Can you pls make a site where you can play with these things, and if you do, put it in the description and reply to this comment, telling you created it? I WANT TO PLAY WITH THESE THINGS SO BAD
At some point, one of these structures will randomly be extremely stable and self replicating. (Maybe something with its outer wall which allows it to gather more of its kind). That would make the chaos go extinct....
at most of the videos I can see some 'cells' as I would like to call it, eat other particles or cells and it become too big to it splits into 2 cells... Kinda like how cells division work but without the chromosomes bullshits lol
The problem with this is that it is _too_ easy for multiple of the same structure to form from nothing, meaning that there is very little material left for any of them to make clones out of without first cannibalising something ( since they all use unique particles )
Yeah i feel like i could see that in a more complex environment the material in the outer layer of the “cell” could be attracted to the center layer with a middle layer seperating it, so that if the availability of the materials is right it could form a stable loop of getting big then collapsing in on itself in such a way that it replicates
1:58 first eukaryotic cell 3:44 cell walls form 4:08 complex life (and separate species) 5:26 many species (and racism) 5:34 mitosis 6:07 genuses start forming (similar species) 8:51 defense mechanisms arise (the blue "prey" cells start deflecting the red "predator" cells) 10:37 multicellular life emerges 11:14 cell nuclei form 12:52 filter feeders emerge 14:16 dna-like structure emerge edit: how did this get over 15 likes in under a day edit 2: some guy made a reply criticizing this so heres a disclaimer: i am not a scientist i made this comment for fun do not take any of this as something accurately explaining this still dont have to be so salty about it tho also the like count has gone up times ten when i last edited edit 3: Foreword: I am sorry if i misinterpreted anything you said, correct me if i am wrong because some thing sound more hostile than they should logically be from you. to c0dejjshizpostarchive624, i will call you Cody because I'm not gonna say all that, and to Mr_Tophatt, i have seen your argument and i have decided that bot of you are wrong. Cody, nobody would have believed that an all lowercase comment from some guy with the word meme in his name which features the word racism and uses the wrong term for similar species was going to be scientifically accurate, and to tophat, the is no way to "correct" a joke, but the lesser of the two evils is tophat, Cody says that there are no names, and that is true, except i was using those names to compare to the real things. and no this is not a strawman fallacy, i am simplifying what he said. I was comparing the simulation to real things in cellular life, and Cody, you said this was an "egregious misrepresentation" and that feels more like an insult than a regular saying something is incorrect for a good reason. and shut up about me misusing real scientific terms literally not another soul on this planet cares about that. Please stop blabbering about misrepresentation and actually respect me for taking time i will never get back to write a response to someone who will never read it., and looking for timestamps to compare to real life. I could have ignored you but i did not, and i hope you write back so i can understand a bit more about what you are doing. Arguments are supposed to be learned from, so i will leave some criticisms off so this won't go stale. Back to tophat, again, i was comparing the simulations to real life phenomena, and there shouldn't be "correct" and "incorrect" terms, Tophat said less so i say less about him. *_TL/DR:_* Both people who were arguing were wrong in some form. Both had shortcomings but they also had times of being correct. Both people did not win the argument, so i am making an edit. All in all, this comment was a joke and Cody took it too seriously but Tophat had some incorrect things too. Both were wrong and the argument will continue.
Although this is fun and all, I would like to remind everyone that everything in this comment is an egregious misrepresentation of these concepts. I don't believe that OP intended to pass this off as "real", but for those gullible enough, this is absolutely positively incorrect.
These kinds of simulations always spark so many ideas! What if you introduced gravity, each particle being attracted to a point at the centre inversely proportional to distance. What if you introduced energy, which affected the max speed of the particles? What if you grouped the particles in the beginning rather than randomly dispersed them? What if what if what if. This was beautiful. Thanks for putting it together
Gravity is not a point in space, but how much a mass bends the space or pulls. This game is already a gravity-pulling type simulation, but what should happen is that as the mass grows bigger, its pull should be bigger. I think that's what you meant.
@@aezravito9717 I meant what I said. When you program a simulation like this you can represent elements of reality as selectively and abstractly as you want. The current simulation has no direct or accurate representations of real world physics or phenomenon. I used the word “gravity” only to convey the idea. I think the version of gravity you propose would be very computationally expensive, but also very cool. It would be great to see that with an enormous world.
@@GerinoMorn I think that could be done for each particle by multiplying the color's "weights" with random numbers near 1 which would change the interaction but not so much so that it functions completely differently.
Now I wanna see this with the atomic scale now. Protons, neutrons, electrons, or the quarks making them up, then watch them as they show the different properties of gravity, electromagnetic properties, charges, changes in state of matter (solid to liquid, liquid to gas, gas to liquid, liquid to solid, etc.), radioactivity, tranparecy, conductivity, malleability, and more. Imagine how big of a simulation you need just for those things that are surprisingly 99% empty
@@elementgermaniumhere's the great thing about simulations though! You don't, you just need to know it's effects! This simulation you watched was simplified as hell, and CELLS formed!
They can simulate a human organ I think on the molecular level. One of my prof told that in a lecture 10 years ago. But they had research computation clusters. I bet its even more achievable today.
@@4984christian It depends on what you mean by "life"- In a simulation like this, I've personally seen simple "creatures" that eventually get so unstable they split into 3+, but more complex ones have a hard time It would be AMAZING to one day see spontaneous generation with our OWN EYES, computer or not!
really makes you think about sentience like, at what point do you go from mass to life? to brains? to concience? Edit: this discussion that has started in the replies is civil??? never seen this before.
I personnaly think that what makes us sentient is just the size of our brain, like at one point we became smarter that other animals and we gained concience and emotions to better understand, but we need to understand because we have concience.. idk if it's clear I tried my best to explain my thoughts...
@@darkgobelin4439The size of our brain doesn't really affect our intelligence. Popular misunderstatement. Though, i don't really know about what makes us far more intelligent than any other animal. I might guess that it's the amount of neurons in our brain.
Many of these structures appear pretty stable, I think if this simulation had a way of making new particles out of existing ones, self-replication could be achieved
The simulation works a lot like a closed system, or a cave underground. What this needs for it to be more realistic is to have new elements constantly appearing (kinda like energy from the sun hitting the earth)
self replication was achieved, it just involved murder at one point the orange yellow and blue cells were wplit by the red and cyan cells creating two new cells, mitosis being acheived partially
@@creature-zf8rs It's less of mitosis, and more of forced cytokinesis, the cell is violently ripped in half by some external structure and particles are somewhat evenly dispersed between the daughters.
You could, if more than 20 particles are clumped together. Change their value so that the different colors form “covalent bonds” with other particles of that color within the clump (which would just be like some string like code so they can’t wander off) Then you get molecules and it can get a little more complicated. You could also have some structures which you know are useful like bonds which two poles be forms that can form easily. If you know it can form in real life due to reactions it should be allowed to make it form in the simulation. Life could get more complex that way
As conway's game of life you can never really seem to be able to estimate how many generations this simulation will have or if it will be stable, if a cell will grow indefinetly, etc. Super duper cool particle chaos.
It's provably impossible to decide for all cases because of halting problem, not to mention NP problems like collatz conjecture, which is essentially a subset of halting problem.
I can't imagine a better way of simulating and explaining real-world biology than this. The first part literally teaches you about genetic traits in a way that is so undeniably simple that even a 3rd grader could learn it
I like to think of this more of as quantum traits rather than genetic ones because the simulation defines simple rules of how different particles interact with each other, the entire simulation then arises from these simple rules
I think one of the limiting factors of this simulation is the lack of collisions. As you have said, you have added a repulsion for when two particles get too close. Yet, many times, once there is a big enough number of particles attracting each other they seem to reach a sort of "size limit" from which the addition of particles will only make the mass of particles denser and not bigger. This, seems to greatly limit the size and complexity of the "organisms" that emerge. Probably the addition of actual collisions would suppose a significant strain on the GPU but I think it might pay off. Thank you so much. Loved the video.
With hundreds of different particle types and millions of particles and given the „correct“ matrix for their interactions with each other, with enough time and just by chance - some structure could form, that can collect other particles and can thus replicate itself - which would make it a cell. And with that, real simulated life could „evolve“ over time just in this simple simulation. That’s how life emerged in our reality, probably.
6:39 I see that all the particle collections are preferentially moving towards the right, this suggests either a flaw in seeding or general anisotropy in the direction of the interaction strengths. It is probably a bug in the software. ie. when I was making my Lenia simulator, I used fourier transforms for convolution, and that meant that when I tried to use a kernel with (iirc) odd side length then it would make the whole simulation have a preferential spatial direction.
I made my own web implementation of particle life. I had to put the link in a community post on my channel as yt automatically deletes comments with links apparently. It's pretty fun to mess around with, but it isn't gpu accelerated so you can't have a massive amount of particles, the number varies on how good your cpu is.
4:24 you can see that in this configuration the particles either become 'ships', rectangles with sorted colours, or cells, with which are circular-ish and group differently
I only think you'd see many more species if the colours weren't evenly mixed in with each other. If there was an uneven distribution it would be more likely that different species could emerge, and then engage in interspecific competition based on their attraction/repulsion properties.
You could possibly make a kind of molecular behaviour by giving them properties which come into effect when they are "bonded" to another particle. A bond would be defined as a certain level of force interaction between 2 particles for at least x seconds (1N over 0.5 seconds bonds them until it stops or drops to 0). Say when you have a structure made of reds and blues, it changes some attraction properties, or even adds some interaction properties with other pairings.
even with such simple ruleset it almost looks like life sometimes. Imagine how it would get after adding more complex rules, like bonds, or multi-layer movement (like what's already here, but the repel/attract rules change at higher distances. making things repel to a point, and further than that attract, or vice versa, btw. this is an actual quantum property)
Do you know the reason why it stopped increasing in complexity at a certain level? Not because they were too few or too slow but because they don’t have purpose in their existence - complexity is created by a purpose or, at the very least, a feedback that is stored in them. That’s what I’m currently working on... for 15 years:)
I tried to recreate this project for my own enjoyment over on codepen. I've gotten it pretty similar but I don't know how you got such complex patterns that stay stable. Particularly, the organisms with striped sections like the one in the thumb nail. If you have any advice I'd love to hear it!
This reminds me of the hypothesis of fine-tuned universe. According to this hypothesis, the values of all physical constants are so finely matched that even the slightest deviation would lead to the impossibility of the existence not only of life, but also of fundamental structures such as quarks and atom
@@alexanderthemidIIt has been proved, well sort of... No tangible research has been done on it but through equations of physics any physicist can tell that if the constants in our reality were slightly different nothing would exist... It's like knowing that if you cut off the base of a cup you can never fill it with water. Everything is just perfectly tuned... perfectly.
@@Blankoo3di agree with that. But I also have a question is that if all contants are increased 1% in perfect ratio, would the universe will work? (And it's the "interactions of these constants" that shape the world)
@@alexanderthemidI the problem with "life finds a way" into relation with the fundamental presets is that - they are constants, which mean since the beginning of this universe they have been the same, so it doesn't make sense for the universe to do try and error until they find the perfect values.
@@lixun7390 There is literally nothing saying there can't be multiple universes that may have different laws of physics, and we happened to be in the one that supports this kind of life.
I'd be really curious to see how environmental pressures would impact the simulation. Like an area that changes attraction or changes a particle from one to another.
It would be interesting to see particles be able to change into other particles when certain conditions are met, such as 2 red particles changing into a green for instance, or after a set amount of time. Combining this with rules similar to Conway's game of life would prove insightful to seeing how homeostasis naturally emerges.
Right? There were so many cases of perpetual motion machines that made me annoyed. However, as a rough approximation of what unicellular life behaves, it still does an okay job. In reality, we have positive, neutral, or negative charges, and that's really all we get to play with. Ah well. At least it is pretty.
@@Smiley_404 It seems like the "cell" itself was barely holding itself together, but due to the intense charge of the "membrane" attracting to the "nucleus" (via the bridging orange particles that attract both the "membrane" and "nucleus), it remained stable. Once the "cell" got a hold of more cyan particles (those that make up the "nucleus") the force overwhelmed that of the "membrane" and it violently exploded. From this, the "membrane" and "bridge" particles formed new, smaller "cells", which quickly picked up a "nucleus" of a few cyan particles. This system is unbelievably impressive, notice how the daughters (terminating the use of quotation marks for convenience) without a nucleus of cyan particles form far more fluid structures, as the nucleus was a necessity for a proper membrane to form. Speaking of the membrane, the membrane appears perfectly formed to allow for fission of the cell. At large sizes (when the cell would want to divide), it forms slits that easily allow cyan particles into the nucleus, until the force overpowers the membrane and the cell divides. The life cycle of this structure of particles works specifically to grow in size, until it is too large, then divide. This, however, is similar to a cell being struck by an external structure that forcefully divides it. The major difference here is how the cell itself works to allow for this. The only issue with this cell is its inability to defend itself, despite how impressive its abilities of fission are, most or all of the offspring die before maturity (in which it divides).
This is my attempt to catalog all the “lifeforms” that can evolve here. There are three main categories of life, which I will classify as Floaters, Coasters and Absorbers. Floaters are the simplest type of lifeform, being stationary creatures that just consist of one or two types of cells. These seem to be most common in simulations with low particle color counts. Species include 3:21 Flavum Botrus and 3:54 Red Centrum. Floaters do not necessarily have to be simple however, as the species Longa Forma demonstrates at 14:15. Coasters are species that revolve around being very fast, being able to speed across the map rapidly. These are surprisingly durable, being able to reform after an impact fairly easily. These can vary in speed, from being only moderately fast, like the species Neo Corpus at 4:13, or very fast, or the species family Purpura Caput at 6:35. Finally we have predators, which revolve around feeding on members of its own kind, as other prey species tend to destroy them. These can either be stationary or moving, as long as they can effectively catch prey and reform it into a part of itself. The species include Rubrum Luna at 11:02 and its ambush predator variant at 11:08 at the bottom of the simulation. My favorite species was the Navigare Navis at 10:45, a species of Coaster with a unique sail-shaped appearance. (God I spent way too long doing this lol)
While I must admit that this is a very interesting concept, and that It does appear to simulate life, you forgot a crucial detail about life. You see, just because they are able to move and are also able to create complex formations, they are unable to act of their own accord. The particles in this video are only moving simply based on the laws of attraction and repulsion. For example, if I were to take a bunch of north pole and south pole magnets, while they could repel and attract each other to form complex structures, they are unable to move and act of their own accord. The same problem would occur, even if there were a dozen different types of magnetic poles. On the contrary, life is able to move independent of outside forces. While some organisms have simple jobs that only require them to do a certain number of tasks, others are able to think for themselves and choose where they want to go, regardless of the forces of attraction and repulsion. So, I ask you this one simple question. How do these particles evolve into complex organisms that are able to move independently of each other? I fail to see how these structures are able to produce a T-Rex that is able to think for itself even if they had a billion years to do it.
I think "external forces" is inaccurate. Your atoms are pulling on the Earth while it pulls on you. Your atoms are affecting a magnet while it's affecting you. I think there is no real internal vs external distinction, it's all the same field, and you can algebra the + / - signs to describe it from the perspective of this or that object. But that's an arbitrary choice, and exactly the same level of complexity will emerge from exactly the same level of simplicity, regardless of which side you put all the minus signs in your conceptual representation of whatever nature is actually doing.
@@BlackMatt2k Basically what I’m trying to say is that there’s more to life than just pulling and repelling. The video only shows how particles will act based solely on pulling and repelling alone.
@@Osprey2511 I think the point is that even with only pulling and repelling, only 2 dimensions, only X compute, counterintuitively complex states and behaviors still arise. Of course it's not "real life", but understanding how "fake life" emerges from simple rules provides conceptual frameworks for people to investigate "real life". Playing with LEGOs isn't building skyscrapers, but if you ask 2 kids to analyze a skyscraper, the one who played with LEGOs will notice different things, ask different questions. These are _models_ of _aspects_ of a thing, and in *2D*, which of course you can't get "real life" out of anyway, because all the twisting and folding of real particle configurations allows for waaay more complexity.
This simple simulation explains so much about life 14:37 This simulation even shows a distinction between plant and animal species! purple-red-blue organisms move around to find more nutrients (more particles of its own kind) while green-blue organisms form layers to grow outwards and reach more nutrients this way. It is absolutely astonishing and promises a future where simulation of inner workings of our own universe will explain what we could not before! Imagine this kind of simulation but with a lot of different particles, where some of them expire and create different particles. Organisms would need to find nutrients before their own body expires. Meanwhile other organisms would feast on the disposed matter. just wow
I would love to see a simulation like this but with more colours. Imagine 30 different particles with different properties, how would that change the randomness of the simulation?
I have to admit I am curious. what would happen if you applied 35 different attract or repel values to 10 different colors in a 100 total particle simulation?
I noticed that there are many shots where cells tend to travel from left to right. I wonder if there's a bias in phisics engine that favours that direction, or it is an intended feature, a background gradient. But still, nothing that makes it less amazing.
I liked the video because it was interesting mathematically, however, it still doesn't explain the origin of life but only that information tends to cluster whenever it's possible/favourable, or that pattern emerges when things are viewed from a larger scale (Ramsey). The process behind the formation of the earliest life is way more complex than what you presented since a lot of things need to be just right.
This is incredible. I am thinking of additional rules. Perhaps if two certain unlike particles meet, they turn into different particles, or maybe just one does. Perhaps some particles lose or gain attraction forces as they touch other like particles.
What happens if you make the matrix symmetric (i.e. Newton's third law)? This would prevent the particles from gaining energy from nowhere. Also, I think you may have some kind of update order bug? I noticed that all the groups of particles move more right than left in every simulation, and that tends to happen to me as well. You can usually get rid of it by breaking the loop into three separate loops, one to update the position, one to update the velocity, and one to update the acceleration, but this doesn't always work.
This is a great example of something that has bugged me for ages. It's a common belief in evolutionary science that all life on Earth is descended from the first single cell organism, but what these p-life sims clearly demonstrate is that when the conditions are perfect for *one* single cell organism to form, the same conditions are also very favourable for multiple other similar, or even identical organisms to form independently.
To note something; Every aspect of our world can be related to a clock. Certainly at its core, it's just a gear turning, but when you take so many simple things, and blend them together in various ways, the complexity compounds. Going from a simple mechanical motion to the telling of time. Life is no different. At the end of the day it is a basic chemical process which life starts, many of these come together to create something truly awe inspiring.
I know it's beyond the scope of what you were doing, but did you put any consideration into conservation of energy? I notice that some of those "cells" are propelling themselves around like someone blowing into their own sails - creating energy from nothing. That's fine, of course, given your scope, but I thought it was interesting to wonder about.
You should add “chemistry” when when a particle gets close enough to another they can change into different particles not do anything or have only one particle change
This would be really good to teach about various forces on molecules/particles, I would really hope you publish it, or sell it even on something like steam for people to use to teach ewlements of physics or biology
This is fascinating! I"m wondering whether large some of these could exhibit traits we typically associate with early biological life, aka homeostasis / natural selection / self duplication. Did any of these sets of rules produce multiple different sets of particle groups that "fought" over "dominance" ( aka was "phenotype" distribution variable over time, like a predator-prey cycle ? )
I tried to programm it myself and it looks so unbelievable cool with the right tweaking with the variables. Thanks for the vid with the explainations :)
I wonder if you add into this some kind of chemical reaction where when some specific particles get too close they would transform to other ones. I would like to see if something like reproduction can easily appear in this simulations also
I’ve been thinking about doing this for a month. Thank you for showing me the possibilities; this is fantastic! Also I love the music. Where might I find some? Or even, what’s this genre called? Thanks.
What if you add the idea of energy, functions for each particle (like nucleus and mitochondria dna production etc), and the ability for cells to come together but they have like a membrane against cells of the same type / membrane
iv thought of some things to try that may make it even more realistic. or et very least may have intertesting results. nonlinear force curve. could result in more subtle interaction types "coupling" force emulation (i.e particles that get very close "stick" aggressively and rigidly to each other until they are far enough apart) tailored particle fields based on real world organic elements like hydrogen and carbon. multi field models. electromagnetic and strong/weak force emulation.
This shows that you don't need overly complex things to make something beautiful (and amusing to my single celled brain)
i like this comment, it gets a dollar sign emoji
💲
fr
@@practicemodebutton7559 💲
Fr twice in one comment section
@@practicemodebutton7559also a satellite emoji
🛰
Can you pls make a site where you can play with these things, and if you do, put it in the description and reply to this comment, telling you created it? I WANT TO PLAY WITH THESE THINGS SO BAD
from the looks of it, it would take a lot of computing power to run, so it's not likely many people would get as much use from it as he did
@@BoredYoshi it runs on the GPU so not really
@@holl7w true
I AGREE BRO
He can make a smaller version
At some point, one of these structures will randomly be extremely stable and self replicating. (Maybe something with its outer wall which allows it to gather more of its kind).
That would make the chaos go extinct....
at most of the videos I can see some 'cells' as I would like to call it, eat other particles or cells and it become too big to it splits into 2 cells...
Kinda like how cells division work but without the chromosomes bullshits lol
The problem with this is that it is _too_ easy for multiple of the same structure to form from nothing, meaning that there is very little material left for any of them to make clones out of without first cannibalising something ( since they all use unique particles )
Unfortunately he didn't add any sort of particle creation system that the particles can access.
Yeah i feel like i could see that in a more complex environment the material in the outer layer of the “cell” could be attracted to the center layer with a middle layer seperating it, so that if the availability of the materials is right it could form a stable loop of getting big then collapsing in on itself in such a way that it replicates
I think the complex "cells" emerging from this soup of particles and "evolving" to survive in the randomness and chaos is quite cool.
If you think about it, pretty much everything in the universe is a bunch a particles following strict rules, so this is perfect.
1:58 first eukaryotic cell
3:44 cell walls form
4:08 complex life (and separate species)
5:26 many species (and racism)
5:34 mitosis
6:07 genuses start forming (similar species)
8:51 defense mechanisms arise (the blue "prey" cells start deflecting the red "predator" cells)
10:37 multicellular life emerges
11:14 cell nuclei form
12:52 filter feeders emerge
14:16 dna-like structure emerge
edit: how did this get over 15 likes in under a day
edit 2: some guy made a reply criticizing this so heres a disclaimer: i am not a scientist i made this comment for fun do not take any of this as something accurately explaining this
still dont have to be so salty about it tho
also the like count has gone up times ten when i last edited
edit 3:
Foreword: I am sorry if i misinterpreted anything you said, correct me if i am wrong because some thing sound more hostile than they should logically be from you.
to c0dejjshizpostarchive624, i will call you Cody because I'm not gonna say all that, and to Mr_Tophatt, i have seen your argument and i have decided that bot of you are wrong. Cody, nobody would have believed that an all lowercase comment from some guy with the word meme in his name which features the word racism and uses the wrong term for similar species was going to be scientifically accurate, and to tophat, the is no way to "correct" a joke, but the lesser of the two evils is tophat, Cody says that there are no names, and that is true, except i was using those names to compare to the real things. and no this is not a strawman fallacy, i am simplifying what he said. I was comparing the simulation to real things in cellular life, and Cody, you said this was an "egregious misrepresentation" and that feels more like an insult than a regular saying something is incorrect for a good reason. and shut up about me misusing real scientific terms literally not another soul on this planet cares about that. Please stop blabbering about misrepresentation and actually respect me for taking time i will never get back to write a response to someone who will never read it., and looking for timestamps to compare to real life. I could have ignored you but i did not, and i hope you write back so i can understand a bit more about what you are doing. Arguments are supposed to be learned from, so i will leave some criticisms off so this won't go stale. Back to tophat, again, i was comparing the simulations to real life phenomena, and there shouldn't be "correct" and "incorrect" terms, Tophat said less so i say less about him.
*_TL/DR:_* Both people who were arguing were wrong in some form. Both had shortcomings but they also had times of being correct. Both people did not win the argument, so i am making an edit. All in all, this comment was a joke and Cody took it too seriously but Tophat had some incorrect things too. Both were wrong and the argument will continue.
Thank you
Dam bro
bro made an edit for 15 likes
That filter feeders are something else entirely)
Although this is fun and all, I would like to remind everyone that everything in this comment is an egregious misrepresentation of these concepts. I don't believe that OP intended to pass this off as "real", but for those gullible enough, this is absolutely positively incorrect.
10:47 I love these guys. They look like they have umbrellas
_Umbrellium qwertii_
@@malechex611 :D
I made my reply before seeing urs
they act like fishing nets! i can't believe that such behavior could appear from just these simple rules.
They look like the little mushrooms from mario
Saw the title and the thumbnail and the channel and I knew I was gonna see something good
If only more content creators knew these basic secrets to luring in more viewers.
These kinds of simulations always spark so many ideas! What if you introduced gravity, each particle being attracted to a point at the centre inversely proportional to distance. What if you introduced energy, which affected the max speed of the particles? What if you grouped the particles in the beginning rather than randomly dispersed them? What if what if what if. This was beautiful. Thanks for putting it together
What I think can be a gamechanger is mixing colours and their "electronegativity"
Gravity is not a point in space, but how much a mass bends the space or pulls.
This game is already a gravity-pulling type simulation, but what should happen is that as the mass grows bigger, its pull should be bigger. I think that's what you meant.
@@aezravito9717 I meant what I said. When you program a simulation like this you can represent elements of reality as selectively and abstractly as you want. The current simulation has no direct or accurate representations of real world physics or phenomenon. I used the word “gravity” only to convey the idea. I think the version of gravity you propose would be very computationally expensive, but also very cool. It would be great to see that with an enormous world.
@@GerinoMorn I think that could be done for each particle by multiplying the color's "weights" with random numbers near 1 which would change the interaction but not so much so that it functions completely differently.
Now I wanna see this with the atomic scale now. Protons, neutrons, electrons, or the quarks making them up, then watch them as they show the different properties of gravity, electromagnetic properties, charges, changes in state of matter (solid to liquid, liquid to gas, gas to liquid, liquid to solid, etc.), radioactivity, tranparecy, conductivity, malleability, and more. Imagine how big of a simulation you need just for those things that are surprisingly 99% empty
To be fair, you’d need to know how quantum gravity works, and no one knows how quantum gravity works lol
@@elementgermaniumhere's the great thing about simulations though! You don't, you just need to know it's effects! This simulation you watched was simplified as hell, and CELLS formed!
They can simulate a human organ I think on the molecular level. One of my prof told that in a lecture 10 years ago. But they had research computation clusters. I bet its even more achievable today.
The question is: will life emerge by doing such a simulation.
@@4984christian It depends on what you mean by "life"- In a simulation like this, I've personally seen simple "creatures" that eventually get so unstable they split into 3+, but more complex ones have a hard time
It would be AMAZING to one day see spontaneous generation with our OWN EYES, computer or not!
really makes you think about sentience
like, at what point do you go from mass to life? to brains? to concience?
Edit: this discussion that has started in the replies is civil??? never seen this before.
fr
I personnaly think that what makes us sentient is just the size of our brain, like at one point we became smarter that other animals and we gained concience and emotions to better understand, but we need to understand because we have concience.. idk if it's clear I tried my best to explain my thoughts...
@@darkgobelin4439 we know that, @junhongwu1882 is asking at which point consciousness starts
@@darkgobelin4439The size of our brain doesn't really affect our intelligence. Popular misunderstatement. Though, i don't really know about what makes us far more intelligent than any other animal. I might guess that it's the amount of neurons in our brain.
@@PEIIIKA It's approximately the ratio of the brain-mass compared to body-mass that makes things intelligent
Many of these structures appear pretty stable, I think if this simulation had a way of making new particles out of existing ones, self-replication could be achieved
The simulation works a lot like a closed system, or a cave underground.
What this needs for it to be more realistic is to have new elements constantly appearing (kinda like energy from the sun hitting the earth)
self replication was achieved, it just involved murder
at one point the orange yellow and blue cells were wplit by the red and cyan cells creating two new cells, mitosis being acheived partially
@@funnifunnifunni mitosis kinda happens through and requires stimuli as seen in the simulation because without it they won't split
@@creature-zf8rs It's less of mitosis, and more of forced cytokinesis, the cell is violently ripped in half by some external structure and particles are somewhat evenly dispersed between the daughters.
Mitosis would be extremely complex, but there is an experiment where the structures reproduced when catching "food" and they started evolving
You could, if more than 20 particles are clumped together. Change their value so that the different colors form “covalent bonds” with other particles of that color within the clump (which would just be like some string like code so they can’t wander off)
Then you get molecules and it can get a little more complicated. You could also have some structures which you know are useful like bonds which two poles be forms that can form easily. If you know it can form in real life due to reactions it should be allowed to make it form in the simulation. Life could get more complex that way
One of the coolest videos I have ever seen highlighting emergent properties
As conway's game of life you can never really seem to be able to estimate how many generations this simulation will have or if it will be stable, if a cell will grow indefinetly, etc. Super duper cool particle chaos.
It's provably impossible to decide for all cases because of halting problem, not to mention NP problems like collatz conjecture, which is essentially a subset of halting problem.
Appreciate your attention to details! The foley sound effects add depth and professionalism to the video, loved it overall!
youtube's compression algorithm hates this man
love the animations and the background music in this one! another amazing upload
fr
I can't imagine a better way of simulating and explaining real-world biology than this. The first part literally teaches you about genetic traits in a way that is so undeniably simple that even a 3rd grader could learn it
I like to think of this more of as quantum traits rather than genetic ones because the simulation defines simple rules of how different particles interact with each other, the entire simulation then arises from these simple rules
I think one of the limiting factors of this simulation is the lack of collisions. As you have said, you have added a repulsion for when two particles get too close. Yet, many times, once there is a big enough number of particles attracting each other they seem to reach a sort of "size limit" from which the addition of particles will only make the mass of particles denser and not bigger. This, seems to greatly limit the size and complexity of the "organisms" that emerge. Probably the addition of actual collisions would suppose a significant strain on the GPU but I think it might pay off. Thank you so much. Loved the video.
This actually might be the way that life started. The visualisation is great and you need to be more popular for what you are doing❤
fr
Ya
Alah who Het universum maken
@@Alihussei0n Nope
@@pidx YOU MEAN THAT ALAH DIDN'T DO IT!!!!!
When you think about it life is just particles of different types that like each other or hate. It's so simple, but so fascinating.
With hundreds of different particle types and millions of particles and given the „correct“ matrix for their interactions with each other, with enough time and just by chance - some structure could form, that can collect other particles and can thus replicate itself - which would make it a cell. And with that, real simulated life could „evolve“ over time just in this simple simulation. That’s how life emerged in our reality, probably.
Only if mutations would happen in these copies
17:50 two celled organism at the bottom moving up!
6:39 I see that all the particle collections are preferentially moving towards the right, this suggests either a flaw in seeding or general anisotropy in the direction of the interaction strengths. It is probably a bug in the software.
ie. when I was making my Lenia simulator, I used fourier transforms for convolution, and that meant that when I tried to use a kernel with (iirc) odd side length then it would make the whole simulation have a preferential spatial direction.
I made my own web implementation of particle life. I had to put the link in a community post on my channel as yt automatically deletes comments with links apparently. It's pretty fun to mess around with, but it isn't gpu accelerated so you can't have a massive amount of particles, the number varies on how good your cpu is.
4:24 you can see that in this configuration the particles either become 'ships', rectangles with sorted colours, or cells, with which are circular-ish and group differently
This is amazing! I'd like to see it with way more colors, even if you can't tell the colors apart you might see a lot of different species emerge
I only think you'd see many more species if the colours weren't evenly mixed in with each other. If there was an uneven distribution it would be more likely that different species could emerge, and then engage in interspecific competition based on their attraction/repulsion properties.
I like how most of these have at least a single species in them, showing that those exact species with those settings are the most stable ones
This Simulation is beatiful, not gonna lie, yet a real cell is so much more complex then this, that its hard to comprehend.
Really nice visualization, great music too.
Can we do this in 3d?
I think it’s sorta on 3d surface lol. It’s all one surface at least. No edges
Way more lag more combinations
You'll need one more dimension of computational power
Y Æ S
@@mitchellparadise3801no it’s on a looping 2d surface
You could possibly make a kind of molecular behaviour by giving them properties which come into effect when they are "bonded" to another particle. A bond would be defined as a certain level of force interaction between 2 particles for at least x seconds (1N over 0.5 seconds bonds them until it stops or drops to 0).
Say when you have a structure made of reds and blues, it changes some attraction properties, or even adds some interaction properties with other pairings.
GREAT VIDEO!
Bot
Can you add polarity to the particles, like repelling from one side and attracting in another?
1:28 We're only a minute and a half in and already the lifeforms have learned how to simulate racism
Lol
14:22 I like those. Serene ordered caterpillars of bright colors and chaotic entropic beings of dark colors...
Conway been real quiet since this dropped.
Bro he died
10:49 the sail boat creatures were my fav.
even with such simple ruleset it almost looks like life sometimes.
Imagine how it would get after adding more complex rules, like bonds, or multi-layer movement (like what's already here, but the repel/attract rules change at higher distances. making things repel to a point, and further than that attract, or vice versa, btw. this is an actual quantum property)
5:34 those cells in the middle gonna have a talk with Pepsi in court
I love the way the ones at 10:40 trail things behind them. So sci-fi like!
Do you know the reason why it stopped increasing in complexity at a certain level? Not because they were too few or too slow but because they don’t have purpose in their existence - complexity is created by a purpose or, at the very least, a feedback that is stored in them. That’s what I’m currently working on... for 15 years:)
This is so awesome and interesting!
I tried to recreate this project for my own enjoyment over on codepen. I've gotten it pretty similar but I don't know how you got such complex patterns that stay stable. Particularly, the organisms with striped sections like the one in the thumb nail. If you have any advice I'd love to hear it!
This reminds me of the hypothesis of fine-tuned universe. According to this hypothesis, the values of all physical constants are so finely matched that even the slightest deviation would lead to the impossibility of the existence not only of life, but also of fundamental structures such as quarks and atom
We can't really prove that though. Even if the universe chose a different preset, you know what they say; life finds a way, I guess.
@@alexanderthemidIIt has been proved, well sort of... No tangible research has been done on it but through equations of physics any physicist can tell that if the constants in our reality were slightly different nothing would exist... It's like knowing that if you cut off the base of a cup you can never fill it with water. Everything is just perfectly tuned... perfectly.
@@Blankoo3di agree with that. But I also have a question is that if all contants are increased 1% in perfect ratio, would the universe will work? (And it's the "interactions of these constants" that shape the world)
@@alexanderthemidI the problem with "life finds a way" into relation with the fundamental presets is that - they are constants, which mean since the beginning of this universe they have been the same, so it doesn't make sense for the universe to do try and error until they find the perfect values.
@@lixun7390 There is literally nothing saying there can't be multiple universes that may have different laws of physics, and we happened to be in the one that supports this kind of life.
I'd be really curious to see how environmental pressures would impact the simulation. Like an area that changes attraction or changes a particle from one to another.
Imagine a bullet hell game that you need to avoid those particles
It would be so cool
ah sweet
running from a perpetually exploding shrapnel bomb that ricochets all over the place
Intense Touhou gameplay💀💀💀
It would be interesting to see particles be able to change into other particles when certain conditions are met, such as 2 red particles changing into a green for instance, or after a set amount of time. Combining this with rules similar to Conway's game of life would prove insightful to seeing how homeostasis naturally emerges.
this is why i love evolution simulations
I need this as my phone and PC wallpaper, simulated in real time. So fun to watch!
the fact that you skewed the laws of physics by not abiding to the reaction force is a no go for me. It still looks amazinng and feels alive
Right?
There were so many cases of perpetual motion machines that made me annoyed.
However, as a rough approximation of what unicellular life behaves, it still does an okay job.
In reality, we have positive, neutral, or negative charges, and that's really all we get to play with. Ah well. At least it is pretty.
This channel is so underrated
5:19 i love this
The creature on the bottom left literally exploded from eating too much
@@bluestone-gamingbg3498it was more like "reproduction" notice the parts formed 2 other of itself, that's similar to how some real cells reproduce.
@@Smiley_404 It seems like the "cell" itself was barely holding itself together, but due to the intense charge of the "membrane" attracting to the "nucleus" (via the bridging orange particles that attract both the "membrane" and "nucleus), it remained stable. Once the "cell" got a hold of more cyan particles (those that make up the "nucleus") the force overwhelmed that of the "membrane" and it violently exploded. From this, the "membrane" and "bridge" particles formed new, smaller "cells", which quickly picked up a "nucleus" of a few cyan particles.
This system is unbelievably impressive, notice how the daughters (terminating the use of quotation marks for convenience) without a nucleus of cyan particles form far more fluid structures, as the nucleus was a necessity for a proper membrane to form. Speaking of the membrane, the membrane appears perfectly formed to allow for fission of the cell. At large sizes (when the cell would want to divide), it forms slits that easily allow cyan particles into the nucleus, until the force overpowers the membrane and the cell divides.
The life cycle of this structure of particles works specifically to grow in size, until it is too large, then divide. This, however, is similar to a cell being struck by an external structure that forcefully divides it. The major difference here is how the cell itself works to allow for this.
The only issue with this cell is its inability to defend itself, despite how impressive its abilities of fission are, most or all of the offspring die before maturity (in which it divides).
This is my attempt to catalog all the “lifeforms” that can evolve here. There are three main categories of life, which I will classify as Floaters, Coasters and Absorbers. Floaters are the simplest type of lifeform, being stationary creatures that just consist of one or two types of cells. These seem to be most common in simulations with low particle color counts. Species include 3:21 Flavum Botrus and 3:54 Red Centrum. Floaters do not necessarily have to be simple however, as the species Longa Forma demonstrates at 14:15. Coasters are species that revolve around being very fast, being able to speed across the map rapidly. These are surprisingly durable, being able to reform after an impact fairly easily. These can vary in speed, from being only moderately fast, like the species Neo Corpus at 4:13, or very fast, or the species family Purpura Caput at 6:35. Finally we have predators, which revolve around feeding on members of its own kind, as other prey species tend to destroy them. These can either be stationary or moving, as long as they can effectively catch prey and reform it into a part of itself. The species include Rubrum Luna at 11:02 and its ambush predator variant at 11:08 at the bottom of the simulation. My favorite species was the Navigare Navis at 10:45, a species of Coaster with a unique sail-shaped appearance. (God I spent way too long doing this lol)
While I must admit that this is a very interesting concept, and that It does appear to simulate life, you forgot a crucial detail about life. You see, just because they are able to move and are also able to create complex formations, they are unable to act of their own accord. The particles in this video are only moving simply based on the laws of attraction and repulsion. For example, if I were to take a bunch of north pole and south pole magnets, while they could repel and attract each other to form complex structures, they are unable to move and act of their own accord. The same problem would occur, even if there were a dozen different types of magnetic poles.
On the contrary, life is able to move independent of outside forces. While some organisms have simple jobs that only require them to do a certain number of tasks, others are able to think for themselves and choose where they want to go, regardless of the forces of attraction and repulsion.
So, I ask you this one simple question. How do these particles evolve into complex organisms that are able to move independently of each other? I fail to see how these structures are able to produce a T-Rex that is able to think for itself even if they had a billion years to do it.
I think "external forces" is inaccurate. Your atoms are pulling on the Earth while it pulls on you. Your atoms are affecting a magnet while it's affecting you. I think there is no real internal vs external distinction, it's all the same field, and you can algebra the + / - signs to describe it from the perspective of this or that object. But that's an arbitrary choice, and exactly the same level of complexity will emerge from exactly the same level of simplicity, regardless of which side you put all the minus signs in your conceptual representation of whatever nature is actually doing.
@@BlackMatt2k Basically what I’m trying to say is that there’s more to life than just pulling and repelling. The video only shows how particles will act based solely on pulling and repelling alone.
@@Osprey2511 I think the point is that even with only pulling and repelling, only 2 dimensions, only X compute, counterintuitively complex states and behaviors still arise. Of course it's not "real life", but understanding how "fake life" emerges from simple rules provides conceptual frameworks for people to investigate "real life". Playing with LEGOs isn't building skyscrapers, but if you ask 2 kids to analyze a skyscraper, the one who played with LEGOs will notice different things, ask different questions.
These are _models_ of _aspects_ of a thing, and in *2D*, which of course you can't get "real life" out of anyway, because all the twisting and folding of real particle configurations allows for waaay more complexity.
@@BlackMatt2k Because you mentioned Lego, I will put this debate to rest. Well played.
This simple simulation explains so much about life
14:37 This simulation even shows a distinction between plant and animal species!
purple-red-blue organisms move around to find more nutrients (more particles of its own kind) while green-blue organisms form layers to grow outwards and reach more nutrients this way. It is absolutely astonishing and promises a future where simulation of inner workings of our own universe will explain what we could not before!
Imagine this kind of simulation but with a lot of different particles, where some of them expire and create different particles. Organisms would need to find nutrients before their own body expires. Meanwhile other organisms would feast on the disposed matter.
just wow
No it doesn't lmao.
I would love to see a simulation like this but with more colours. Imagine 30 different particles with different properties, how would that change the randomness of the simulation?
Pseudo-asexual reproduction by division at 17:26? Around the center of the screen.
I have to admit I am curious. what would happen if you applied 35 different attract or repel values to 10 different colors in a 100 total particle simulation?
5:52 COVID-19 at the bottom right
I noticed that there are many shots where cells tend to travel from left to right. I wonder if there's a bias in phisics engine that favours that direction, or it is an intended feature, a background gradient. But still, nothing that makes it less amazing.
10:29 cell turned into earth
I liked the video because it was interesting mathematically, however, it still doesn't explain the origin of life but only that information tends to cluster whenever it's possible/favourable, or that pattern emerges when things are viewed from a larger scale (Ramsey). The process behind the formation of the earliest life is way more complex than what you presented since a lot of things need to be just right.
Imagine introducing rules for life and death. It'd be really interesting to see
Actually, they are already there. A cell is alive while it's whole and moving, and dies when it's eaten by another or collapses
@@Aaa-hl6oj that's true but I meant the particles rather then cells
Super cool!
Btw what’s the background music?
fun fact: no one finished watching the video yet
Now some one can
Yes
I did
about 2k people watched it now
@@BoneEaters context: i made this comment within 4 minutes when the video was uploaded
Theists just can't comprehend the beauty of life and this universe, it's just so incredible
6:38 Cells are migrating off screen
They reappear on the other side
6:00 youtube compression loves this bit
This is incredible. I am thinking of additional rules. Perhaps if two certain unlike particles meet, they turn into different particles, or maybe just one does.
Perhaps some particles lose or gain attraction forces as they touch other like particles.
What happens if you make the matrix symmetric (i.e. Newton's third law)? This would prevent the particles from gaining energy from nowhere.
Also, I think you may have some kind of update order bug? I noticed that all the groups of particles move more right than left in every simulation, and that tends to happen to me as well. You can usually get rid of it by breaking the loop into three separate loops, one to update the position, one to update the velocity, and one to update the acceleration, but this doesn't always work.
This is a great example of something that has bugged me for ages. It's a common belief in evolutionary science that all life on Earth is descended from the first single cell organism, but what these p-life sims clearly demonstrate is that when the conditions are perfect for *one* single cell organism to form, the same conditions are also very favourable for multiple other similar, or even identical organisms to form independently.
To note something; Every aspect of our world can be related to a clock. Certainly at its core, it's just a gear turning, but when you take so many simple things, and blend them together in various ways, the complexity compounds. Going from a simple mechanical motion to the telling of time. Life is no different. At the end of the day it is a basic chemical process which life starts, many of these come together to create something truly awe inspiring.
You can really apply everything happening here to life. This is so cool to watch.
Nice music at 5.00 min. What is the name?
Also nice Video.:-)
This video is underrated. It deserves more views
You should give the particles access to some sort of particle creating power so that they start small but burst into big creatures
I know it's beyond the scope of what you were doing, but did you put any consideration into conservation of energy?
I notice that some of those "cells" are propelling themselves around like someone blowing into their own sails - creating energy from nothing.
That's fine, of course, given your scope, but I thought it was interesting to wonder about.
You would call that chaos. Like in real life molecules and atoms.
@@ooRobertoo Chaos doesn't create energy in real life. This has no real-life analogue.
@@delphicdescant Chaos would mean things like gravity, van der Waal forces, magnetic forces, etc.
You should add “chemistry” when when a particle gets close enough to another they can change into different particles not do anything or have only one particle change
Great video, simply explained yet complex and beautiful. You could have add the force matrix you used for the different simulations.
Bro actually found a way to say particle a billion times in one video, LOVE IT!
Do you plan on releasing the code for this?
6:29 anyone know the name of the song?
Can you make a follow-up video where you give concrete examples of molecules that would have traits like the ones illustrated in your content?
9:04 love how the neon green and cyan cells kinda just sit there
This would be really good to teach about various forces on molecules/particles, I would really hope you publish it, or sell it even on something like steam for people to use to teach ewlements of physics or biology
This is fascinating! I"m wondering whether large some of these could exhibit traits we typically associate with early biological life, aka homeostasis / natural selection / self duplication. Did any of these sets of rules produce multiple different sets of particle groups that "fought" over "dominance" ( aka was "phenotype" distribution variable over time, like a predator-prey cycle ? )
Amazing! I love seeing all the different 'creatures' that appear.
I tried to programm it myself and it looks so unbelievable cool with the right tweaking with the variables. Thanks for the vid with the explainations :)
I could watch this simulation running for hours, specially with this nice soft music
Why not add reactions? In where one particle can change the type of another. I wonder what complexity could arise then!
I understand pretty much nothing that you did but i would like to suggest you to add a second matrix for short range interaction and see what happens
I wonder if you add into this some kind of chemical reaction where when some specific particles get too close they would transform to other ones. I would like to see if something like reproduction can easily appear in this simulations also
Thank you for making this video, this is one of the coolest things I've seen in a while
I’ve been thinking about doing this for a month. Thank you for showing me the possibilities; this is fantastic!
Also I love the music. Where might I find some? Or even, what’s this genre called?
Thanks.
The dude explains everything simply and makes school look like a fool
Where do all of these particles get the energy to move from?
What if you add the idea of energy, functions for each particle (like nucleus and mitochondria dna production etc), and the ability for cells to come together but they have like a membrane against cells of the same type / membrane
iv thought of some things to try that may make it even more realistic. or et very least may have intertesting results.
nonlinear force curve. could result in more subtle interaction types
"coupling" force emulation (i.e particles that get very close "stick" aggressively and rigidly to each other until they are far enough apart)
tailored particle fields based on real world organic elements like hydrogen and carbon.
multi field models. electromagnetic and strong/weak force emulation.
Do the particles have velocities of their own? Or is everything in response to the others