This is the most sophisticated artificial life project I have seen so far. The fact that you actually visualize the food in the stomach is insane to me!
@@Kram1032 cell lab: evolution sandbox is an android app simulator of microscopic multicellular organisms. It is extremely complicated and you can design proper regeneration, immortality, etc
Have you tried leaving bibites corpses behind when they die instead of just exploding into giblets? The biting off system seems like it would work well with it, if not just treating a corpse like a single large meat pellet. Remaining as a larger piece would (hopefully) incentivize mouths evolved to tear of chunks instead of just being able to eat the little giblets that they explode into. Edit: it could also allow for some evolved physical traits to be impactful after death (poisonous, hard shell/tough flesh, etc.) possibly resulting in niches like scavengers due to corpses not being able to be fully eaten effectively or more specialized bibites that can eat and digest stuff others can't (poison, etc.).
And bones! Some animals eat those and even NEED those in their diet. Wanna know a little fact you will hate for the rest of your life? Deer eat baby birds that fall off the trees, they are basically free food packets, and if im not mistaken its because deer sometimes need an extra bit of calcium in their diet, so... you know... crunch crunch. Yeah i told you it was a horrible thing to know, but knowing why it happens opens a lot of possibilities. For this simulation we could start developing a sistem where food is more than just a bar to be filled, it could have different properties and materials which are needed for different functions, just like in the real world. And yeah, sadly its gonna be so fcking complex to make, just thinking about all the adaptations needed, all the different variables to add and how they interact with each other. Its gonna be tough. I bet it would be best to add this after the different plant types idea comes out, as there could be where those different vitamins and stuff could grow and appear.
@@auri1075 Introducing nutrients instead of treating all pellets as balanced diet packets should be on the roadmap at some point! Although I don't think it will have an effect on behaviours like predation, but create more complexity in terms of developing a food chain of different species predating on selected preys, and herbivores searching for specific plants
@@BaroTheMadmanif lasting bodies instead of meat pellets become a thing along with evolving plants, they could mostly replace pellets with their meat/plant matter keeping the traits of the bibite/plant that can influence how bibites react and interact with it.
I like how each bibite has a physically simulated stomach, perhaps parasites could evolve to latch onto food and prey on other bibites from the inside out and be visible inside the stomach simulation eating the food that enters or whatever
@@TheYeetedMeat about that it would be cool if bibtes could evolve diffrent methods of making children like some would be pregnat witch would make them slower but some would lay eggs witch they woulf have to protect and it could start a new neshe witch would be eating eggs
The tale of a lineage of Bibites generated from pure randomness: I started a huge world. Ran it for awhile to get something with the RNG. After about 30 hours IGT at 11.18x timewarp, something had reached generation 8 and there were fewer and fewer randomized genomes spawning. By generation 30, it was unplayably laggy, so I grabbed what I felt like was a representative sample of the diversity there and copied it over to a standard size world with standard settings and viruses on. I ran this world for awhile. And the Bibites grew in population enough that they began to randomly collide. Head to head collisions usually became a stalemate, where they would simply push at each other until one or both Bibites died of starvation. This resulted in evolutionary innovation. To solve this conflict, the Bibites developed an instinct to attack whenever there were lots of pellets around. This ensured one Bibite would win the fight by killing the other. Populations continued to grow. But this created a new problem. At a population of over 150, deaths in combat became more frequent, and Bibites became more lethal in combat. Sometimes even killing their own offspring or parents. Some Bibites had evolved innovative techniques. One of them went really fast (20 u/s) if there was a Bibite nearby within its vision and really slow (3 u/s) otherwise. This increased the likelihood that it would either win the fight by hitting them in the side, or avoid the fight by running past them. Bibites also learned that laying their eggs near meat was a bad idea. Another Bibite had a different idea. Instead of going faster when another Bibite was nearby, it would go in reverse. This lineage eventually died out though, but that does not invalidate its actions. Many generations later, the Bibites had settled on a radical new solution: altruism guided by self-preservation. Upon approaching each other, they would use a mix of herding behavior and lowering their speed and turning in order to negotiate past one another. They even made cute little head shaking motions. These Bibites had incredibly complex brains compared to their ancestors, with 24 separate synapses. The population rose to 300 while the plant pellet count dropped to 25000 Energy out of 450000. Most energy was unused, as the Bibites were essentially using their huge, peacefully cooperating population to extract all the resources they could harmoniously. They had even forgotten their combat instincts completely by Hour 144 of this world. I declared that world a success, but moved on and imported a representative sample of the now highly monomorphic population to a new world. Size 500 but with 10x the density and growth rate of plants, which means 10% more overall resources. The one very important difference is that this world has merge pellets enabled, meaning the pellets combine into gigantic planetoids. For awhile, the population just grew and grew. It eventually reached 500 as they got a bit more comfortable with physical contact and stabilized around 550, waiting their turn for more food or scrambling to try to get to the super pellets first or find open spaces. That was before an important new development... The reemergence of violence. The Bibites redeveloped their violent tendencies around 80 hours in, murdering their way to the food planetoids instead of waiting for a space to open up. With so many Bibites and so little grazing space. This lowered the population to 120 individuals. Now, their growth rate began to skyrocket, as there was little hope for an efficient, long, and prosperous life anymore. Instead, they tried to live fast, have kids within 6 minutes birth (maturity was under 3 minutes), and inevitably die young from the violence. Size also works as a decent defense against violence, as the higher health means an attack is not as lethal. This also changed the layout. There were more and small planetoids to eat instead of a few big ones, and more normal-sized pellets too. Things carried on this way for awhile. One consequence of rapid growth is that they could reach huge sizes and didn't stop growing into adulthood. Consequently, I've seen at least one example of a Bibite with a weight of 35, a Maturity of 9.5, and an age of.... 45 minutes. Yeah. They almost never live past an hour even though senescence is 8 or 9 hours out. They either wander into the void and starve due to over-reproduction draining their energy before they can find their way back, or get ganged up on and killed. Another more recent development is the use of grabbing and ungrabbing every clock Tic except when they're full. This allows them to throw other bibites or themselves away from food or into the void, or get a second approach at another Bibite's tail or side while it's eating. It also makes a way to unlock a grab on a Bibite to avoid starving. At around 260 hours, I went into the save file and increased the minimum efficiency of meat from 15% to 45% to hopefully make it possible for predators and scavengers to evolve. Though none have evolved yet at 300 hours of 11.18x warp. TL;DR: evolution taught inept bibites to survive, confused Bibites to fight, warring Bibites to make peace, peaceful Bibites to fight even harder, and fighting bibites to throw things.
You had an entire Bibite civilization running on your computer in absolute harmony, and the moment food became big it was doomed by greed. You could write a book on this.
I'm excited about BIOME! It's important for a digital lifeform to have the ability to grow past its developer's perspective, and giving them the ability to make new body parts and genes themselves is a big step toward hitting that milestone.
Just thinking about physical formations that restrict their movement like rocky areas or water areas. I know the idea is to start with more basic warm cold kinda biomes, but its the first step into complex environments and im super excited about it, as in real life, environment is one of the most important variables for an animal to evolve and develop.
@@auri1075 like some creatures would have to evlove fur to survive in cold but thanks to it would begin to die if in desert and mabye in the desert there would be less plants
Other alternative approaches include: 1. Allowing bibites to engineer structures and machines 2. Allow bibites to self-assemble into colonies, complete with communication and actuation. There's at least 1 arxiv paper on this self-assembly
Idea, apply water in certain pools, water is a competitive resource everyone needs. this allows for more competition for the same area, which allows for feeding to happen more often by herbivores. edit: to all those who question if bibites are aquatic, IDK if they are, just wanted to say that, if they aren't, this would be cool.
That would be very interesting to see. I dont know how it would work with the bibites, but water could have different depths, some depths can be crossed by walking some dont. You could even add waterflow. The effect of different terrains, muddy areas would make it harder to move. Though i have doubts of what the bibites are so im not sure if it would apply to them. I mean, they look like microorganisms, in which case most of what i said can not be applied to them.
@@spironspirit6502 epidemics simulation in a population of bibites? Add some variables that affect behaviour, stats, and other capacities of bibites these are illnesses and they pass onto nearby bibites and evolve separately. They stick to bibites, and then infect them, but if the bibites wash themselves on water they dont get the illness. Would be nice to see if some illnesses would remain or if some would dissappear instantly because they are way too detrimental making the bibite not even able to infect others.
@@the_cheese_cultist if that is true, then it wont be for long. biomes have been confirmed to come in a future update. this will add things like deserts and forests, which are obviously above ground.
I have a lot to say about this. First, it's obvious you put an enormous amount of efforts in your video, from editing, simulations, transitions, little 5 seconds jokes that must have took an hour each to get right, it's all really impressive and it's obviously not easy. Now, for digestion, having been interested in A-life for about a decade, you are the first one I saw implementing digestion, I never thought of that, and I would have thought of it as an oversimulation problem, where someone goes too deep on insignificant details, and introduces instability to the simulation without meaningful gain. However, you've shown me that it actually makes a lot of sense. Of course, it's obvious now that digestion is extremely important to the whole process of allowing differentiation between herbivores and predators. More than that, carnivores like cats sometime munch on grass to aid digestion, and some herbivores are opportunist, like horses that would eat a bit of meat if it was just laying there. That it emerged in your simulation is not only eye-opening, it's a big hint that you're on the right track. I wish to discuss more about it because it's a genius bit of work, be proud! The bit about bugs in the initial implementation are funny, but what's interesting is the "multiply food to aid in locomotion" one. It's interesting to me because it's not the only simulation in which it happened. Same with the infinite food bit, that one happened to carykh, where I think his alife was allowed to vomit food. The goal was to feed babies or cooperate and share, but his alife made use of a bug and learned to vomit more than they ate without dying, creating food out of nowhere and trivializing survival. I have never seen the nuclear fission before though, that one was funny. If you'll allow me another story, this time one of mine, I was working on Tamagotchis, that little handheld toy, popular in the 2000s. Except mines were nothing more than bitpacked structs representing their health, hunger, sickness, dirtiness, etc. They also had a NEAT brain in charge of the body. The experiment was (at first), really simple : The Tamagotchi had to learn to cry for food if it was hungry, or cry for medicine if it was sick, you get the point. However, they only had 10 food pellets total. The ones who survived the longest would reproduce with the other winners, rinse and repeat. Classic stuff so far. One Tamagotchi was a tiny being in memory, something like
A simple way to mix advanced vision with plant evolution is to make different materials (wood, leaves, meat, etc.) have different colors, and to implement peripheral vision, wich would tell the bibit what color the material is, instead of what the material is. It could make some bibits evolve to have worse vision, but better food collection, and others to have better vision, but only able to get certain foods (i.e. herbivore that can eat any green thing, but can’t distinguish between each type of green thing and carnivore that can only eat perfectly fine meat, but can detect it easily). Now I realize that is NOT simple at all lol
For the biomes: Some organ like the elephant ears to cool down when there is too much heat. Fluffy bibites for the cold. Fat: with high water content for arid areas and with high energy content for frozen areas. Day night cycle! And bioluminescence to keep contact with similar bibites in the night. Night vision which is very poor during the day and viceversa for day vision. This could also be a stepping stone for sounds and hearing! Decomposing meat? In hot environments it would happen faster than in colder environments.
That last part reminds me of something I heard about animals who use cold waters as refrigerators to avoid rotting. Could be done here by making some parts of biomes more extreme than others
I just realized, predation, and this simulation, is the perfect example of "when a measure becomes a target, it ceases to be a good measure". Like, "does predation evolve" is a good measure for how lifelike the simulation is, but if you were to hard code in predation, just to make it happen, it would not make the simulation any more lifelike. Not that you are doing that.
A very good point, and one that has made me a bit uneasy in hearing about his quest to "make predation happen", since then it could cease to be a truly emergent phenomenon like he is hoping.
@@HansLemurson The fact it isn't happening shows there's something wrong with his simulation. He's not trying to force predation, he's just trying to make it possible to begin with. So long as it is impossible to sustain, it means that his simulation is inaccurate.
Just an idea, but if you add the vision system idea, you could let bibites influence some of the bits? Like a carnivore could hide in a field of plants, having density very similar to the plants, aswell as other features. This could allow camouflage, waiting patiently until a prey item comes, and the predator strikes! Then 1 last simple bit, mouth strength should probably affect throwing, so a herbivore can't just suddenly evolve to throw in a random direction and instantly kill a carnivore on accident. Just a few ideas though, and have a nice day!
A simple/basic way to do this (To test it out at least) would be in my opinion to assign colors to stuff be it a plant or a Bibit and in the vision system color affinity. That would mean that plants and Bibits can evolve to have different colorations and be colorblind-er to specific colors so that for example. Blue visioned preditors would struggle to see red-colored prey and green visioned prey would struggle to see blue-colored preditors. In the future, this can be refined/overhauled to included shape and size so instead of just not seeing certain colors, they would have to have the ability to distinguish between similar looking plants and Bibits(both prey and preditors) Another note. I would be cool if they would have to find mates to produce offspring who have a random assortment of parents gens. Allthough this would require much larger simulations.
Does the current level of energy reserves affect the amount/quality of the meat dropped on death? Makes realistic sense that a "healty" individual would have more/better meat than a starving one and could potentially encourage scavengers to transition to active hunting if relevant enough.
This is continuing to be a very interesting project to follow. An idea for future filler content: Maybe look at some simulations and pick out a few good species to really dive into? Like go through their life cycle and explain the various characteristics and how they evolved to get them. Would be kinda like a mini nature documentary. Can't wait to see what you show off next.
@@TheBibitesDigitalLife With the most recent version I’mma drop another $20 because shut up and take my money 😅 I wanted to do more but budget is tight so will start uploading unusual Bibites to Google docs, recording with OBS, and just really anything I can do to share data and to spread it to other people also interested in this stuff. Again, you are 100% on point and heading in the right direction with this.
Another few ideas: what if the bibits had a gene that controlled egg incubation time and egg hardness. Then make bibits able to eat eggs and the hardness is a roadblock that makes them require stronger jaws to break they harder they get, eventually requiring jaws so strong it is a specialization. Make jaw strength proportional to damage. Make big bibits able to swallow tiny bibits whole. Reintergrate disease and parasites.
idea for plants: fruits, veggies, and mushrooms, fruits are easier to digest and give more type "A" food, while veggies are larger but give are slower to digest with more "B" food than A, both give both types, just have a more focus on one at a time. Mushrooms would give both "A" and "B" type food and grow off of meat and plants after a time. They give little resource but are super easy to digest.
Mushrooms on earth aren’t plants though, they’re more closely related to animals than plants, and don’t just grow wherever decaying stuff is but instead grow where spores are able to form networks. They seem a bit too complicated to be added with the evolution of plants
@@zackbuildit88 true, but with the format I had, I just decided not to put the fact in parenthesis, probably because I didn't think of it. Thanks for the info, anyway!
@@spironspirit6502 what if instead of just having mushrooms growing like a plant they could grow on parts of meat that haven't been eaten or smaller parts of plants
@@revimfadli4666 biodegraders almost identical to fungi have evolved thousands of times independently so actually it's more likely to be exact same situation than radically different
This is only the 3rd of your videos that I've seen so far, and what I've seen feels like what the "cell stage" of Spore SHOULD have been. I remember seeing the very early game development footage way back in the 00's and being so excited for the kind of simulation they seemed to be making, only to be so disappointed when the game finally came out and the cell stage was over and done with in 15 minutes. Bravo dude, you're doing what EA failed to do and making me feel that same excitement again 15 years later! ^^
I really want to know how this project fits into actual scientific research. I certainly have never heard of anything even close to it, which obviously doesn't mean much because I'm not a biologist, but I'm just personally convinced that you're advancing the field. At the end of all this you should totally try publishing a paper detailing your processes and results. Especially if that BIOME system amounts to anything, I believe that alone would advance the field of evolutionary simulation a ton, possibly some other fields like Machine Learning, who knows what the scientific community can make of it. Your work is a true gift to humanity, thank you so much.
I actually presented the project at the ALIFE2020 conference! So much awesome stuff and research. A few students have also asked me if they could use the project for their research/papers, which I absolutely agreed to, but I don't think I saw the results yet 🤔
Yea I agree that this is an interesting project and could be a serious research project. If you haven't already, maybe reach out to some biologists working on similar computational modeling projects and get their input. As a biologist it definitely interests me, I can see how much you're learning and implementing and I feel like there are some core fundamentals of ecology, feeding behavior, nutrient sensing, evolution, and neurobiology that would be cool to implement and might allow for greater flexibility/similarity to real population dynamics. You're definitely on the right track by looking at digestion, but there are a lot of other dynamics that go into predator/prey interactions that are as important, such as the number of offspring, lifespan, food scarcity, mating, etc. You could definitely find some experts who would be interested in talking with you about the projects or letting you audit some courses
@@TheBibitesDigitalLife Nvm I’m editing this because I realised you talked about making plants harder to digest, but here are some ideas that I think would be great: 1. Saturation. When eating, a pellet should yield a status effect that decreases the amount of energy used in by multiplying it by some fraction, but not in thinking and reproducing. The time this lasts should be higher in meat, and there should be a few genes associated: Rumination (might not be a good name) increases the energy loss of initiating saturation upon feeding in exchange for the fraction energy use is multiplied by being smaller, but should actually be a debuff unless feeding is extending saturation time rather than starting it, essentially making clustered food better than spaced out food, and Saturation Child Imparting (again a bad name), which determines how much extra energy a parent bibite would use up in making it’s child bibite in exchange for giving it saturation.
@@TheBibitesDigitalLife and 2. Plants should probably have evolvable vascular systems and thorns. Thorns on plants deal damage on contact and in the digestive system, and can be detected by bibites, but bibites can also specialise to eat thorns better with less damage and absolutely no cost over their lifetime, but requiring more energy to be born without the extra energy being imparted upon them. The vascular system meanwhile should be for multi-segmented plants and would let them pass water and sugars (energy) through their bodies in exchange for a small bit of said resources, even if part of their body is in a dry area where it cannot get water by itself.
I think something you're missing in your simulation of food is a system to model nutrients. If energy was all that mattered we'd all be optimized to consume as much fat as possible, but we need nutrients and proteins to function properly. A way to implement this would be to have organism have a "nutrient hirerchy" that determines what genes they'll put their nutrients to support. More developed and specialized genes would require more nutrients to maintain. So instead it maxing each trait organisms would have to choose what is most beneficial given the limited resources. Also consider adding a sleep mechanism. Could be implemented by having energy efficiency decrease as the organism gets more "tired". So the longer they're awake the more expensive it becomes to function. To go along with this a night mode in which organisms will have a harder time "seeing" their surroundings
Marvelous! Keep it up! I'm incredibly impressed with the sophistication of your project, I can't think of anybody better qualified than a systems engineer.
I'm really hyped for how this project will develop! An idea for intermediate filler content you could try to put out between these big videos, I think it would be interesting to see what real organisms and behaviors have influenced what you have done thus far and/or what you'd like to do in the future. It might be a good way to educate your base and to get people thinking about new dynamics that could help the project.
This is really cool! I can’t wait to see how you add more senses like smell and hearing to the bibits, I think it would add a lot more depth to the different niches and behaviors.
I know these videos take a crazy amount of time to make, but you do a damn good job here! This is entertaining and educational and I patiently look forward on the future of this project and channel.
New materials sound like a cool idea, perhaps creatures could evolve to grip rocks and throw them at other creatures, dealing massive damage. I think I've seen that happen before here but rocks could probably just do more damage than plants or meat.
Maybe damage would be relative to the material's hardness and weight, the surface of impact and the amount of energy absorbed on impact (if I throw a cube of plastic and it scratches your side, you wouldn't get as hurt as if you took it in the face or belly)
This is something I have been thinking for a while now, and now I’m waiting for the premiere to begin. And this is the kind of content I watch:) I had something to ask you to implement but I forgot what it was
This is amazing. For all of the ideas that were being thrown around about how to get predation, this is by far the simplest solution. This project is in good hands for sure.
One idea to maybe add: Rocks. - For digestion. Like, chicken and some other birds will actually eat gravel to aid digestion, grinding up the stuff they eat, increasing efficiency, freeing up more nutrients overall. And likewise, indigestible plant matter actually can also be helpful with, uh, gastric flow I guess? Some stuff tends to get stuck more easily, causing blockages. And the indigestible parts of plants may help that along. So I think actually simulating waste disposal, complete with keeping track of how many nutrients are left in it and how tricky it is to break down, would be an excellent addition.
The rock idea would work best for animals which eat tougher food pellets. As those proposed in the vid. This would make so many different behaviours for survival i love it.
Hello there! I am a biology student at BYU and I was wondering if corpses could be added as a meal sorce on the simulation. Instead of creating chunks of meat when an animal die, you could make a corpse of the dead creature. This will open way more niches for onivores and opportunistic animals. Also, as a corpse rots, the less energy it can provide, or at least, the harder it is to digest a rotten meat. I love the simulation and hope it gets more attention on youtube!
7:17 i have a GREAT feature idea: if a bibite tries to load more acid than there is room in the stomach, it will vomit. Bibites could even learn to vomit as a defence mechanism, if you add acid damage
im amazed that you manage to not only create a very deep simulation of life, but you also explain it very well and with good editing, its the hole package
It's the result of a lot of work and practice 😅 If you look at my first videos the simulation was not there yet, and my editing skills were non-existent!
maybe an interesting idea: to help make scavenging a more viable niche, have bibites lay eggs when they have children instead of just spawning them, they could have even more nutrient density than meat but wouldn't be viable for herbivores to eat because of the need for specialised jaws to break the egg's shell.
They do actually lay eggs. I haven't used the simulator much so I don't know why the eggs are there (though sometimes there isn't enough energy in them so it dies), but the PDF linked in the Bibite neural network viewer doesn't mention looking for eggs or anything, so I guess it hasn't been implemented.
@@elliotlea5457 i know they lay eggs yet feldbiss's idea is more realistic due to eggs taking some time to hatch. this would actualy make the scavenger niche alot more viable and make it easier to survive as a ferocious carnivore.
I like seeing the Dani editing style on someone who's calmer. I think it's pretty inherently funny. Otherwise, also, epic content. Love to see sophisticated evolution simulations in friendly, understandable game-like environments. Keep up the good work!
Dude, that digestion system is overpowered! I do love the BIOME plan, too. Edit: Thinking about the koala example made me think of a thing you could do if your brain simulations become too complex, which is to "Tax" intelligence by making it more expensive to have a more complex brain.
a simple concept that i had considered long ago was simply making the digestion work as so: for each percentage point of plant digestion efficiency, it costs three units, and for each percentage point of meat digestion efficiency, it costs one unit; where the unit is adjusted to balance things reasonably (expect herbivores to hover around 50% efficiency)
Aha! I'm glad you added digestion, it's such a huge part of the difference between herbivores and carnivors and I'm glad it played out well in the simulation
I've been simulating a single system since this update came out. A really small change that i think would work wonders is another "cheat" neuron/synapse. Just have two vision input neurons attached to the rotate or accelerate output neurons instead of just one. At the moment it's nearly impossible for predictors to successfully evolve a second sight neuron without screwing up their developed pathfinding. But if they should spawn "virgin" with two sight neurons i think it would SERIOUSLY help with hunting behavior.
Just wondering will the new update be coming after the premiere or before the premiere. (just exited for the new visuals as i dont know what is what in my average simulations) Thanks
I really think that the most important part of any evolution simulator is ecosystems, especially varied and changing ecosystems. Right now, the most you have is areas of higher or lower concentrations of food, which doesn't impact a whole lot. Having different areas that affect the organisms promotes different adaptations to fill different niches, allowing for a much wider range of biodiversity. It could help to solve the issue of transitioning from herbivore to carnivore, by allowing the species to gradually evolve through different situations rather than hoping that the jump happens naturally and is sustainable. Some variables that you can have for different environments could be the temperature, having more or less concentrations of different resources (you'd have to add in more resources to compete for before adding this one), or environmental factors like mud that slows down one movement type but promotes another, or hiding places for certain body types. I saw in another comment suggesting adding pools of water, which I think would be a great place to start. Water itself is a valuable resource that can be competed for and specialized around. Stemming from this, you could add mobility options, with a spectrum going from fins to legs, with fins being most useful for water, legs most useful for land, and somewhere in the middle for the shoreline (moving between water and land regularly). Inside the water, there could be one ecosystem with aquatic herbivore and carnivore niches, on the shoreline there could evolve to be a niche that dominates the shoreline as bibites go to drink water, and on land there could be a huge range of biodiversity and niches depending on an area's access to water (the plant matter could have a different spawning frequency or even quality depending on the proximity to water, and the bibites themselves would have to adapt around the availability of water for themselves depending on what area they live in). From there, you could implement environmental change, having the bodies of water shift and change size and shape, which would create an ever changing simulated world with a high biodiversity. And that's just one example stemming from a single idea, you could take the premise in any direction you wanted and run with it. I should probably say that this is all just a theory of mine, and I have absolutely no credibility at all, not in biology or coding. This entire thing is basically a guess formed from what little I know and understand.
something i might reccomend is having tthe digestion also depend on how "chewed" the food is , with smaller pellets digesting with greater efficiency. you might need to adjust how the biting works so that its an actual "decision" on how to evolve and not just always pup everything. not sure if its there already but a bibit that dies of starvation should drop less meat than a hunted bibit that has energy banked. and bibits that have a large amount of energy should be visually bigger to a predator.
Could you implement something like grass? Or something that predators can use as camouflage if evolved in this assassin like traits. This will help predators while herbivores can evolved to avoid grass and maybe evolved high enough sense that can detect the predators even while camouflage
Would be great if combined with other required resources like water. Creating Bibites who prefer to wait in the water and go on the "ground" to eat, and the other way around with predators who are less visible in the water attacking from it. With any creatures who needs water becoming their prey. (Like crocodiles)
Grasses are one of the most advanced plant types ever. They do a weird asexual reproduction colonial thing in addition to having sex. That probably happened, but it is not representative of how the first carnivores appeared (they actually appeared before herbivores, because they appeared before plants, because being a plant is hard).
You could allow a waste excretion system that produces waste from digested food. Bibites can eat this with a cost of their health over time (disease), however they can evolve to prevent this. The waste material would be extremely easy to chew and easy to digest but wouldn't be at all energy dense. Creatures that use waste as their primary source of food require the living of other species to eat and reproduce.
Potentially changing meat after its been eaten it becomes a grey crescent to represent a striped corpse. A high jaw (bite) power variable should be needed to be able eat the corpse. This would mean you have to add 1 new variable but you could try to update the combat by allowing different builds (I want to see a giant herbivore with a massive tail). claw / bite / tail power/speed variables could be added to all the bibits, so you can workout the energy and the power of each attack.
Loving the new update! Good stuff! I have some suggestions for future updates. It would be cool to see the number of eggs a bibite has layed, and/or the number of offspring that is still alive. The main thing I would love to see is an option to turn off the visual simulation. That way we can leave our bibites running while afk or while the game is in the background. I think it could really speed up the simulation and speed up the fun!
Leaving a comment to hopefully help boost this in the algorithm. I've always been fascinated with the idea of extremely in depth evolution simulations and you're really going the extra mile to put in all the details I feel is lacking in any other simulation.
All the things you're adding to the bibites before plant evolution make perfect sense and I think the modules system would be the same level of if not more interesting than plant evolution, but don't overwork yourself, we can wait! The bibites is way too underrated, it's easily the best life simulation so far!
Really love this video! I think with your carrying capacity problem you're running into an issue where predators can completely depopulate a species before dying off themselves. If you can buff the last survivors of a species by making them more difficult to find or kill, you'll see more die off from predators that over-exploit prey. So you can get into regular cycles of increasing and decreasing predator prey populations like real life.
Nectar or fruit seem like a pretty logical material, since it makes it easier to eat plant matter but aids the plant in some way. Rocks being used to help with digestion, a la chicken gizzards, would also be a good addition. And while you're at it, developing poisons as both a factor a material can have and a skill that can be evolved feels like another fun addition spinning off of this inclusion.
I feel like a big thing that could improve practically everything is working on colonies of bibites or something similar. Having multicellular creatures allows more variety in designs, as well as more variety in general. Maybe this could be done by more pheremones, or ways for bibites to transfer food and energy to others
Making plants evolve some sort of self defense mechanisms would hinder the herbivores and more accurately mimic nature. I would have no clue how to code that, but that's why we watch Bibites!!!
this simulation is so cool and complex, I don’t know how you managed to code it all without going insane. The digestion thing is really really neat, expecially the visualization.
Absolutely love it ! As a fellow doctor who has high levels of enthusiasm in evolution and microbiology i absolutely simp for your con bro ! Eished more ppl were to actually care more for such enlightenments than to low tier gaming streams and such
I'm just discovering this channel and project, and I... I don't have enough words What you are doing is insane I had this idea on my mind as a kid, as a fantasy, to be able (humans in general, not me particularly) someday to program an entire world with several species, complex interactions and all... and what you are doing might very well be one of the first projects like this ever!! I cannot wait to see how it keeps on improving, and to try to become involved in this community
I burst out laughing at the Bibite's immediate abuse of splitting food infinitely as a form of transport, that's just perfect-- Not only infinite food, but energy-free travel LOL
This is a great project, many thanks to all contributors and mr. idea father here. Hopefully it will reach a point of essential completion so I can basque in its ever growing glory for many days to come.
I've wanted to do something like this for a long time. One thing to keep in mind is things like digestion are abstractions of the emergent properties of chemistry and physics. The abstraction makes it easier to model but isolates it. So with the meat vs plant model it ignores that their properties come from things like, amount of fats, carbohydrates, proteins, etc, and that these things themselves have different properties. Organisms therefore can be more plant like or more animal like, assuming that 'plant' and 'animal' even continue to have useful meanings. Anyway for a practical suggestion, have you tried implementing a smell based system? Something like a cellular automata which allows an organism to detect concentrations of different materials or different species.
Absolutely love where this channel is heading. A system where bibites can evolve new body parts would be amazing to see, you could even go so far as to have each new bodypart / functional unit of body having its own material stats that also effect how it functions (obvi in the distant future lol). Very exiting!!
You could also make it so bibites able to move in different ways, like flying could use lots of, very fast, but they would only need to use it for short bursts because they can glide. Or slithering like a snake which could give it certain advantages. Making them quadrupeds or bipeds and making it affect their lifestyle and movement patterns. Mating and more complex relationships would also be amazing to see. Making shelters like nests or dens and giving them the abilities to affect the things around them, other than eating them. I think this is super cool and it’s so awesome that you’re doing this.
This project is legendary, You are legendary and anyone who helps You is legendary. You basically made one of the most life-like artificial life simulation and gave it for free. This is going down in history
I always wanted a game where the life in the game was apart of a ever evolving system. This project seems like its paving the ground work for this sort of thing in the future
i think another good thing to apply to the program is adding another resource (lets just call it 'materials' or 'nutrients' for now) and /waste/. this would force the bibits to also manage how many resources they have in their bodies at once, perhaps even splitting the nutrient resources into proteins and minerals or so, and having too much of one or both can cause damage to the bibit, while evolving methods to accommodate having more or less tolerance to either depending on the environment. this would ALSO encourage the diversification of the scavengers, as these resources (plus some free floating energy) would be left behind if something dies and rots as well
You should also consider that smell is as big of a factor for animals finding the correct plants to eat. Smell is an “aura” that is sensed in a certain area depending on wind and smell strength. But I think it should be considered.
I love all this changes ! I just have one remark : in your current simulation, carnivores can't eat plants at all or they will lose energy, while even full herbivores can eat a little bit of meat sometimes. It looks like the opposite of real life, in which many "carnivores" are in fact omnivores that eat slightly more meat than plants (i think all mammalians carnivores work like this for example), so it may be a bit irrealistic in my opinion great work anyway !!
Aside from the plant they get from eating the stomachs of their prey, ferrets aren't able to eat plants at all. They're incapable of digesting them and then get clogged up and die. They're the only true carnivorous mammal I can think of. I mean it's not like plants from the stomach of something that can handle it matter much.. it's already partially digested for them.
I think making the creature's color a deciding factor in camouflage would do great as it would allow for predators and prey to hunt and be hunted more dynamicly.
I have a fun idea: Collab with someone else who has an evolution simulator, and add both your creatures and theirs to a single simulation. You can compete to see which is better at adapting.
That sounds fun, but it's VERY impractical. Each simulation has an entire code architecture that can't be translated between one another. Even if the programming language is the same, everything else is different: the functions, the architectures, the genes, the variables and so on
@@pedroheck3667 true but 'd like to see which simulation is more realistic. like what creatures adapt better. is the te one person or this legend. basically it will be another Alife project with predators. than they would set the simulation to run to see which Alife simulation has better CO-EVOLUTION.
I don't know if you still read this comment section, but making blood will allow for a parasite niche, which will then allow for a symbiosis niche to occur of smaller bibites eating the parasites off of larger bibites, assuming bibites can tell the difference between prey and allies
As you might know, a desirable trait for the system to have which will lead to many emergent properties to share with our own world is a measure of conservation; if energy were conserved within the system, with low entropy energy coming in and high entropy energy leaving in equal measure, you will find that many other desirable priorly arbitrary properties emerge.
i love your videos! these kind of simulations made me fall in love with coding. its like finding your favorite music album all over again. Please keep going!
This is the most sophisticated artificial life project I have seen so far. The fact that you actually visualize the food in the stomach is insane to me!
I'd say cell lab is
@@blacky6552 Haven't seen that one
Look at Ecosystem the game. It is very similar but 3D.
@@Kram1032 cell lab: evolution sandbox is an android app simulator of microscopic multicellular organisms. It is extremely complicated and you can design proper regeneration, immortality, etc
@nøc yeah, its a great one.
Have you tried leaving bibites corpses behind when they die instead of just exploding into giblets? The biting off system seems like it would work well with it, if not just treating a corpse like a single large meat pellet. Remaining as a larger piece would (hopefully) incentivize mouths evolved to tear of chunks instead of just being able to eat the little giblets that they explode into.
Edit: it could also allow for some evolved physical traits to be impactful after death (poisonous, hard shell/tough flesh, etc.) possibly resulting in niches like scavengers due to corpses not being able to be fully eaten effectively or more specialized bibites that can eat and digest stuff others can't (poison, etc.).
what I thought too
Thinking the same thing.
And bones! Some animals eat those and even NEED those in their diet.
Wanna know a little fact you will hate for the rest of your life?
Deer eat baby birds that fall off the trees, they are basically free food packets, and if im not mistaken its because deer sometimes need an extra bit of calcium in their diet, so... you know... crunch crunch.
Yeah i told you it was a horrible thing to know, but knowing why it happens opens a lot of possibilities. For this simulation we could start developing a sistem where food is more than just a bar to be filled, it could have different properties and materials which are needed for different functions, just like in the real world. And yeah, sadly its gonna be so fcking complex to make, just thinking about all the adaptations needed, all the different variables to add and how they interact with each other. Its gonna be tough.
I bet it would be best to add this after the different plant types idea comes out, as there could be where those different vitamins and stuff could grow and appear.
@@auri1075 Introducing nutrients instead of treating all pellets as balanced diet packets should be on the roadmap at some point! Although I don't think it will have an effect on behaviours like predation, but create more complexity in terms of developing a food chain of different species predating on selected preys, and herbivores searching for specific plants
@@BaroTheMadmanif lasting bodies instead of meat pellets become a thing along with evolving plants, they could mostly replace pellets with their meat/plant matter keeping the traits of the bibite/plant that can influence how bibites react and interact with it.
I like how each bibite has a physically simulated stomach, perhaps parasites could evolve to latch onto food and prey on other bibites from the inside out and be visible inside the stomach simulation eating the food that enters or whatever
This is genius!
Was just thinking about a system like this!
bibites inside bibites
@@sampy901 bibiteception and bibite pregnancy
@@TheYeetedMeat about that it would be cool if bibtes could evolve diffrent methods of making children like some would be pregnat witch would make them slower but some would lay eggs witch they woulf have to protect and it could start a new neshe witch would be eating eggs
The tale of a lineage of Bibites generated from pure randomness:
I started a huge world. Ran it for awhile to get something with the RNG. After about 30 hours IGT at 11.18x timewarp, something had reached generation 8 and there were fewer and fewer randomized genomes spawning.
By generation 30, it was unplayably laggy, so I grabbed what I felt like was a representative sample of the diversity there and copied it over to a standard size world with standard settings and viruses on.
I ran this world for awhile. And the Bibites grew in population enough that they began to randomly collide. Head to head collisions usually became a stalemate, where they would simply push at each other until one or both Bibites died of starvation.
This resulted in evolutionary innovation. To solve this conflict, the Bibites developed an instinct to attack whenever there were lots of pellets around. This ensured one Bibite would win the fight by killing the other.
Populations continued to grow. But this created a new problem. At a population of over 150, deaths in combat became more frequent, and Bibites became more lethal in combat. Sometimes even killing their own offspring or parents. Some Bibites had evolved innovative techniques. One of them went really fast (20 u/s) if there was a Bibite nearby within its vision and really slow (3 u/s) otherwise. This increased the likelihood that it would either win the fight by hitting them in the side, or avoid the fight by running past them. Bibites also learned that laying their eggs near meat was a bad idea.
Another Bibite had a different idea. Instead of going faster when another Bibite was nearby, it would go in reverse. This lineage eventually died out though, but that does not invalidate its actions.
Many generations later, the Bibites had settled on a radical new solution: altruism guided by self-preservation. Upon approaching each other, they would use a mix of herding behavior and lowering their speed and turning in order to negotiate past one another. They even made cute little head shaking motions. These Bibites had incredibly complex brains compared to their ancestors, with 24 separate synapses. The population rose to 300 while the plant pellet count dropped to 25000 Energy out of 450000. Most energy was unused, as the Bibites were essentially using their huge, peacefully cooperating population to extract all the resources they could harmoniously. They had even forgotten their combat instincts completely by Hour 144 of this world.
I declared that world a success, but moved on and imported a representative sample of the now highly monomorphic population to a new world. Size 500 but with 10x the density and growth rate of plants, which means 10% more overall resources. The one very important difference is that this world has merge pellets enabled, meaning the pellets combine into gigantic planetoids.
For awhile, the population just grew and grew. It eventually reached 500 as they got a bit more comfortable with physical contact and stabilized around 550, waiting their turn for more food or scrambling to try to get to the super pellets first or find open spaces.
That was before an important new development... The reemergence of violence. The Bibites redeveloped their violent tendencies around 80 hours in, murdering their way to the food planetoids instead of waiting for a space to open up. With so many Bibites and so little grazing space. This lowered the population to 120 individuals. Now, their growth rate began to skyrocket, as there was little hope for an efficient, long, and prosperous life anymore. Instead, they tried to live fast, have kids within 6 minutes birth (maturity was under 3 minutes), and inevitably die young from the violence. Size also works as a decent defense against violence, as the higher health means an attack is not as lethal. This also changed the layout. There were more and small planetoids to eat instead of a few big ones, and more normal-sized pellets too.
Things carried on this way for awhile. One consequence of rapid growth is that they could reach huge sizes and didn't stop growing into adulthood. Consequently, I've seen at least one example of a Bibite with a weight of 35, a Maturity of 9.5, and an age of.... 45 minutes. Yeah. They almost never live past an hour even though senescence is 8 or 9 hours out. They either wander into the void and starve due to over-reproduction draining their energy before they can find their way back, or get ganged up on and killed.
Another more recent development is the use of grabbing and ungrabbing every clock Tic except when they're full. This allows them to throw other bibites or themselves away from food or into the void, or get a second approach at another Bibite's tail or side while it's eating. It also makes a way to unlock a grab on a Bibite to avoid starving.
At around 260 hours, I went into the save file and increased the minimum efficiency of meat from 15% to 45% to hopefully make it possible for predators and scavengers to evolve. Though none have evolved yet at 300 hours of 11.18x warp.
TL;DR: evolution taught inept bibites to survive, confused Bibites to fight, warring Bibites to make peace, peaceful Bibites to fight even harder, and fighting bibites to throw things.
You had an entire Bibite civilization running on your computer in absolute harmony, and the moment food became big it was doomed by greed. You could write a book on this.
Sounds like a Bibite Dwarf Fortress...
The moment when you try to simulate life, but instead accidently make a simulation of nuclear fission. It's impressive.
I'm excited about BIOME! It's important for a digital lifeform to have the ability to grow past its developer's perspective, and giving them the ability to make new body parts and genes themselves is a big step toward hitting that milestone.
Just thinking about physical formations that restrict their movement like rocky areas or water areas. I know the idea is to start with more basic warm cold kinda biomes, but its the first step into complex environments and im super excited about it, as in real life, environment is one of the most important variables for an animal to evolve and develop.
@@auri1075 imagine a mountain biome with rock that could form walls and some creatures could climb them
@@jakubgrono9070 exactly! And that could even add the temperatures variate as you go up.
Theres so much to do, so much potential!
@@auri1075 like some creatures would have to evlove fur to survive in cold but thanks to it would begin to die if in desert and mabye in the desert there would be less plants
Other alternative approaches include:
1. Allowing bibites to engineer structures and machines
2. Allow bibites to self-assemble into colonies, complete with communication and actuation. There's at least 1 arxiv paper on this self-assembly
Thumbnail game is getting good. You been killing it homie.
Thanks! Slowly getting the ropes!
Idea, apply water in certain pools, water is a competitive resource everyone needs. this allows for more competition for the same area, which allows for feeding to happen more often by herbivores.
edit: to all those who question if bibites are aquatic, IDK if they are, just wanted to say that, if they aren't, this would be cool.
That would be very interesting to see. I dont know how it would work with the bibites, but water could have different depths, some depths can be crossed by walking some dont. You could even add waterflow. The effect of different terrains, muddy areas would make it harder to move.
Though i have doubts of what the bibites are so im not sure if it would apply to them. I mean, they look like microorganisms, in which case most of what i said can not be applied to them.
@@auri1075 I also think about it cleaning and reducing chances of illness if you stay in it for a while!
I'm pretty sure the bibites takes place underwater, so that wouldn't make much sense
@@spironspirit6502 epidemics simulation in a population of bibites? Add some variables that affect behaviour, stats, and other capacities of bibites these are illnesses and they pass onto nearby bibites and evolve separately. They stick to bibites, and then infect them, but if the bibites wash themselves on water they dont get the illness.
Would be nice to see if some illnesses would remain or if some would dissappear instantly because they are way too detrimental making the bibite not even able to infect others.
@@the_cheese_cultist if that is true, then it wont be for long. biomes have been confirmed to come in a future update. this will add things like deserts and forests, which are obviously above ground.
I have a lot to say about this.
First, it's obvious you put an enormous amount of efforts in your video, from editing, simulations, transitions, little 5 seconds jokes that must have took an hour each to get right, it's all really impressive and it's obviously not easy.
Now, for digestion, having been interested in A-life for about a decade, you are the first one I saw implementing digestion, I never thought of that, and I would have thought of it as an oversimulation problem, where someone goes too deep on insignificant details, and introduces instability to the simulation without meaningful gain.
However, you've shown me that it actually makes a lot of sense. Of course, it's obvious now that digestion is extremely important to the whole process of allowing differentiation between herbivores and predators. More than that, carnivores like cats sometime munch on grass to aid digestion, and some herbivores are opportunist, like horses that would eat a bit of meat if it was just laying there. That it emerged in your simulation is not only eye-opening, it's a big hint that you're on the right track. I wish to discuss more about it because it's a genius bit of work, be proud!
The bit about bugs in the initial implementation are funny, but what's interesting is the "multiply food to aid in locomotion" one. It's interesting to me because it's not the only simulation in which it happened. Same with the infinite food bit, that one happened to carykh, where I think his alife was allowed to vomit food. The goal was to feed babies or cooperate and share, but his alife made use of a bug and learned to vomit more than they ate without dying, creating food out of nowhere and trivializing survival. I have never seen the nuclear fission before though, that one was funny.
If you'll allow me another story, this time one of mine, I was working on Tamagotchis, that little handheld toy, popular in the 2000s. Except mines were nothing more than bitpacked structs representing their health, hunger, sickness, dirtiness, etc. They also had a NEAT brain in charge of the body. The experiment was (at first), really simple : The Tamagotchi had to learn to cry for food if it was hungry, or cry for medicine if it was sick, you get the point. However, they only had 10 food pellets total. The ones who survived the longest would reproduce with the other winners, rinse and repeat. Classic stuff so far. One Tamagotchi was a tiny being in memory, something like
Thanks so much 💪💪💪
And awesome wall of text, awesome examples! It's impressive to me that agents will always evolve to use the bugs in the system
"...like horses that would eat a bit of meat if it was just laying there" Is that comment from you seeing that one video of that baby chick?
@@elliotlea5457 Haha, yeah, 2 such videos
@@bertrandrodruiguez2614 Yeah that one stuck in my mind lol
A simple way to mix advanced vision with plant evolution is to make different materials (wood, leaves, meat, etc.) have different colors, and to implement peripheral vision, wich would tell the bibit what color the material is, instead of what the material is. It could make some bibits evolve to have worse vision, but better food collection, and others to have better vision, but only able to get certain foods (i.e. herbivore that can eat any green thing, but can’t distinguish between each type of green thing and carnivore that can only eat perfectly fine meat, but can detect it easily).
Now I realize that is NOT simple at all lol
For the biomes:
Some organ like the elephant ears to cool down when there is too much heat.
Fluffy bibites for the cold.
Fat: with high water content for arid areas and with high energy content for frozen areas.
Day night cycle! And bioluminescence to keep contact with similar bibites in the night. Night vision which is very poor during the day and viceversa for day vision. This could also be a stepping stone for sounds and hearing!
Decomposing meat? In hot environments it would happen faster than in colder environments.
That last part reminds me of something I heard about animals who use cold waters as refrigerators to avoid rotting. Could be done here by making some parts of biomes more extreme than others
FLUFFY BIBITES
I just realized, predation, and this simulation, is the perfect example of "when a measure becomes a target, it ceases to be a good measure". Like, "does predation evolve" is a good measure for how lifelike the simulation is, but if you were to hard code in predation, just to make it happen, it would not make the simulation any more lifelike. Not that you are doing that.
A very good point, and one that has made me a bit uneasy in hearing about his quest to "make predation happen", since then it could cease to be a truly emergent phenomenon like he is hoping.
@@HansLemurson Well It seems he wants it to emerge, so thats good.
@@HansLemurson The fact it isn't happening shows there's something wrong with his simulation. He's not trying to force predation, he's just trying to make it possible to begin with. So long as it is impossible to sustain, it means that his simulation is inaccurate.
one guy and a bunch of virtual ovals made science interesting to me, how? i honestly don't know.
That's awesome to hear!
I love science
@@TheBibitesDigitalLife underwater themed?
You’ll love Biblidaridion’s evolving an alien biosphere experiment, I highly recommend checking it out!
@@TheBibitesDigitalLifeCan you have the plants evolve too, so the herbivores can diversify more.
Just an idea, but if you add the vision system idea, you could let bibites influence some of the bits? Like a carnivore could hide in a field of plants, having density very similar to the plants, aswell as other features. This could allow camouflage, waiting patiently until a prey item comes, and the predator strikes! Then 1 last simple bit, mouth strength should probably affect throwing, so a herbivore can't just suddenly evolve to throw in a random direction and instantly kill a carnivore on accident. Just a few ideas though, and have a nice day!
A simple/basic way to do this (To test it out at least) would be in my opinion to assign colors to stuff be it a plant or a Bibit and in the vision system color affinity. That would mean that plants and Bibits can evolve to have different colorations and be colorblind-er to specific colors so that for example. Blue visioned preditors would struggle to see red-colored prey and green visioned prey would struggle to see blue-colored preditors.
In the future, this can be refined/overhauled to included shape and size so instead of just not seeing certain colors, they would have to have the ability to distinguish between similar looking plants and Bibits(both prey and preditors)
Another note.
I would be cool if they would have to find mates to produce offspring who have a random assortment of parents gens. Allthough this would require much larger simulations.
Does the current level of energy reserves affect the amount/quality of the meat dropped on death? Makes realistic sense that a "healty" individual would have more/better meat than a starving one and could potentially encourage scavengers to transition to active hunting if relevant enough.
Yep that's the case!
That’s an awesome idea, I’m sure you’re awesome too, your very existence makes this world a better place to live in
This is continuing to be a very interesting project to follow. An idea for future filler content: Maybe look at some simulations and pick out a few good species to really dive into? Like go through their life cycle and explain the various characteristics and how they evolved to get them. Would be kinda like a mini nature documentary. Can't wait to see what you show off next.
Appreciate you’ve very obviously been reading the feedback from your users
Everything is 100% on point and your progress so far has been hella good
I read EVERYTHING
@@TheBibitesDigitalLife With the most recent version I’mma drop another $20 because shut up and take my money 😅
I wanted to do more but budget is tight so will start uploading unusual Bibites to Google docs, recording with OBS, and just really anything I can do to share data and to spread it to other people also interested in this stuff.
Again, you are 100% on point and heading in the right direction with this.
Another few ideas: what if the bibits had a gene that controlled egg incubation time and egg hardness. Then make bibits able to eat eggs and the hardness is a roadblock that makes them require stronger jaws to break they harder they get, eventually requiring jaws so strong it is a specialization.
Make jaw strength proportional to damage.
Make big bibits able to swallow tiny bibits whole.
Reintergrate disease and parasites.
The thumbnail is captivating
Thanks! Slowly getting better
@@TheBibitesDigitalLife👇🚫
idea for plants: fruits, veggies, and mushrooms, fruits are easier to digest and give more type "A" food, while veggies are larger but give are slower to digest with more "B" food than A, both give both types, just have a more focus on one at a time. Mushrooms would give both "A" and "B" type food and grow off of meat and plants after a time. They give little resource but are super easy to digest.
Mushrooms on earth aren’t plants though, they’re more closely related to animals than plants, and don’t just grow wherever decaying stuff is but instead grow where spores are able to form networks. They seem a bit too complicated to be added with the evolution of plants
@@zackbuildit88 true, but with the format I had, I just decided not to put the fact in parenthesis, probably because I didn't think of it. Thanks for the info, anyway!
@@spironspirit6502 what if instead of just having mushrooms growing like a plant they could grow on parts of meat that haven't been eaten or smaller parts of plants
@@zackbuildit88or they could jus be aliens with only superficial similarities to our creatures, like wobbledogs
@@revimfadli4666 biodegraders almost identical to fungi have evolved thousands of times independently so actually it's more likely to be exact same situation than radically different
This is only the 3rd of your videos that I've seen so far, and what I've seen feels like what the "cell stage" of Spore SHOULD have been. I remember seeing the very early game development footage way back in the 00's and being so excited for the kind of simulation they seemed to be making, only to be so disappointed when the game finally came out and the cell stage was over and done with in 15 minutes. Bravo dude, you're doing what EA failed to do and making me feel that same excitement again 15 years later! ^^
I’m in love with this content. It’s great. The cheesy “stock presentation” little bits you do are awesome
I really want to know how this project fits into actual scientific research. I certainly have never heard of anything even close to it, which obviously doesn't mean much because I'm not a biologist, but I'm just personally convinced that you're advancing the field. At the end of all this you should totally try publishing a paper detailing your processes and results.
Especially if that BIOME system amounts to anything, I believe that alone would advance the field of evolutionary simulation a ton, possibly some other fields like Machine Learning, who knows what the scientific community can make of it. Your work is a true gift to humanity, thank you so much.
I actually presented the project at the ALIFE2020 conference! So much awesome stuff and research.
A few students have also asked me if they could use the project for their research/papers, which I absolutely agreed to, but I don't think I saw the results yet 🤔
Yea I agree that this is an interesting project and could be a serious research project. If you haven't already, maybe reach out to some biologists working on similar computational modeling projects and get their input. As a biologist it definitely interests me, I can see how much you're learning and implementing and I feel like there are some core fundamentals of ecology, feeding behavior, nutrient sensing, evolution, and neurobiology that would be cool to implement and might allow for greater flexibility/similarity to real population dynamics. You're definitely on the right track by looking at digestion, but there are a lot of other dynamics that go into predator/prey interactions that are as important, such as the number of offspring, lifespan, food scarcity, mating, etc. You could definitely find some experts who would be interested in talking with you about the projects or letting you audit some courses
@@TheBibitesDigitalLife Nvm I’m editing this because I realised you talked about making plants harder to digest, but here are some ideas that I think would be great: 1. Saturation. When eating, a pellet should yield a status effect that decreases the amount of energy used in by multiplying it by some fraction, but not in thinking and reproducing. The time this lasts should be higher in meat, and there should be a few genes associated: Rumination (might not be a good name) increases the energy loss of initiating saturation upon feeding in exchange for the fraction energy use is multiplied by being smaller, but should actually be a debuff unless feeding is extending saturation time rather than starting it, essentially making clustered food better than spaced out food, and Saturation Child Imparting (again a bad name), which determines how much extra energy a parent bibite would use up in making it’s child bibite in exchange for giving it saturation.
@@TheBibitesDigitalLife and 2. Plants should probably have evolvable vascular systems and thorns. Thorns on plants deal damage on contact and in the digestive system, and can be detected by bibites, but bibites can also specialise to eat thorns better with less damage and absolutely no cost over their lifetime, but requiring more energy to be born without the extra energy being imparted upon them. The vascular system meanwhile should be for multi-segmented plants and would let them pass water and sugars (energy) through their bodies in exchange for a small bit of said resources, even if part of their body is in a dry area where it cannot get water by itself.
I think something you're missing in your simulation of food is a system to model nutrients. If energy was all that mattered we'd all be optimized to consume as much fat as possible, but we need nutrients and proteins to function properly. A way to implement this would be to have organism have a "nutrient hirerchy" that determines what genes they'll put their nutrients to support. More developed and specialized genes would require more nutrients to maintain. So instead it maxing each trait organisms would have to choose what is most beneficial given the limited resources. Also consider adding a sleep mechanism. Could be implemented by having energy efficiency decrease as the organism gets more "tired". So the longer they're awake the more expensive it becomes to function. To go along with this a night mode in which organisms will have a harder time "seeing" their surroundings
Nutrients remind me of BioMaker CA and Neuraquarium.
Also, speaking of dreams, I wonder if Norn-style dreaming could be applied to Bibites too
Marvelous! Keep it up! I'm incredibly impressed with the sophistication of your project, I can't think of anybody better qualified than a systems engineer.
I'm really hyped for how this project will develop! An idea for intermediate filler content you could try to put out between these big videos, I think it would be interesting to see what real organisms and behaviors have influenced what you have done thus far and/or what you'd like to do in the future. It might be a good way to educate your base and to get people thinking about new dynamics that could help the project.
This is really cool! I can’t wait to see how you add more senses like smell and hearing to the bibits, I think it would add a lot more depth to the different niches and behaviors.
I know these videos take a crazy amount of time to make, but you do a damn good job here! This is entertaining and educational and I patiently look forward on the future of this project and channel.
Thank you so much
New materials sound like a cool idea, perhaps creatures could evolve to grip rocks and throw them at other creatures, dealing massive damage. I think I've seen that happen before here but rocks could probably just do more damage than plants or meat.
Maybe damage would be relative to the material's hardness and weight, the surface of impact and the amount of energy absorbed on impact (if I throw a cube of plastic and it scratches your side, you wouldn't get as hurt as if you took it in the face or belly)
This is something I have been thinking for a while now, and now I’m waiting for the premiere to begin.
And this is the kind of content I watch:)
I had something to ask you to implement but I forgot what it was
This is amazing. For all of the ideas that were being thrown around about how to get predation, this is by far the simplest solution. This project is in good hands for sure.
One idea to maybe add: Rocks. - For digestion.
Like, chicken and some other birds will actually eat gravel to aid digestion, grinding up the stuff they eat, increasing efficiency, freeing up more nutrients overall.
And likewise, indigestible plant matter actually can also be helpful with, uh, gastric flow I guess? Some stuff tends to get stuck more easily, causing blockages. And the indigestible parts of plants may help that along.
So I think actually simulating waste disposal, complete with keeping track of how many nutrients are left in it and how tricky it is to break down, would be an excellent addition.
The rock idea would work best for animals which eat tougher food pellets. As those proposed in the vid. This would make so many different behaviours for survival i love it.
And having waste with nutrients left over would create a niche for things like a Dung-bibite
@@Hirosjimma also could play into plant evolution where the dung may fertilize the ground, giving rise to more nutrient-dependent plants
This game's production is one of the most engaging stories.
Hello there!
I am a biology student at BYU and I was wondering if corpses could be added as a meal sorce on the simulation.
Instead of creating chunks of meat when an animal die, you could make a corpse of the dead creature. This will open way more niches for onivores and opportunistic animals.
Also, as a corpse rots, the less energy it can provide, or at least, the harder it is to digest a rotten meat.
I love the simulation and hope it gets more attention on youtube!
I second this, it seems like the best way to make predation more advantageous than scavenging and would model real ecosystems
Commenting to help boost the algorithm.
These are good changes
7:17 i have a GREAT feature idea:
if a bibite tries to load more acid than there is room in the stomach, it will vomit. Bibites could even learn to vomit as a defence mechanism, if you add acid damage
Or they could feed other bibites
@@nathanwi1147 yes, yes! birds
@@MrRyanroberson1 Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds Birds
im amazed that you manage to not only create a very deep simulation of life, but you also explain it very well and with good editing, its the hole package
It's the result of a lot of work and practice 😅
If you look at my first videos the simulation was not there yet, and my editing skills were non-existent!
maybe an interesting idea: to help make scavenging a more viable niche, have bibites lay eggs when they have children instead of just spawning them, they could have even more nutrient density than meat but wouldn't be viable for herbivores to eat because of the need for specialised jaws to break the egg's shell.
They do actually lay eggs. I haven't used the simulator much so I don't know why the eggs are there (though sometimes there isn't enough energy in them so it dies), but the PDF linked in the Bibite neural network viewer doesn't mention looking for eggs or anything, so I guess it hasn't been implemented.
@@elliotlea5457 i know they lay eggs yet feldbiss's idea is more realistic due to eggs taking some time to hatch. this would actualy make the scavenger niche alot more viable and make it easier to survive as a ferocious carnivore.
I like seeing the Dani editing style on someone who's calmer. I think it's pretty inherently funny. Otherwise, also, epic content. Love to see sophisticated evolution simulations in friendly, understandable game-like environments. Keep up the good work!
Just looked up Dani, I admit there's a similitude 🤣
And thanks overall for the kind comments 😁
Dude, that digestion system is overpowered!
I do love the BIOME plan, too.
Edit:
Thinking about the koala example made me think of a thing you could do if your brain simulations become too complex, which is to "Tax" intelligence by making it more expensive to have a more complex brain.
a simple concept that i had considered long ago was simply making the digestion work as so: for each percentage point of plant digestion efficiency, it costs three units, and for each percentage point of meat digestion efficiency, it costs one unit; where the unit is adjusted to balance things reasonably (expect herbivores to hover around 50% efficiency)
Yes! Finally! I hope it is not short.
It is not necessary.
P.S.: The best update so far.
The actual content is ~15min long, then thanks, and then the patreon simulation
@@TheBibitesDigitalLife Great!
Aha! I'm glad you added digestion, it's such a huge part of the difference between herbivores and carnivors and I'm glad it played out well in the simulation
I've been simulating a single system since this update came out. A really small change that i think would work wonders is another "cheat" neuron/synapse. Just have two vision input neurons attached to the rotate or accelerate output neurons instead of just one. At the moment it's nearly impossible for predictors to successfully evolve a second sight neuron without screwing up their developed pathfinding. But if they should spawn "virgin" with two sight neurons i think it would SERIOUSLY help with hunting behavior.
6:10 Infinity food split for locomotion and nuclear reaction simulation. The deadpan delivery brought a smile to my face!
Just wondering will the new update be coming after the premiere or before the premiere. (just exited for the new visuals as i dont know what is what in my average simulations) Thanks
About the same time, will do my best to upload as soon as possible! Working on it right now 😁!
I really think that the most important part of any evolution simulator is ecosystems, especially varied and changing ecosystems. Right now, the most you have is areas of higher or lower concentrations of food, which doesn't impact a whole lot. Having different areas that affect the organisms promotes different adaptations to fill different niches, allowing for a much wider range of biodiversity. It could help to solve the issue of transitioning from herbivore to carnivore, by allowing the species to gradually evolve through different situations rather than hoping that the jump happens naturally and is sustainable.
Some variables that you can have for different environments could be the temperature, having more or less concentrations of different resources (you'd have to add in more resources to compete for before adding this one), or environmental factors like mud that slows down one movement type but promotes another, or hiding places for certain body types. I saw in another comment suggesting adding pools of water, which I think would be a great place to start. Water itself is a valuable resource that can be competed for and specialized around. Stemming from this, you could add mobility options, with a spectrum going from fins to legs, with fins being most useful for water, legs most useful for land, and somewhere in the middle for the shoreline (moving between water and land regularly). Inside the water, there could be one ecosystem with aquatic herbivore and carnivore niches, on the shoreline there could evolve to be a niche that dominates the shoreline as bibites go to drink water, and on land there could be a huge range of biodiversity and niches depending on an area's access to water (the plant matter could have a different spawning frequency or even quality depending on the proximity to water, and the bibites themselves would have to adapt around the availability of water for themselves depending on what area they live in). From there, you could implement environmental change, having the bodies of water shift and change size and shape, which would create an ever changing simulated world with a high biodiversity. And that's just one example stemming from a single idea, you could take the premise in any direction you wanted and run with it.
I should probably say that this is all just a theory of mine, and I have absolutely no credibility at all, not in biology or coding. This entire thing is basically a guess formed from what little I know and understand.
something i might reccomend is having tthe digestion also depend on how "chewed" the food is , with smaller pellets digesting with greater efficiency. you might need to adjust how the biting works so that its an actual "decision" on how to evolve and not just always pup everything.
not sure if its there already but a bibit that dies of starvation should drop less meat than a hunted bibit that has energy banked. and bibits that have a large amount of energy should be visually bigger to a predator.
added in 0.6
One of the best simulations on UA-cam I can’t wait to see what you do with this amazing project in the future!
Could you implement something like grass? Or something that predators can use as camouflage if evolved in this assassin like traits. This will help predators while herbivores can evolved to avoid grass and maybe evolved high enough sense that can detect the predators even while camouflage
This is just a suggestion that probably won't be implemented, I thought it will be cool
Would be great if combined with other required resources like water. Creating Bibites who prefer to wait in the water and go on the "ground" to eat, and the other way around with predators who are less visible in the water attacking from it. With any creatures who needs water becoming their prey. (Like crocodiles)
Grasses are one of the most advanced plant types ever. They do a weird asexual reproduction colonial thing in addition to having sex. That probably happened, but it is not representative of how the first carnivores appeared (they actually appeared before herbivores, because they appeared before plants, because being a plant is hard).
Or herbivores learn to hide too.
@@ripmadhatter4948 ye they can camouflage too
Good luck with that project!
Have you implemented reproductive drive? That seems pretty important.
You could allow a waste excretion system that produces waste from digested food. Bibites can eat this with a cost of their health over time (disease), however they can evolve to prevent this. The waste material would be extremely easy to chew and easy to digest but wouldn't be at all energy dense. Creatures that use waste as their primary source of food require the living of other species to eat and reproduce.
Eating waste also allows for pathogen evolution, also fascinating
Potentially changing meat after its been eaten it becomes a grey crescent to represent a striped corpse. A high jaw (bite) power variable should be needed to be able eat the corpse. This would mean you have to add 1 new variable but you could try to update the combat by allowing different builds (I want to see a giant herbivore with a massive tail). claw / bite / tail power/speed variables could be added to all the bibits, so you can workout the energy and the power of each attack.
Loving the new update! Good stuff!
I have some suggestions for future updates. It would be cool to see the number of eggs a bibite has layed, and/or the number of offspring that is still alive.
The main thing I would love to see is an option to turn off the visual simulation. That way we can leave our bibites running while afk or while the game is in the background.
I think it could really speed up the simulation and speed up the fun!
Hype time, can't wait😁
Leaving a comment to hopefully help boost this in the algorithm. I've always been fascinated with the idea of extremely in depth evolution simulations and you're really going the extra mile to put in all the details I feel is lacking in any other simulation.
Thanks! You're a hero 💪💪💪
The editing and green screen work just keeps getting better and better with every video, keep it up! I love learning new stuff from videos like this
All the things you're adding to the bibites before plant evolution make perfect sense and I think the modules system would be the same level of if not more interesting than plant evolution, but don't overwork yourself, we can wait!
The bibites is way too underrated, it's easily the best life simulation so far!
Really love this video! I think with your carrying capacity problem you're running into an issue where predators can completely depopulate a species before dying off themselves. If you can buff the last survivors of a species by making them more difficult to find or kill, you'll see more die off from predators that over-exploit prey. So you can get into regular cycles of increasing and decreasing predator prey populations like real life.
Nectar or fruit seem like a pretty logical material, since it makes it easier to eat plant matter but aids the plant in some way. Rocks being used to help with digestion, a la chicken gizzards, would also be a good addition. And while you're at it, developing poisons as both a factor a material can have and a skill that can be evolved feels like another fun addition spinning off of this inclusion.
I feel like a big thing that could improve practically everything is working on colonies of bibites or something similar. Having multicellular creatures allows more variety in designs, as well as more variety in general. Maybe this could be done by more pheremones, or ways for bibites to transfer food and energy to others
He's already implemented herding behaviour. Well, the ability to evolve herding.
@@Xeridanusmaybe go 1 step further and make them physically attach? Like EvoLife and Dylan Cope
Making plants evolve some sort of self defense mechanisms would hinder the herbivores and more accurately mimic nature. I would have no clue how to code that, but that's why we watch Bibites!!!
yooo im so exited
this simulation is so cool and complex, I don’t know how you managed to code it all without going insane. The digestion thing is really really neat, expecially the visualization.
Not first but 4th :) IM SO EXCITED
yo i friggin love this experiment. props to all your hard work bro, the world is a better place thanks to you.
🦆
🦆
🦆
🦆 q u a c k
🦆
🦆
Absolutely love it ! As a fellow doctor who has high levels of enthusiasm in evolution and microbiology i absolutely simp for your con bro ! Eished more ppl were to actually care more for such enlightenments than to low tier gaming streams and such
waiting for the memes
I'm just discovering this channel and project, and I... I don't have enough words
What you are doing is insane
I had this idea on my mind as a kid, as a fantasy, to be able (humans in general, not me particularly) someday to program an entire world with several species, complex interactions and all... and what you are doing might very well be one of the first projects like this ever!! I cannot wait to see how it keeps on improving, and to try to become involved in this community
Extremely underrated channel. I love these kind of projects, there are not very many out there. Greetings from Argentina.
Thanks so much 🤩
i love how he does all his own stock footage
People tell me I have a stockfootage face
I burst out laughing at the Bibite's immediate abuse of splitting food infinitely as a form of transport, that's just perfect-- Not only infinite food, but energy-free travel LOL
This is a great project, many thanks to all contributors and mr. idea father here. Hopefully it will reach a point of essential completion so I can basque in its ever growing glory for many days to come.
Love where you took this idea. You did a lot of really cool things with it that I never would have considered.
Thanks so much for the nice comment 🥰
I've wanted to do something like this for a long time. One thing to keep in mind is things like digestion are abstractions of the emergent properties of chemistry and physics. The abstraction makes it easier to model but isolates it. So with the meat vs plant model it ignores that their properties come from things like, amount of fats, carbohydrates, proteins, etc, and that these things themselves have different properties. Organisms therefore can be more plant like or more animal like, assuming that 'plant' and 'animal' even continue to have useful meanings.
Anyway for a practical suggestion, have you tried implementing a smell based system? Something like a cellular automata which allows an organism to detect concentrations of different materials or different species.
Absolutely love where this channel is heading. A system where bibites can evolve new body parts would be amazing to see, you could even go so far as to have each new bodypart / functional unit of body having its own material stats that also effect how it functions (obvi in the distant future lol). Very exiting!!
Take as long as you need, it is worth the wait. And i wish you luck on this incredible endeavor! :)
You could also make it so bibites able to move in different ways, like flying could use lots of, very fast, but they would only need to use it for short bursts because they can glide. Or slithering like a snake which could give it certain advantages. Making them quadrupeds or bipeds and making it affect their lifestyle and movement patterns. Mating and more complex relationships would also be amazing to see. Making shelters like nests or dens and giving them the abilities to affect the things around them, other than eating them. I think this is super cool and it’s so awesome that you’re doing this.
This project is legendary, You are legendary and anyone who helps You is legendary. You basically made one of the most life-like artificial life simulation and gave it for free. This is going down in history
Thank you so much 🥰
Trying my best
I have been counting the days until your next video, i have saved this one for a special time, and am so glad to be a follower of yours.
Thanks so much, next one shouldn't be too far 😁
I always wanted a game where the life in the game was apart of a ever evolving system. This project seems like its paving the ground work for this sort of thing in the future
i think another good thing to apply to the program is adding another resource (lets just call it 'materials' or 'nutrients' for now) and /waste/.
this would force the bibits to also manage how many resources they have in their bodies at once, perhaps even splitting the nutrient resources into proteins and minerals or so, and having too much of one or both can cause damage to the bibit, while evolving methods to accommodate having more or less tolerance to either depending on the environment.
this would ALSO encourage the diversification of the scavengers, as these resources (plus some free floating energy) would be left behind if something dies and rots as well
Came to the comments to shout this is simply awesome! Look forward to more of this project.
Thanks!
This is insanely cool. I don't know what else to say, I just want the algorithm's attention
You should also consider that smell is as big of a factor for animals finding the correct plants to eat. Smell is an “aura” that is sensed in a certain area depending on wind and smell strength. But I think it should be considered.
I'm glad that I've found your channel. I absolutely love playing with the simulation.
Thanks, I am glad you found it too 😁
If the bibitts takes place underwater, you could make salinity a factor when you get around to environments.
i just had a thought that life discovered itself, named itself, was fascinated by itself, and decided to try and recreate itself.
I love all this changes ! I just have one remark : in your current simulation, carnivores can't eat plants at all or they will lose energy, while even full herbivores can eat a little bit of meat sometimes. It looks like the opposite of real life, in which many "carnivores" are in fact omnivores that eat slightly more meat than plants (i think all mammalians carnivores work like this for example), so it may be a bit irrealistic in my opinion
great work anyway !!
Aside from the plant they get from eating the stomachs of their prey, ferrets aren't able to eat plants at all. They're incapable of digesting them and then get clogged up and die. They're the only true carnivorous mammal I can think of. I mean it's not like plants from the stomach of something that can handle it matter much.. it's already partially digested for them.
I think making the creature's color a deciding factor in camouflage would do great as it would allow for predators and prey to hunt and be hunted more dynamicly.
I have a fun idea:
Collab with someone else who has an evolution simulator, and add both your creatures and theirs to a single simulation. You can compete to see which is better at adapting.
That sounds fun, but it's VERY impractical. Each simulation has an entire code architecture that can't be translated between one another. Even if the programming language is the same, everything else is different: the functions, the architectures, the genes, the variables and so on
@@pedroheck3667 true but 'd like to see which simulation is more realistic. like what creatures adapt better. is the te one person or this legend. basically it will be another Alife project with predators. than they would set the simulation to run to see which Alife simulation has better CO-EVOLUTION.
Plant evolution is coooool B) . Also, I've said it before, but these videos are an incredibly useful educational tool for me!
Honestly, you bringing up adding rocks as a material makes me think of, like, a species that evolved to survive off of *literal stone*
Every update makes me more and more excited for the final thing
I don't know if you still read this comment section, but making blood will allow for a parasite niche, which will then allow for a symbiosis niche to occur of smaller bibites eating the parasites off of larger bibites, assuming bibites can tell the difference between prey and allies
one of the coolest simulation projects ive seen in a while. Keep it up!
As you might know, a desirable trait for the system to have which will lead to many emergent properties to share with our own world is a measure of conservation; if energy were conserved within the system, with low entropy energy coming in and high entropy energy leaving in equal measure, you will find that many other desirable priorly arbitrary properties emerge.
Found your videos, and immediately downloaded your simmy gammy thing, six hours in I'm loving it keep it up
i love your videos! these kind of simulations made me fall in love with coding. its like finding your favorite music album all over again. Please keep going!
Absolutely will 💪