"I can probably do it better" every programmer, hell, every person creating something says this and immediately regrets their decision until it's complete
I remember that I was still with only 2 years of experience on Roblox Lua when I saw that time's best vehicle module. I wasted a month or two trying to make a better one just to abandon it and realise that I can do it now, but if I attempt it's not even going to benefit me in any way.
You could probably achieve more stability if you could somehow make the ball leave an imprint of its momentum on any object it comes into contact with, on the understanding that contact between two objects must achieve a null state before contact is broken (like with objects that are artificially forced into place, like walls that hang in the air, returning equivalent negative values). In nature, we are supposed to subtract the influence of the struck object(s) from the original trajectory of the striking object to find its path through space. If we don't, the moving object inherits that extra energy of being struck back by the wall and internalizes it, without losing any of its original velocity, causing the structure to eventually warble out of control the longer it moves. I don't know how it works in code, though.
I did physics sim stuff too (and want to do more) and it seems like our brains are of similar smoothness, so I understand your pain. It's comforting to see someone else struggle with the same stuff.
Can only agree! I'm working with an advanced optics sim at the moment and some days I just wish I could close my eyes and forget my existence for a little while... But when it works, then it's all worth it again!
I think you need some damping/friction forces. It looks like the springs in the ball are perfectly elastic right now, so each impact will just increase the internal pressure, causing each triangular section to oscillate more and more (like simple harmonic motion, but complicated since it's in a connected mesh of springs). Dissipating the energy as "heat" should keep the ball's internal pressure more stable and prevent the triangular sections from vibrating faster and faster.
#1) There seems to be not enough decay to the energy in the springs #2) The Energy is stored in the single springs and not dissipated to the other springs, allowing for resonance. And as those who resonate more will re-collide earlier, there is a positive feedback loop to this resonance.
imagine having to code all the quantum particles and then atoms and physics and complexity of this universe, those beings may as well be gods compared to us
Disclaimer: I know very little about numerical integration. But maybe using a symplectic integrator (e.g. velocity verlet) instead of Runge-Kutta would help with the oscillations you're getting.
@@Gonkee Although RK4 is a big improvement on Euler method, and is good enough most of the time, it still can be improved (usually in the cost of computing effort and complexity of the algorithm). Specifically, at 7:10, as a result of increasing the spring coefficient to 999999, the solution probably now have very high gradients and your time-step is probably not small enough to capture it and the solution completely diverges. In the Numeric Analysis world, we say the problem/equation is "stiff". If you want to somewhat keep the simplicity of the integrator, you might be interested in the adaptive RK45 method. If you are willing to invest in more complex methods, maybe consider researching into implicit RK methods. This is the first time I hear about the Verlet's method proposed above, but maybe it's also a good fit. Disclaimer: I never used implicit RK methods, but I heard from a professor that they are used to solve stiff problem.
@@Gonkee i think the problem is that you're trying to solve a set of partial differential equations by treating each PDE as a separate ODE. im by no means an expert but afaik there are algorithms designed for solving PDEs that are much more stable
Hey Gonkee, I came across your video some time ago and it inspired me so much. I loved seeing your entire raw process, struggling and learning an entire language from scratch. It gave me a huge motivation boost and made me realise that it's possible to pull off something as crazy as this through sheer will. Thanks a lot for this. You're a genius.
Im in my first year in college studying software engineering and I thought chemistry was gonna be useless to me.. but I now saw you using the ideal gas formula and now I’m motivated to study 👍 thanks, and great video!
Thank you Gonkee. You have made all the pain of teaching physics worth it. I am thinking that I can use your video as motivation for my students. Would be great for the computational physics class that I taught a couple years ago. I do feel a bit guilty, that I can't go help fix your code right now. I'm sure you'll figure it out...you are the best type of student. Very proud of you young man.
the points on the sphere accumulate kinetic energy when time passes with no way to get rid of it (that's why it might go haywire or "explode"). in the real world a ball in air and earth has more than 100 ways to get rid of this energy through collision and contacting. it's like the ball gets refreshed each time it does something but still hold on to its positional value and it's center of mass general velocity vector's direction. hope this physics fact inspire you.
You got style, patience and brains man. Keep it up. I’m not even in this field but watching you think through the problems really made me smile. You’ll go far.
Nice video! Yeah Runge-Kutta 4... when I learned this, it was like cake and roses. Joking, it was like pain in my pineal gland. But I managed to get it working. It actually took me about a half a year to actually understand the papers of Runge-Kutta 4 and how to advance it. I mean you are taking one second and dividing it to pieces and then summarising the pieced simulations together and passing as a single second... But it sure does make a huge difference! I integrated it to my vehicle simulation and I actually had to rewrite about 90% of the simulation, just to make it compatible and there still are some minor issues, just like you are having haha. Maybe you can advance your soft body to a metal simulation, by adding some fading dampening, you could make a seriously cool damage system. Fluids are the hardest, would love to see how you approach it.
1. Fix: (Not easy to implement) When solving the harmonic oscillators (these spring mass thingies in the icosphere) with an explicit scheme like RK4, the systems energy increases from numerical errors until it diverges to infinity. When solving with an implicit scheme (much harder to implement), the systems energy converges to zero and is therefore to be preferred. 2. Fix: (Turbo easy to implement) I assume You did not implement a damping term. In reality, a pendulum or an oscillator never oscillates for ever, it loses energy due to friction and shit. If You include a small damping term into the oscillators You will see iz gunna wok. 3. Fix: (Bruh) Just decrease the time step size. The RK4 scheme should do better at 120Hz or 240Hz. He actually implemented a daming term in 6:07 The parameter 'dampingConstant' is probably set too low - much too low
@@madmax43v3r Yep: If the step size is too large on an explicit scheme, everything goes boom. There are situations where a damping term generates a converging solution.
@@TheRainHarvester In this case: The machine error is orders of magnitudes smaller than then error of the integration scheme. => Doubles would probably not make a significant difference. In Your case: The major source of errors was the machine accuracy for small float numbers. Hence, double precision solves the issue.
The ball exploded because the time step wasn’t small enough to equilibrate/mesh grid wasn’t small enough. To get it to not blow up you can decrease time step to small small Increase nodes Adjust the equilibrium position on the springs Alternatively you can use an FOV average to assign values to your ball as opposed to using a spring mass system.
"How many of the average person's fart gasses are needed to make the pressure in an ico sphere go from 1 atmospheric pressure to 2 atmospheric pressure?"
You have an awesome and interesting way of communicating your ideas and experience. I would love to see some NN, or other machine learning subjects tackled by you.
This is amazing! (I also want to make my own physics engine :)) Also, it's really cool how your new channel is *already* more popular than your old one :O
Nice one bro. That's an awesome project. During lockdown, I decided to make a program that tries to predict which stocks are going up and down. Programmers be crazy...
Bro, ur insane. I love that type of humor so much, the editing is well paced and those random images there and here alongside all the dumb memes, that shit is insane. Keep going on, the content is insane as well man
It's truly a joy to see others go through the pain I'm subjecting myself to, okay that sounds cruel, but at least it makes me feel better! You did a great job though, I'm looking forward to seeing more simulation stuff (and I would unironically watch 2h of this)
As a fellow sleep deprived math enthusiast and programmer who also likes to inflict pain upon himself with impossible projects; this video and your channel as a whole is great. Thank you "Inspecto" :)
Friction and dampening, my friend, that is the key to stabilizing. There are multiple different kinds, and often they are small enough to be dismissed. But whether it is air resistance or the springiness being dampened by the rubber converting motion into heat, or whatever, it is usually some variation of a force (/acceleration) in the opposite direction of the velocity, and is some proportion of the velocity or the velocity squared. So it will always attempt to counter the velocity, and bringing the system towards a stop (and gravity in this case keeps adding to the motion of the system). And since the velocity becomes slower and slower, the friction/dampening proportional to the velocity becomes lower and lower, thereby making the slowdown exponentially decreasing (sort of asymptomatic towards stand still).
6:45 - it's not the accuracy of explicit methods that's bad - it's instability. You raise your dt too high - you get an unstable system. And Runge-Kutta is excplicit too, so you'll inevitably will bump into problems with stability when you crank certain numbers too high or too low. Also, if your spring constant is very high - it's better in terms of calculation to implement them not as springs, but as a rigid rods. Also, you can add some dampening to your system to "vent" some instabilities that accumulate over time. I'm basically trying to accurately simulate problems like that (inflatable shells dynamics) and it's a living hell (But i'm going to use FEA since I don't need any real-time simulation).
@@johnsherfey3675 it's not just rounding. But he could use double floats. Sometimes it's the approach and departure i describe in my micro gravity simulation.
Good job! All you are missing is surface tension, a surface tangent force between vertices to distribute them homogeneous, in these case a spring like force pointing from vertice to its neighboring vertice on the mesh.
8:54 I have an idea was is going on: could it be that the precision of your coordinates is getting worse the more your object moves away from the initial point? I remember watching a guy developing a fps engine and he had the same problem, that when he walked around in the world at some point the objects were getting "wobbly".. then he realized that the better approach is, freezing the player character at 0,0,0 and just moving the world around him. It's not solving your problem at all (unless your blob is THE central object) but could be an explanation..
When you have the issue of the mass exploding, it is likely because the spring mass system of the ball is a stiff system, although RK4 will increase the accuracy of the system (thus making it harder to explode) we can reduce the dependence on the systems spectral radius by using an implicit method. Backward Euler is very simple, but will work better than even RK4 on a stiff equation. There are many higher order implicit methods for differential equations if that is what is desired as well.
One downside to the ico sphere, you don't get fine control over the resolution of it, if you want to increase the number of vertices, you have to subdivide it. Whereas UV and square spheres can have X by Y number of vertices, which is better for things like spherical terrain.
Dude what you did is amazing. Especially knowing what you know in the amount of time you did it in. This is really impressive! The most I can do is draw a square. Actually, I can't even do that anymore because I forgot all my high school Java. Keep it up! You will be big brain one day! (youre already big brain tho
I once made a cloth simulator (spring-mass system) and it had similar instability with ramping oscillation. It became a lot more stable when I introduced some sort of fake friction (just multiplied the velocity of each point by 0.99 every time step).
You might want to use implicit methods. The common technique in Physics-Based animations is Backward Euler. This would be much more stable. If you really want to keep using explicit methods, make sure your time step is low enough and maybe add some damping. David Levin from UoT has uploaded superb videos for course on his youtube account. You should be able to learn everything you need to implement the other parts of your engine. PBD might have been a nice alternative way to make everything fit seamlessly in the same simulator, but I have personally made it work with FEM.
awesome video!!! Maybe to fix the ball you can add some sort of communication between adjacent points so that the small oscilations dont get out of hand?
I just completed a raytracer for a class, and the file that calculated intersections and collisions ended up around 900 lines long. I now have PTSD flashbacks whenever someone mentions collision 😭
You could fix the ball by clamping the values so the points cant move past the center of the ball. You can calculate the center by averaging the position of several points for more accuracy. You can also clamp forces in other ways. In real life, this is like heat and friction losses. Most things dont perfectly transfer or transform energy. You can clamp the max spring length and cap the forces a little bit. You can use the length multiplied by a small constant to make the effect gradual, also forcibly clamping the positions of points to be on the right side of center, basically physical normal mapping, can all work together to make an object feel more solid and not gain energy incorrectly. So it feels solid, heavy, and realistic, and absorbs energy sometimes.
Looks like you could add a "damping" term to your Mass-Spring modeling for the tensile forces. It's a fairly common system. Instead of being a Mass-Spring system, it becomes a Mass-Spring-Damper system (sometimes referred to as MSD). The damping term creates a force opposing the direction of travel. The damping term is c*x_dot (damping coefficient [constant] * velocity of mass). So as the mass (vertice in this case) begins traveling at a high speed, a large oppose force (created by the damping term) begins to get generated, which will cause the mass to decelerate. If you implement a damping term to your physics engine, you might find that you no longer have the ball shaking itself to death! I might try and do this for you if I have some time to take a look at the code. Cool video regardless! :)
Be a badass and program your own tcad and spice software. You could make it big. There's already some open source. You just want to have a complete tcad and/or spice together that's competitive with the commercial stuff, then you make a big break through in universities by getting lots of grad students, postdocs to use the stuff since they can program their own models. Like Oliphant did numpy. You'd also wan to make sure to have thermals sims possible too. Separately a matching electromagnetic wave simulator with its own spice and thermal sims is a whole second project. Not sure it's yet possible to have combined electromagnetic wave studies and tcad level modeling for semiconductors.
Maybe try using fewer triangles to construct the icosphere? I would also include some dissipative forces in addition to Hooke forces if you haven't already.
What's great about programming in Java is that you can use ASM to inject code wherever you want in any part of the program. Who needs to write from scratch when you can just get admin access and edit stuff?
The strings between points not lose energer through time maybe that's the reason why your ball begin being "shaky". Every bounces add more energy to your bounding system, and the ball can explode due to too much accumulation. The string are like vibrating, go look forward energy dissipation, like dampeners inside a vehicule. Great Job with programming !
Physics mayor here, you basically implemented a kind of not very stable bubble, not a boucy ball! The ball may be stimulated more accurately by defining a 3d mesh (as if it was a cristal lattice of some sort even if it's not) and adding "spring forces" to each one, and not only with the first neighboring points, but to the 2nds and 3ds also. I would love to see the behaviour then! I loved the video, keep up the good work!
OMG, I've just started coding two months ago. First time seeing something like this! Being a physics lover, I want to turn my physics knowledge into something like this. I subscribed to learn moreee, can't waittttttttttttttttt....
Ah, the programmer's credo:
"We do things not because they are easy, but because we thought they were going to be easy."
ON POINT
Not the programmers only👍
Yeah that's about everything in a programmer's brain
Then we see thousands of mistakes
@@tex1297 literally just anyone working with stuff that involves a large amount of math
@@filipthunell8631 right +any creative people of any field
imagine how many cool libraries we wouldn't have if the authors had realized how much effort it would be
"I can probably do it better" every programmer, hell, every person creating something says this and immediately regrets their decision until it's complete
It's the only thing that keeps world from not using 500 year old software lol
Instead, we have e.g. JS frameworks mess, being flooded by D-K syndrome folks on one hand and driven by shiny object syndrome folks on the other :P
I remember that I was still with only 2 years of experience on Roblox Lua when I saw that time's best vehicle module. I wasted a month or two trying to make a better one just to abandon it and realise that I can do it now, but if I attempt it's not even going to benefit me in any way.
Except that's how innovation is done
@@dimaryk11 not always but sometimes it does result in innovation.
I was cracking up at 8:53. So that's how a bouncy ball behaves when it exists in an atmosphere that somehow lacks entropy.
You could probably achieve more stability if you could somehow make the ball leave an imprint of its momentum on any object it comes into contact with, on the understanding that contact between two objects must achieve a null state before contact is broken (like with objects that are artificially forced into place, like walls that hang in the air, returning equivalent negative values). In nature, we are supposed to subtract the influence of the struck object(s) from the original trajectory of the striking object to find its path through space. If we don't, the moving object inherits that extra energy of being struck back by the wall and internalizes it, without losing any of its original velocity, causing the structure to eventually warble out of control the longer it moves.
I don't know how it works in code, though.
Oh right. So his sphere heated up.
@@matthewscott336 that's the best way to say it
I did physics sim stuff too (and want to do more) and it seems like our brains are of similar smoothness, so I understand your pain. It's comforting to see someone else struggle with the same stuff.
Can only agree! I'm working with an advanced optics sim at the moment and some days I just wish I could close my eyes and forget my existence for a little while... But when it works, then it's all worth it again!
i dont even code, im a civil engineering student
just come here to see people in pain and having fun
ahhhhh one of my kind
Smart minds think alike
Subbed
8:10 why does the ball look like the coronavirus 💀
I think you need some damping/friction forces. It looks like the springs in the ball are perfectly elastic right now, so each impact will just increase the internal pressure, causing each triangular section to oscillate more and more (like simple harmonic motion, but complicated since it's in a connected mesh of springs). Dissipating the energy as "heat" should keep the ball's internal pressure more stable and prevent the triangular sections from vibrating faster and faster.
At high enough friction/viscousity this ball will behave like a solid object, by the way.
@@blinded6502 with explicit integration it wont ever be able to use a high enough viscosity/stiffness to make it behave like a rigid body.
Yeah I think they should have some friction relatively to the other points. If they go against the stream then friction. Have fun! xD
@@gaussgreen7921 Of course. Unless you advance simulation by very tiny time intervals, and possibly even increase precision of the coordinate system.
Yeah also the imperfections of solving the differential equations numerically could be causing a feedback loop of growing energy in the springs
3:07
Hexagons, are the bestagons
When I first read the title I thought it said "Coding an entire physics engine IN scratch", that would have been insane
Omg same so now i sad
oooauaiaiaiaiauauauaauauauauau how to unread things
@Biglyp people have made 3d stuff in scratch
@@katteisace4563 yeah buts it’s not true 3D
same
That epic montage was amazing, it captures every aspect of "learning" a new skill :´)
#1) There seems to be not enough decay to the energy in the springs
#2) The Energy is stored in the single springs and not dissipated to the other springs, allowing for resonance. And as those who resonate more will re-collide earlier, there is a positive feedback loop to this resonance.
Ok science boy
I am not sure about this but do you adjust the area in the force formula, according with the mesh count?
It just might be that there’s no damping at all
nerd
when are you gonna upload on your yt channel again? your songs are very nice
This dude's got some hilarious editing. Keep it up!
lets all just take a moment to aprecciate the devs of the simulation we live in
Thanks 😊 . I will not delete your soul until you are 90 year old
@@motherisape can you boost me so that I look like I'm 50 when I'm 90?
@@motherisape Yo, Tell the guy who created my code that his choice of skill point distribution sucks.
imagine having to code all the quantum particles and then atoms and physics and complexity of this universe, those beings may as well be gods compared to us
@@casualbydefault my bad, we accidentally used an older version of the skill point distribution algorithm when we made you
I’m a business major and I still enjoyed the video. Good humor & editing, keep it up!
Disclaimer: I know very little about numerical integration. But maybe using a symplectic integrator (e.g. velocity verlet) instead of Runge-Kutta would help with the oscillations you're getting.
What, I thought RK4 was meant to be the best one
@@Gonkee Depends on how you define "best", there's always pros and cons
@@Gonkee Also there are symplectic versions of Runge-Kutta fwiw
@@Gonkee Although RK4 is a big improvement on Euler method, and is good enough most of the time, it still can be improved (usually in the cost of computing effort and complexity of the algorithm).
Specifically, at 7:10, as a result of increasing the spring coefficient to 999999, the solution probably now have very high gradients and your time-step is probably not small enough to capture it and the solution completely diverges. In the Numeric Analysis world, we say the problem/equation is "stiff".
If you want to somewhat keep the simplicity of the integrator, you might be interested in the adaptive RK45 method. If you are willing to invest in more complex methods, maybe consider researching into implicit RK methods.
This is the first time I hear about the Verlet's method proposed above, but maybe it's also a good fit.
Disclaimer: I never used implicit RK methods, but I heard from a professor that they are used to solve stiff problem.
@@Gonkee i think the problem is that you're trying to solve a set of partial differential equations by treating each PDE as a separate ODE. im by no means an expert but afaik there are algorithms designed for solving PDEs that are much more stable
Hey Gonkee, I came across your video some time ago and it inspired me so much. I loved seeing your entire raw process, struggling and learning an entire language from scratch. It gave me a huge motivation boost and made me realise that it's possible to pull off something as crazy as this through sheer will. Thanks a lot for this. You're a genius.
Amazing video. I love doing a funny and unnecessary project that takes part of my soul while I'm making it. Subscribed
I feel your pain :) You still managed to code an awesome demo 💪
Im in my first year in college studying software engineering and I thought chemistry was gonna be useless to me.. but I now saw you using the ideal gas formula and now I’m motivated to study 👍 thanks, and great video!
You are new, to this. So I make an exception. The first rule of thermodynamics is: Noone talks about thermodynamics.
Good stuff man. I'd find some Discord group with coders that geek out on physics and ask around. In the mean time, your video style is great.
boredom really brings you to a journey of pain and suffering. amazing video as always man, i'm loving this format so far :D
“it’s 2 weeks later, and i’ve probably lost 5 years off of my lifespan”
ah yes, mathematics
Thank you Gonkee. You have made all the pain of teaching physics worth it. I am thinking that I can use your video as motivation for my students. Would be great for the computational physics class that I taught a couple years ago. I do feel a bit guilty, that I can't go help fix your code right now. I'm sure you'll figure it out...you are the best type of student. Very proud of you young man.
the points on the sphere accumulate kinetic energy when time passes with no way to get rid of it (that's why it might go haywire or "explode"). in the real world a ball in air and earth has more than 100 ways to get rid of this energy through collision and contacting. it's like the ball gets refreshed each time it does something but still hold on to its positional value and it's center of mass general velocity vector's direction. hope this physics fact inspire you.
You got style, patience and brains man. Keep it up. I’m not even in this field but watching you think through the problems really made me smile. You’ll go far.
3:20 "how to make an iron golem"
what a gamer
Gonkee: "I'm coding an entire physics engine from scratch"
Me: "You poor poor soul"
Nice video! Yeah Runge-Kutta 4... when I learned this, it was like cake and roses. Joking, it was like pain in my pineal gland. But I managed to get it working. It actually took me about a half a year to actually understand the papers of Runge-Kutta 4 and how to advance it. I mean you are taking one second and dividing it to pieces and then summarising the pieced simulations together and passing as a single second... But it sure does make a huge difference! I integrated it to my vehicle simulation and I actually had to rewrite about 90% of the simulation, just to make it compatible and there still are some minor issues, just like you are having haha. Maybe you can advance your soft body to a metal simulation, by adding some fading dampening, you could make a seriously cool damage system. Fluids are the hardest, would love to see how you approach it.
1. Fix: (Not easy to implement)
When solving the harmonic oscillators (these spring mass thingies in the icosphere) with an explicit scheme like RK4, the systems energy increases from numerical errors until it diverges to infinity. When solving with an implicit scheme (much harder to implement), the systems energy converges to zero and is therefore to be preferred.
2. Fix: (Turbo easy to implement)
I assume You did not implement a damping term. In reality, a pendulum or an oscillator never oscillates for ever, it loses energy due to friction and shit. If You include a small damping term into the oscillators You will see iz gunna wok.
3. Fix: (Bruh)
Just decrease the time step size. The RK4 scheme should do better at 120Hz or 240Hz.
He actually implemented a daming term in 6:07
The parameter 'dampingConstant' is probably set too low - much too low
A high damping coefficient will make it go boom too, the more force is generated the smaller your timestep has to be to keep it stable.
@@madmax43v3r Yep:
If the step size is too large on an explicit scheme, everything goes boom.
There are situations where a damping term generates a converging solution.
Would using double floats help? (It helped on my micro gravity videos... Kept 1/r2 from going to infinity at smaller proton scales.)
@@TheRainHarvester In this case:
The machine error is orders of magnitudes smaller than then error of the integration scheme.
=> Doubles would probably not make a significant difference.
In Your case:
The major source of errors was the machine accuracy for small float numbers. Hence, double precision solves the issue.
@@Xbox360SlimFan I'm dealing with integration errors now: unequal approach and departure simulation ticks.
dude. great video. You're a sadist...and I love it, because this is educational and fun (and super funny (I LOL'd a ton))! Also subbed
7:40 as someone who took numerical integration classes from physicists, good luck with that : p
Oh no you make it sound like it'll be a terrible experience
@@Gonkee It will.
@@Gonkee to be honest, it's kind of a coin-flip in my experience, but when it's bad it's really bad
I’ve stumbled upon your channel & have never been happier. 1 minute into this video and you have me cracking up
I laughed my ass of. I don't care how good or bad this actually is, it's the funniest shit watching
Is it just me or is that comment T H I C..
Dani: no not now I’m working on karlson
CREEPER MAN
@@Jake2534 yoooo
The ball exploded because the time step wasn’t small enough to equilibrate/mesh grid wasn’t small enough.
To get it to not blow up you can decrease time step to small small
Increase nodes
Adjust the equilibrium position on the springs
Alternatively you can use an FOV average to assign values to your ball as opposed to using a spring mass system.
Or you can also implement a mesh that changes over time
I like how his equation for the ideal gas constant has all the capital letters for FART in the right order at 5:35
"How many of the average person's fart gasses are needed to make the pressure in an ico sphere go from 1 atmospheric pressure to 2 atmospheric pressure?"
@@xXJ4FARGAMERXx a lot
damn
@@xXJ4FARGAMERXx that is a genius question. I don’t know why the worlds best scientists aren’t working on it now. Or maybe they are…
@@xXJ4FARGAMERXxhaha text in between username go brrrrr
First time watching your video, you have done a remarkable job. I'm sure your channel will stand out as it is already.
You have an awesome and interesting way of communicating your ideas and experience. I would love to see some NN, or other machine learning subjects tackled by you.
This is amazing! (I also want to make my own physics engine :)) Also, it's really cool how your new channel is *already* more popular than your old one :O
This makes me feel way better about my spring-mass system I made. It ALSO was sensitive to the parameters and exploded in a very similar way :)
Nice one bro. That's an awesome project. During lockdown, I decided to make a program that tries to predict which stocks are going up and down. Programmers be crazy...
Can’t wait to see the next update video! It looks like a really cool project.
Dude, you are hilarious and I have so much respect for you doing this!
Your humor is spot on and your content quite unique. Subscribed.
Vids like this motivated me to code,thanks buddy
As a CGI Artist I have to say you made it insanely good! Nice work because you did it alone and didn't knew stuff about it before.
Bro, ur insane. I love that type of humor so much, the editing is well paced and those random images there and here alongside all the dumb memes, that shit is insane. Keep going on, the content is insane as well man
It's truly a joy to see others go through the pain I'm subjecting myself to, okay that sounds cruel, but at least it makes me feel better! You did a great job though, I'm looking forward to seeing more simulation stuff (and I would unironically watch 2h of this)
Never saw you before, I'm 4 minutes into the video, already subscribed. Keep up the good work man ✌️
Your style of comedy is amazing. Time to binge yet another creator's entire existence in just a few days.
nah the kid is pretty cringe i'm glad i'm using sponsorblock to skip his fillers
@@moonshot3159 ironic, 'cause calling people cringe for doing what they love and sharing it with others is pretty cringe if you ask me.
@@Lin_The_Cat_ what you just said has earned you a ticket to cringetown.
Your videos are SO funny!! You capture all the frustration I am feeling but make me laugh while watching 😁💀👍
As a fellow sleep deprived math enthusiast and programmer who also likes to inflict pain upon himself with impossible projects; this video and your channel as a whole is great. Thank you "Inspecto" :)
It looks like the sphere is trying hard to contain all the heat it gained from colliding with the walls
Friction and dampening, my friend, that is the key to stabilizing. There are multiple different kinds, and often they are small enough to be dismissed. But whether it is air resistance or the springiness being dampened by the rubber converting motion into heat, or whatever, it is usually some variation of a force (/acceleration) in the opposite direction of the velocity, and is some proportion of the velocity or the velocity squared. So it will always attempt to counter the velocity, and bringing the system towards a stop (and gravity in this case keeps adding to the motion of the system). And since the velocity becomes slower and slower, the friction/dampening proportional to the velocity becomes lower and lower, thereby making the slowdown exponentially decreasing (sort of asymptomatic towards stand still).
when a triangle decrease in size, the amount of force that it gets gets lower, and the points touch each other... is that why everything colapse ?
6:45 - it's not the accuracy of explicit methods that's bad - it's instability. You raise your dt too high - you get an unstable system. And Runge-Kutta is excplicit too, so you'll inevitably will bump into problems with stability when you crank certain numbers too high or too low. Also, if your spring constant is very high - it's better in terms of calculation to implement them not as springs, but as a rigid rods. Also, you can add some dampening to your system to "vent" some instabilities that accumulate over time. I'm basically trying to accurately simulate problems like that (inflatable shells dynamics) and it's a living hell (But i'm going to use FEA since I don't need any real-time simulation).
Basically you're avoiding at some level rounding errors that can occure.
@@johnsherfey3675 it's not just rounding. But he could use double floats.
Sometimes it's the approach and departure i describe in my micro gravity simulation.
Thank you. You just took all my motivation learning to code.
Ah yes, C# and OpenGL, I regretted doing that every single time, yet I still keep doing it
Excellent sense of humor. Keep it up, buddy!
i'd also enjoy longer videos on stuff like this
Your code is so good it simulated material durability 🤣.
In all honesty that's actually fantastic
nice video gonkee. I studied 1984 in high school as well, not as my physics book though haha. keep it up
Wow what a coincidence I also studied 1984 in highschool
Wow! Same here. What a small world.
@@cardscook7721 No way! I too studied 1984 at high school.
@@JordanChung Bruh at this point these comments are looking like those scams in crypto videos "I too traded with mrs laura". lol🤣
Good job! All you are missing is surface tension, a surface tangent force between vertices to distribute them homogeneous, in these case a spring like force pointing from vertice to its neighboring vertice on the mesh.
ah yes, the reaction to the mathematics of collision, i know that one from experience
8:54 I have an idea was is going on: could it be that the precision of your coordinates is getting worse the more your object moves away from the initial point?
I remember watching a guy developing a fps engine and he had the same problem, that when he walked around in the world at some point the objects were getting "wobbly".. then he realized that the better approach is, freezing the player character at 0,0,0 and just moving the world around him. It's not solving your problem at all (unless your blob is THE central object) but could be an explanation..
7:30 - I wish I had another like button to press here.
When you have the issue of the mass exploding, it is likely because the spring mass system of the ball is a stiff system, although RK4 will increase the accuracy of the system (thus making it harder to explode) we can reduce the dependence on the systems spectral radius by using an implicit method. Backward Euler is very simple, but will work better than even RK4 on a stiff equation. There are many higher order implicit methods for differential equations if that is what is desired as well.
I should also note you may need to increase the damping of the springs if resonance is the problem
"Okay i don't know how to make any of these but im sure ill figure it out."
- Every programmer ever
awesome and entertaining video man! hope you the best
2:37 Fun fact. you can turn a quadritateral into 2 triangles with a simple line.
Then each triangle wouldn't be equilateral though.
8:28 the way the ball moves makes me feel incredibly uncomfortable no idea why
Do you have triophobia
One downside to the ico sphere, you don't get fine control over the resolution of it, if you want to increase the number of vertices, you have to subdivide it. Whereas UV and square spheres can have X by Y number of vertices, which is better for things like spherical terrain.
Dude what you did is amazing. Especially knowing what you know in the amount of time you did it in. This is really impressive! The most I can do is draw a square. Actually, I can't even do that anymore because I forgot all my high school Java. Keep it up! You will be big brain one day! (youre already big brain tho
I once made a cloth simulator (spring-mass system) and it had similar instability with ramping oscillation. It became a lot more stable when I introduced some sort of fake friction (just multiplied the velocity of each point by 0.99 every time step).
Ew, that's a horrible way to do it
@@FunnyFavor
I'm not saying he should do this exactly in his project, it's still insightful information though.
And for my project, it was perfect.
@@FunnyFavor i would do that
Can we post the result of your cloth simulator? 🤩
oh my god. youre the guy that did the awesome godot water shader tutorial. cool channel!!
now do it IN SCRATCH
A combo of two people I never thought I'd see
Dani and Dapz
Dapzi
You might want to use implicit methods. The common technique in Physics-Based animations is Backward Euler. This would be much more stable. If you really want to keep using explicit methods, make sure your time step is low enough and maybe add some damping. David Levin from UoT has uploaded superb videos for course on his youtube account. You should be able to learn everything you need to implement the other parts of your engine. PBD might have been a nice alternative way to make everything fit seamlessly in the same simulator, but I have personally made it work with FEM.
Why would that be more stable?
awesome video!!! Maybe to fix the ball you can add some sort of communication between adjacent points so that the small oscilations dont get out of hand?
Now do it IN scratch
Oh man I laughed so much at this video. I love your sense of humour. Subbed.
I just completed a raytracer for a class, and the file that calculated intersections and collisions ended up around 900 lines long. I now have PTSD flashbacks whenever someone mentions collision 😭
You should code an entire physics engine in scratch
Gonkee: *Makes physics simulation and talks about it.*
Also Gonkee: You know what, let's secretly rickroll the viewers!
You could fix the ball by clamping the values so the points cant move past the center of the ball. You can calculate the center by averaging the position of several points for more accuracy.
You can also clamp forces in other ways. In real life, this is like heat and friction losses. Most things dont perfectly transfer or transform energy.
You can clamp the max spring length and cap the forces a little bit. You can use the length multiplied by a small constant to make the effect gradual, also forcibly clamping the positions of points to be on the right side of center, basically physical normal mapping, can all work together to make an object feel more solid and not gain energy incorrectly. So it feels solid, heavy, and realistic, and absorbs energy sometimes.
1:59 yep I got rickrolled
Looks like you could add a "damping" term to your Mass-Spring modeling for the tensile forces. It's a fairly common system. Instead of being a Mass-Spring system, it becomes a Mass-Spring-Damper system (sometimes referred to as MSD). The damping term creates a force opposing the direction of travel. The damping term is c*x_dot (damping coefficient [constant] * velocity of mass). So as the mass (vertice in this case) begins traveling at a high speed, a large oppose force (created by the damping term) begins to get generated, which will cause the mass to decelerate. If you implement a damping term to your physics engine, you might find that you no longer have the ball shaking itself to death! I might try and do this for you if I have some time to take a look at the code. Cool video regardless! :)
"I have no redeeming qualities."
~The guy who has the skills to create an entire physics engine from scratch
Just... amazing! Thanks for your content.
Be a badass and program your own tcad and spice software. You could make it big. There's already some open source. You just want to have a complete tcad and/or spice together that's competitive with the commercial stuff, then you make a big break through in universities by getting lots of grad students, postdocs to use the stuff since they can program their own models. Like Oliphant did numpy. You'd also wan to make sure to have thermals sims possible too. Separately a matching electromagnetic wave simulator with its own spice and thermal sims is a whole second project. Not sure it's yet possible to have combined electromagnetic wave studies and tcad level modeling for semiconductors.
Maybe try using fewer triangles to construct the icosphere? I would also include some dissipative forces in addition to Hooke forces if you haven't already.
Great video and learning exercise. I’d recommend implementing a vector class so you can reduce your code by a factor of three!
What's great about programming in Java is that you can use ASM to inject code wherever you want in any part of the program. Who needs to write from scratch when you can just get admin access and edit stuff?
Oh wow, you made the github repo accessible thank you so much...
I can't wait to code this myself
I appreciate the use of the 'Winter is Coming' theme for vs
The strings between points not lose energer through time maybe that's the reason why your ball begin being "shaky". Every bounces add more energy to your bounding system, and the ball can explode due to too much accumulation. The string are like vibrating, go look forward energy dissipation, like dampeners inside a vehicule. Great Job with programming !
I'm in the 3:36 mark and I'm lucking my ass off! You have talent my friend!
Physics mayor here, you basically implemented a kind of not very stable bubble, not a boucy ball! The ball may be stimulated more accurately by defining a 3d mesh (as if it was a cristal lattice of some sort even if it's not) and adding "spring forces" to each one, and not only with the first neighboring points, but to the 2nds and 3ds also. I would love to see the behaviour then! I loved the video, keep up the good work!
this is the first video I see of yours, and I really love the content, you def deserve a sub
OMG, I've just started coding two months ago. First time seeing something like this! Being a physics lover, I want to turn my physics knowledge into something like this.
I subscribed to learn moreee, can't waittttttttttttttttt....
I love how you try but then it turns out to be hard. That’s the only way to learn your blind spots. Hope you can improve it in the future!