game has so many interesting design choices.
You'd think so, right? Glad you liked the video!
If you move far away from þe system, þe orbits will actually get messed up due to floating point inprecision
@@Nikola_M Why are you using "þ"
@@thepotatoportal69 why aren't you?
@@sabiro2315 Because it looks too much like "b" or "p" and is confusing for dyslexic people. And besides, it would save basically no characters.
Newton hates him
one quick trick to destroy elliptical orbits
And here I thought the planets were just a precariously balanced orbital equivalent of stacking rocks. Turns out they have no affect on each other, and don't even share the same laws of physics from one body to another. I'm impressed. Great games truly are smoke and mirrors. Great video.
Outer Wilds has some of the best smoke and mirrors I've seen for sure
Damn, i was SURE the gravity only affected the players and the planets orbit was properly pre-defined movement. the fact that they actually are following gravity from the sun (and sometimes each other) EVERY FRAME is absurd to me, this game is crazy
I vaguely remember that they said in the NoClip documentary that linear gravity was stronger than inverse square when closer to the planet and weaker when farther away. The idea is that when you're on a planet, it should be the main force of gravity you feel, but when you're far away you shouldn't feel its pull too much. The sun has inverse square gravity so you always feel its pull, if just slightly.
I'm not sure if this is accurate, it's just what I remember.
Makes sense, I should go back and rewatch that! Also tracks with the Attlerock having inverse square gravity, to have you always slightly feel its pull on TH.
From the video it looked like you primarily checked the orbits but you could have tested how it feels to fly and walk around with the different gravity settings. I think that's where their design decisions will make sense.
I think I remember them saying that too, but it's the wrong way round. 1/R^2 decays faster, so it can be stronger when you're on the planet and weaker further away. Idk why they got confused. It's possible they wanted you to feel the gravity of the planets more while in space, to make the solar system feel more interconnected.
@@blblblblblbl7505 Yeah, I put 1/r^2 and 1/r in desmos calculator and noticed the discrepancy too. I wasn't sure what to think after that.
@@blblblblblbl7505 maybe it was make sure the artificially increased weight of the attlerock and the sun when you're nearby them isn't so strong that it's impossible to escape/land on the sun station? i might have misunderstood this video though
i think a lot of it is about the speeds that the player is traveling at. because the game doesn't have a time warp feature (like kerbal does, for example), you're gonna be scuttling around the solar system at speeds that are within a few orders of magnitude of each other. with newtonian gravity, you barely feel the effects of the planet until you get right up next to it and suddenly you get slingshotted around at ludicrous speed. you actually ended up noticing a really important detail with the constant orbital velocity in inverse-linear gravity, since orbital velocity can also be thought of as the speed at which inertial forces start to outweigh gravitational ones. with inverse-square gravity, inertial forces would start to always win out the moment you get away from the planet at all.
It‘s also kinda odd that the sun has higher surface gravity than the black hole inside of Brittle Hollow
We need realistic Outer Wilds where instead of a supernova the sun just collapses into a black hole instantly
@@xen-42 all that work saving star astroobjects from being slurped and for what
gravity rocks make it less heavy[???]
@@xen-42 As far as I know that would not cause the end of the solar system. Gravitatiional pull for the planets around would remain the same as the mass remains the same, only being pulled into the black hole as it reaches the event horizon
The moon is 20x heavier than timber hearth? Sick vid
Thanks! the moon should have a black hole inside it instead of brittle hollow smh
Kinda makes sense in-universe; the Nomai moved a significant portion of mass from Timber Hearth to Ash Twin. ;;)
As you keep making outer wilds videos you seem to be steadily losing your mind, and I'm here for it.
it takes a toll on the human psyche
Outer Wilds is and we are.
1. In order to have accurate physics AND celestial distances, they would have to start somehow factoring in relativity to the stupid speeds you would need to travel.
2. Also, the planets themselves would need to be massive as well, which means traversing a planet on foot should take actual years. By multiplying the strength of Gravity by one `distance`, you can hypothetically shrink everything to the sqrt of its original size, and your equations will give you velocities and momentums that make it look like you are in ‘big world’ moving at ‘big speeds,’ as long as your camera is zoomed way in. To the player, they experience moving around the local environment as a normal person that exists at correct proportions to their environment would, and the outside environment appears to move relative normal to itself.
- If you were an omnipotent, Bird’s Eye camera, you would see a model of a solar system that moves similar to our own-in terms of how the planets are positioned relative to each other, and how that changes over time (but sped up, ofc.) However, you would see the protagonist zipping all over the solar system at impossible speeds, sprinting like an F-35 when on foot and zapping across space like a tachyon when traveling by spaceship.
-The simulation would be very
confusing as a physicists, until you inspected the code and revealed the trickery. You would then see that the problem is our programmers faked the part that made the orbiting motion look real by playing with the `distance` variable. In a bird’s eye cam-that is, what is presumably meant to be a proper physics sim-this would be silly and a bug. In game design however, it’s a brilliant and even somewhat elegant solution (particularly since tech is filled with engineers that “hate math.”) By altering the physics slightly-which already is a more involved technical task; providing physics, animations, etc. is arguably the largest problem people solve by using a game engine-the developers were able to take what is likely the impossible task of organizing all the code, camera work, changes, transitions, etc. that would have to go into making your character travel realistic distances to realistically-sized planets, yet fit into your screen and not take 40 billion years to complete without looking incomprehensible.
3. Linear and inverse are not mutually exclusive. G = m*M/r; the relationship between changes in r with changes in G is still linear.
The moment I fell in love with this game is the details of gravity that I experienced firsthand, when I land on the Interloper comet, I was reading the text from the shuttle as I see my spaceship indicator just moving, turns out this comet was traveling pass the Giant’s Deep, the gravitational force of this planet is so strong, it pull my ship into their orbit, while I have to fly off and match the velocity, move painfully slow closer to the ship, well at least I just sit there in silence and appreciate the detail of this game like Gabbro appreciate the time and nature.
Yeah the physics are so great, its crazy to me that some people really complain about the controls and stuff even though it allows for cool events like that to happen
Remember that this also affects the Coriolis effect, since the Coriolis force experienced would be Fc = m * a′ (derivative of a) and it would be REALLY annoying to pilot a ship while stiring in a weird curve when you intent to be just falling in a line without modifying the "path" you're intending to do.
What I'm trying to say it's that if you stir the ship while moving in orbit, it would make you curve left or right and change your trayectory instead of just stirring as you're advancing forward, I'm not even sure how the "matching velocity" would work in this scenario?.
Edit: I just worked it out, in an universe where gravity decays at 1/r, it would be:
Angular Velocity: Av = √GM/R^2
θ being the angle between the velocity vector and the axis of rotation.
Coriolis Force: Fc = −2*Av*m*v*sin(θ) (this one doesn't change, but A does)
the normal Av would be √GM/R^3 so it DOES change a lot the things that spin, I wonder if you could try shooting a vessel inside the stranger to see if the trayectory it takes feels "weirder" than usual, or the falling of the things in Brittle Hollow, idk
Mvp comment right here definitely needs more likes
This was oddly terrifying to watch, much like OW in general. Was constantly expecting every celestial body to suddenly get sucked into each other because of some typo in the equation or something.
trying to land on the sun station made me so mad because it became obvious that gravity from the sun worked differently for me, my ship, and the sun station
I could not understand why the ship absolutely would not just settle down on the station with matched velocity, or why in a sun-station-height almost-stable orbit in the ship when you unbuckle you're flung against the ship wall like you're under huge acceleration. I ended up wondering if there was some kind of turbulent solar wind (but only very close to the sun), and I also didn't know why you aren't flung against the wall if you jump in the sun station if that's basically what happens in your ship (it shouldn't actually happen in either case)... tbh though I still don't understand what makes it so hard unless it's the sun station's inverse linear gravity overpowering the sun's and messing up your orbit? but then shouldn't it be easier to land and not harder? Yeah I don't get it.
I can pretty much match the station's orbit, I just can't get out of the ship and into the station without taking on a bunch of momentum that I can't understand or mitigate
@@silphv i think it helps to match velocities a little bit past the sun station and tilt the ship so it's looking just below the horizon before getting out of your seat. that way you should be able to walk with the sun's gravity and when you leave your ship the sun station should be below you. but god it's so annoying
Awesome, I saw the stuff about inverse square/linear in the codebase and never bothered questioning its inconsistency until now. Was also not expecting to see my mod appear in a video I'm watching lmao
Mobius digital must be using reverse square gravity the way they're pulling my leg with that "linear" thing
Pulling your locomotive limb as the Nomai would say
I think what he meant by linear is that they take the reciprocal of a scalar instead of an exponential. And it would simulate a stronger gravity than inverse square since as an example; 1/60 is 0.016 and 1/60^2 is 0.00027 or put another way
1/60 = (1/100 + 6/1,000)
1/60^2 = (2/10,000 + 7/100,000)
im getting my physics phd right now and ur derivation sent me into tears it was so funny
It would be cool if we could see a time-lapse of the planets orbits with N-body physics; imagine seeing how the orbits evolve over time
Babe wake up new xen 42 outer wilds video
I was literally just wondering if anyone had found a way to use the planets' gravity to knock one of them out of orbit, and this video perfectly answered that question. Everytime I find out a new way the developers "cheated" I come away more impressed by their design decisions.
This has proved (to me) that newtons laws are not required to function. At least in outer wilds, this is great, I wonder what other things the devs tried when trying to get thing to orbit nice.
Some places, like the white hole station, just ignore gravity entirely and put the player into zero gravity. Else you'd always be sliding towards the sun since its not moving.
@@xen-42 actually that reminds me of echoes of the eye, if you go to the dw in the real world, then you’ll slide towards the sun.
The accuracy is so shockingly close to realism that it feels like the dev really wanted it to have an already 'settled' feel. It's like it's naturally progressing, just like our own exploration is intended.
In Europe there's a nintendo game called 'Yoshi's universal gravitation'
i wonder if yoshi uses 1/r gravity too
Damn no wonder their universe ended so quickly if their laws of physics are like this
I would be very interested to see a mod that makes all the gravity the realistic type, but adjusts the numbers so everything remains playable. I feel like I could get a better intuition for orbital mechanics that way.
Even if it isn't entirely accurate, I do like the term linear gravity. If you just say inverse gravity, it sounds more like gravity working the opposite way where things would get pushed apart. Thank you friend.
Cool derivation on the orbital speed! So if the sun wasn’t inverse squared, manually inserting yourself into the sun station’s orbit wouldn’t work right.
Think you'd still be able to but itd be much more chaotic since any elliptical orbit would keep precessing!
I don't personally know anything about maths or astrophysics but I always had a sneaking suspicion that if the physics were really realistic, the player would start spinning like a top at bizarre tilt angles with incredible speed due to any slight imperfection in their movement. It would become immediately impossible to deal with as a human(hearthian) at the wheel.
Asteroid Arcade realistic physics update when???
ten thousand years
@@xen-42 I've marked it on my calendar
Really cool video (this is a comment for the algorithm)
Thank you! (this is a reply for the algorithm)
One thing I was wondering about (SPOILERS for base game and dlc ofc)
When strangers and Nomai arrived to the Solar System, they did not comment on the absurdly short distance between planets, nor the small size of them. Thus, I wonder: is the whole OW universe this small? I mean, this is a "toy" solar system with respect to real ones, and the nomai (which travelled the whole cosmos at this point) should know that
4:05 MADE IN HEAVEN!!
It has 1/r gravity? That explains why whenever I tried to slingshot around the black hole to get around there easier I always ended up on really weird paths that I didn't expect despite my thousands of hours of ksp
At least the sun has 1/r^2 so ksp knowledge still works for getting to the sun station!
physics...? who needs it???
uh maybe ALBERT EINSTEIN?!?!?!
You should release the mod honestly.
Very interesting! How stable were the orbits over the 22 min window with everything on inverse square? Did the planets influence each other too much in this scaled down universe? How well do gimmicks like the 0G cave work with inverse square gravity? That probably would have to have been much smaller with proper gravity. Oh, I guess the orbital velocity is one of the major reasons for the sun to be on inverse square gravity from a game design standpoint, as you have planets passing each other which is good for exploration.
With inverse square everything stays stable just moves a bit slower since the planets don't influence each other. When I set it to do that, the solar system falls apart immediately. The 0G cave is entirely separate from the gravity of the planet. There's like, four types of forces going on there: long distance gravity when you're in the air (this is 1/r or 1/r^2), then theres the surface gravity (its just constant), then the interior gravity (it starts going decreasing past a certain depth) and then it immediately cuts off to 0G inside the cave.
Yeah inverse square makes sense for the sun to have the planets pass each other in a more realistic looking way. Harder to notice the effects of 1/r when it only affects the moon of a planet, since none of the planets have more than one moon so you can't really compare the effects.
so the 0G cave is artificial? that makes me a bit sad, I thought it was just the result of the planet being hollow
Technically, not _every_ body is simulated directly. A few are scripted, like the you-know-what around the sun (its _debris_ is orbitally simulated though), as well as the DLC area... and I thought that the Hourglass Twins were scripted _with respect to each other_ (they'll always be an exact distance apart), while only their _combined midpoint_ was subject to orbital physics?
Also fun experiment in vanilla: Fly far, fly fast, and look back via map view once you're at least 25,000km out. Deep Space Kraken has come to play!
Is the you-know-what the sun station, because what makes you say it's scripted? Looking at it I see no way in which it differs from any of the other orbiting bodies. The Twins are a bit weird in that they orbit each other (with forces simulated) and then the force from the sun felt by their combined midpoint is applied to them. The Stranger and Dreamworld are unique yeah in that neither experiences gravity from the sun and the Stranger is scripted to start accelerating away.
The floating point error deep space kraken stuff is fun I agree!
Does the quantum moon gravity effect those balls in the museum?
@@emosh598 no, its gravity does not extend past its clouds!
Someone able to develop a mod, understand and implement the math of astrophysics and explain it in a simple yet entertaining way? Count me in!
Side note, you seem to have a slightly monotone voice when you record. I can't say it's a fault but that's the only flaw I was able to find in your content.
Love your style anyway! ^^
Thank you! but i do be a monotone kind of guy
WHAT?
It's the Attlerock, not the Atterlock!?!?!?!?
My whole life was a lie :(
I love your modding videos! Would you be able to illuminate the eye of the universe so we can see the whole model in game?
Good idea, I tried something similar in a mod I made for a Lore Explorer video (see his Scale Showcase Mod video) but for some reason when he went to record the eye stopped appearing so it didnt make it into the video. Its definitely something I could consider revisiting.
"You can see why they chose to not have the game play like this"
Hm... no, actually I can't. What am I supposed to notice? Are the planets too far apart or something?
hmm yeah lol looking back at that part of the video i dont really let it play out enough, but everything runs into each other and falls into the sun or gets kicked out of the solar system, you can at least see that the twins have completely separated from each other too
@@xen-42 Ah yeah, that's what happened in most of my n-body simulations too! I just didn't see it happen in the video so I assumed you meant something else. Thanks for the reply!
Nuh uh it's real
u from the alternate reality where gravity is over r???
@@xen-42 Narp I am real
I thought you got pulled off of the Interloper due to tidal force.
Yeah, but those tidal forces from Giants Deep only affect the player (and the scout and the ship) and don't affect the Interloper, else it would dramatically affect it's orbit!
great video but it felt really confusing to know when exactly the simulation was considering inverse law or inverse square law, maybe adding a visual cue or just specifying exactly what planets/stars/comets followed which law in the original game and in the simulation. I thought at first all the cosmic objects had inverse gravity in the original game, but I kinda got lost at the part you showed attlerock, like, did it follow 1/r² originally?
Also, there's a easier way of showing why v = sqrt(GM) in inverse law case, simply taking GMm/r = mv²/r (centripetal force)
Everything has inverse gravity except the sun and the attlerock which have inverse square. Fair point about the derivation potentially being easier, I originally did the math to account for elliptical orbits as well hence the complicated math but simplified it down to circular for the video
Fun fact you can break the laws of physics But I'm not gonna tell you how
aw man
"what's ln of 1"😂😂 I say s*** like that all the time
how does gravity work on the Stranger? I think the "gravity" there is MUCH too strong for the rotational speeds they're using and it suddenly cuts in when you get close to a surface you're supposed to be allowed to land/stand on
Yeah you're absolutely right, they fake it and add in some extra gravity in a cylindrical volume that pushes u outwards. The amount of artificial gravity the spin gives is barely enough to stop you just flying everywhere.
the ATP centrifuge is real though, right? I don't remember anything weird when I despun myself with the jetpack
ive been wondering, does the... ahem... *stranger* planet get affected by the sun's gravity at all?
Nope, plus when you enter into its cloaking field there's also a zero-gravity volume that prevents you from being affected by the sun's gravity anymore too. Else you'd just get pulled out of it and sucked into the sun. If you leave your ship floating inside the field, when the Stranger starts to leave midway through the loop your ship will fall out of the zero-gravity field and get sucked into the sun.
Hold the fuck up. Elliptical orbits are not possible* with F=1/r, only with F=1/r^2 and F=1/r^3which is why you cannot orbit any of the planets. Now what I am stuck on is the hourglass twins. These can be reduced to a 2D Lagrangian problem, masses connected by a spring, but what fucks me up is that they seem to remain at their equilibrium separation even after the perturbation begins and one of them starts gaining mass. Can you answer this? Are those two on rails as a singular object at the center of mass? Is there some trick where the sand is weightless? Thanks
the twins don't gain or lose mass. stand on either one, look at the g's and you'll see it doesn't change. the sand doesn't add mass.
@@greenyxd7298 I mean, the Nomai said the Twins behave like a one object after all
Do you have example code of how to get the gravity area to work? I created a rigid body player and a planet with a gravity field, but the player doesn’t move toward the planet. I set the player gravity scale to 1 and the gravity field to priority 1
Plenty of example code in New Horizons github.com/Outer-Wilds-New-Horizons/new-horizons
wow the outuer wilds is yout playground
its a ton of fun to mess around in!
@@xen-42 yeah i go into it all the time
How did you calculate the escape velocity for gravity which is inverse with the distance?
ln(infinity) = infinity, so Fdx with x from r to infinity is infinity.
Also, there' was a typo in your escape velocity for normal gravity, it's v = sqrt(2GM/r) (but that's just a nitpick)
Oh nah what I showed was meant to be circular orbit velocity, should've made that more clear in the video. You make an interesting point though, I hadn't realized that the integral just doesn't converge. So mathematically there is no such thing as an escape velocity when orbiting any of the planets in Outer Wilds, although mechanically the devs just cheat that and have a strict cutoff radius where you stop feeling the effects of gravity.
If you still want the derivation for the non-escape orbital velocity I uploaded it over here: github.com/xen-42/xen-42/blob/main/Linear_gravity.pdf
@@xen-42 Thanks, that was exactly what I was looking for.
Taking the limit when going from (9) to (10) was non-trivial. Had to use L'Hôpital's rule and calculate myself that lim (x->a) (ln(x/a)(x²/x²-a²)) was indeed 1/2. Was fun, though :)
Didn't expect the outcome to be so simple and elegant and resemble the escape velocity of normal gravity, even though we calculated something else.
So does this mean that you can deorbit the planets if you use your ship thrusters?
Nah, each planet is only effected by the planet/sun that it orbits. Some are affected by water which is a weird one, so the Interloper can get stuck inside Giants Deep if it were to collide with its orbit because its physics controllers detect the water (it uses this to detect entering the sun, since water and the sun are both handled by a "fluid detector" in the game's code it ends up also detecting water as a side effect). Unity (at least the version Outer Wilds was made in) doesn't have built in physics support that would allow the player to push planets unless the devs were to recreate all the planet collision models (it only works for convex meshes), else it'd be easy to just change a variable and allow the player to deorbit any planet.
Smh my head, outer wilds can't even simulate real gravity, truly ruined the immersion, 0/10 game, I wish I had never played this game
So tht I can play it again
Every video game does this.....
does gravity as 1/r instead of 1/r^2? thats news to me!
Suxh a freat video lol!
🤓
how dare you comment this i will NEVER make another video EVER AGAIN lest i be called a NERD EMOJI
"Why didn't they just do This?" Changes this. Then procedes to show WHY "this" wouldn't have worked. This was a bit of the most useless video I've seen. I know I'll watch ANY thing on outer wilds... this video proves why that needs to stop.
Hey I've never had a hate comment before, cheers!
@xen-42 I'm sorry you took this as a hate comment. I'd like to rescind any hate you felt from me, it wasn't intended. It was negative, sure. But the things you complained about, like the physics, are some of the best things in the game to me. So i commented what i did. I apologize if you took that as hate though, it wasn't. I've seen multiple of your videos and thumbs up and commented positively before. I appreciate the time you took to create your video. Thanks for sharing outer wilds content. I hope you can re read my comment hear my apathy and confusion now, instead of hate.
@@GrandNoble I see. I think you've misunderstood me though, because I'm not complaining about the physics, I'm just saying that the gravity is inaccurate to real life physics. Anyway, shame to hear you didn't like the video. Cheers.