What will happen after 100h of evolution? Study of natural selection in an ecosystem simulator

In this video, I decided to showcase and demonstrate what kind of result you can expect from the simulation. So, follow me on this fascinating journey, as I let the simulation run for 100 hours, regularly pausing it so we can see how things develop.
We're also going to get some help from @alejolab in order to fix a certain problem 🤔
part 2: • The result of 100...
=================================
The Bibites is an Artificial Life (Living AI!) simulation where I recreate some biological processes and let the lifeforms live, eat, reproduce, and mutate, leading to active evolution.
They can evolve their body through a genetic algorithm and their behavior through a custom neural network algorithm.
=================================
Follow me on Twitter : TBibites
Support the project on Patreon : www.patreon.com/leocaussan
Join the community on Reddit : www.reddit.com/r/TheBibites/
Download and play the game : leocaussan.itch.io/the-bibites
Subscribe to the channel: / @thebibitesdigita...
=================================
Credits:
Footages from Pexels.com and Videezy.com
Procedural Sprites by Brax - Twitter@Braxiations
All songs by Fireballme - glowingamber.bandcamp.com
#evolution #AI #devlog #simulation #ecosystem

[Tactical Ant]

I think each bibite should be assigned with a random letter and number when they are born from nothing and each of their kid would have the same number. This would make it easier to track migrations and extinction events, as it would act like an ID to identify species with a common ancestor. In your own simulation, it would have been really useful to make sure that the inutilus truly migrated to other islands.

[Timothy McLean]

@The Bibites: Digital Life Ah, but blockchain is decentralized. Which is universally beneficial for some reason _and_ unique to blockchain technology.

[Luke First]

@The Bibites: Digital Life this is so based. fuck blockchain

[The Bibites: Digital Life]

@Gustavo Lopes I indeed am against crypto, but that's not what this is about. I know what a Blockchain is: A technology that builds a ledger and uses hashing (in some cases) to validate ledger blocks that hold information and a reference to the previous blocks. This technology is dumb and has absolutely no real-world uses where a regular database, list, or table, doesn't do a better more efficient job. I'm tired of people thinking a Blockchain is synonymous with storing data. We've been storing data for a long time in computer science, and the solutions we already have are just more efficient data structures. Especially for the case of species, genetic information, and such, can you put forth a single argument that would make a Blockchain a better approach than any other regular data structure?

[Gustavo Lopes]

@The Bibites: Digital Life man... You may be against crypto, and that's okay, but that's not what I'm talking about. I mean the technology that powers the thing (a hash in a block, that contains a hash in the block and etc.). If you search for it, you'll see how it could help you track the genes. It has nothing to do with any finantial thing whatsoever.

[The Bibites: Digital Life]

@Gustavo Lopes Blockchain is a dumb technology

[Lord Ecramox]

There are many evolution projects on KZbin, but the Bibites are by far my favorite. Their pixel art has so much charm and their complexity is unmatched elsewhere. Nice job!

[Boddle]

omg its lord ecramox the almigthy!!111

[Lord Ecramox]

@Matthew Anderson I am pretty biased I suppose.

[Matthew Anderson]

How can you not like procedural pixel art? I mean your PFP is literally rainworld

[Nen Master5]

@The Bibites: Digital Life ...HELP!! For this Latin-Taxomony-Naming-Scheme, is there an Online-App i can use?

[OGumbal]

​@The Bibites: Digital Life hey dude can u release this on steam? im scared of downloadingmit from the site😅

[Josh Landers]

Really hoping for some charts and graphs and maybe even a generation tree or family tree for them. Looking forward to getting sucked back into the simulation too. Watching these events make me want to play it more.

[IRS HQ]

Hot take of the century but i think that the pink eye gene was wiped out by initulis that migrated over and the reason they mutated to the physical likeness the ancestral natives into a pink eyed larger variant that differed from the other islands was a higher infant mortality of the island (from the racism laser)

[Twist_YT]

I wish i can play it but im on mobile so i cant But its its still nice to watch!

[Stick world Animated]

Yes we need that I have a graphing system in C++ if he wants to use it! Or does he use java script idk!

[Rizky Muhammad Arief]

@Andreas M yea, that i just think we can add new id_bibits if there has 10 bibits in same or just add more and more id_bibits for only little changed, um i don't know what i want to tell Xd

[Andreas M]

@Rizky Muhammad Arief i think gene shouldn't be used to make the tree, in Real life it would be the way to do so but in our situation we have id

[Yokoto12343]

Maybe you can encourage predation by lowering plant calories even more and making meat and injured bibites "bleed" a pherhormone trail that allows carnivores to track them

[ToxicWaste]

Good idea

[skyworm]

Real life predators metabolically are constantly in a state of fasting. That is, they are evolved to function fine without constant eating and digesting. This is possible because animals are extremely nutritious compared to plants (especially grass and leaves) and the cost of digestion is low. Humans are actually like this too though our far origin seems to be more herbivorous and we have kept some limited ability to digest cultivated fleshy parts of fruits and seeds (especially with cooking and other preparation). Most other apes have a digestive system built for plants, leaves, though they can digest animals too. Generally digesting animals is easy, it's in an accessible form, but effectively digesting plants requires a more elaborate system for breaking it down and converting it properly as well as breaking down toxins. So the evolutionary cost of eating plants is much higher, as well as more limited (one system can't digest every plant effectively enough for survival to be viable), and requires eating a lot more, constantly eating really. Huge amount of time and energy dedicated to crunching the material down then the digestion which takes a lot more energy and time than digesting animals.

[Death_Bull_7_]

Good idea

[Kitty Giatanidon]

Yes, symbiosis. Commensalism, mutualism, and parasitism. Commensalism is when it affects one in a good way Mutualism is when it affects both good And parasitism is when it affects one poorly and one well

[Jappards]

@Kitty Giatanidon What if they could latch and hold onto larger Bibites to drain their energy? Think Lamprey fish.

[Corwin Keylon]

You might have to separate the diet gene. I talked about it a bit on the last video. The fact of the matter is that it takes way too much time to "un evolve" plant digestion and transform it into meat digestion for carnivores to ever evolve. Instead of making them *choose* between plants or meats, just have an individual gene for each. This would make it possible for a bibite to have maximum efficiency in both plants AND meat, however you can balance this by making such a metabolism extremely costly in energy. I don't know exactly what else can be done to promote evolution to carnivore, but at the very least you shouldn't punish them for trying by having them lose their main food source entirely. I don't know if even this will help fix the predation problem, it's just a guess.

[StarScream]

For the most realism, I think the carnivore gene should have a random chance of developing in an environment with less food than the amount of creatures in the area, similar to how carnivores may have developed in reality. That way, when the creature gets too successful and starts destroying their environment, natural selection can start preferring omnivorous or eventually carnivorous creatures.

[Parker Nelson]

@Björn Jürgens very smart, I love your last idea I think that would be very cool to see

[Giovane Dias]

@Nevir202 humm... maybe lacing a digestion axis as starch/fiber proficiency vs mineral/acid proficiency? as usually those macro nutrients clive different adaptations to enable/increase digestion of those kinds of foods. herbivores usually can't digest larger bones where carnivores can't digest well long starch chains and fibers (no source here, srry), so a continuum can be built where one lineage can migrate from a herbivore towards a carnivore digestion given the right inputs/luck

[Jappards]

@Björn Jürgens Or different kinds of plants have different defenses? Some plants could have a "shell" around them, others may be spiky and others may eject a pheromone that makes the herbivore more likely to be seen. Also useful for plant evolution. Different mouths could be the options for evolution.

[Jackson Davies]

Omnivores for the win.

[Phantom Fox]

what i would love is a family tree to be added into the game so you can see how a species evolved it would make it some much more interesting in my opinion also great video i love what your doing

[Rytis Noreika]

family tree would be amazing

[Mahapushpa Cyavana]

KZbin keeps on putting my comment in the wrong thread. Sorry.

[Master TMO]

Probably the simplest way to provide family tree information is for a single data field to be added with the ID (I presume there is one) of the parent. It would take some stitching together to create a full family tree, but that would be the most space-effective solution I can think of.

[Lurky B0i]

Maybe how a family tree would work is every so often, the simulation will document the most recurring species and monitor them and then after another period of time the simulation will document them but have a record of their past. The simulation would keep track of the children and when change happens to a majority of them a new species is created.

[Fireballme]

I forgot how much I love The Bibites. Maybe someday it could be used as an educational tool or something!

[ToxicWaste]

​@The Bibites: Digital Life ok

[A Stick]

@The Bibites: Digital Life Now thats amazing

[The Bibites: Digital Life]

I know a few teachers have sent me messages that they were using the program in their class!

[Casey Martin]

This is where there is actually a lot of value in having genetic material that is “useless”. The useless parts of the genome are great for determining how (and when) they diverged. Because that DNA doesn’t have any function, it’s very unlikely to be similar as a result of convergent evolution, and much more likely a result of common ancestry.

[nacoran]

Maybe, short of full saves, you could have the program randomly save the genes of a small number of creatures as the simulation progresses. It wouldn't be a full DNA database, but it could have a few samples... the little fossil evidence to help you figure out how things evolved.

[chpsilva]

Maybe compare with the current samples and if it differs enough, create a new entry (I also suggest adding a timestamp for each entry, to help build a chronological graphic)

[waylonmorristexas]

I like that idea

[Vivi]

I absolutely love this series. The only issue with it is the wait between uploads is so long (which is understandable), I have to spend a minute attempting to remember what the last episode taught us! A lot of it is common sense though, so that isn't too difficult :)

[Jag mannen är BRAND]

@The Bibites: Digital Life do you ever think you would do like a few smaller updates in between the big videos? Idk about anyone else but I’d even be interested in the coding that goes into it. Maybe like a less edited video that doesn’t take time to make?

[Gigantopithecus]

@The Bibites: Digital Life I wanna help you

[The Bibites: Digital Life]

Yeah sorry 😭 I'm alone working on this and it takes me a long time to do

[Zerglord Ghi]

The Bibite's ability to swallow food whole made me realize that the implementation of parasitism would be interesting. If a small digestion-resistant bibite gets eaten by a larger bibite, it could eat food inside the bibite, reproduce, then spread after the bibite dies. Just something I thought could be cool to add.

[Nugcon]

yes

[Light_Gladiator]

Yeah, imagine if we had species die off as some of them become super violent or just stop doing anything but spread the tumor

[DaoofFibonnaciCats]

I’m really excited for this since it has been a while since a long simulation(which was only 1 hour and was 2 years ago). Also P.S people on the thebibites reddit have done significantly longer simulations.

[Beans]

The sheer dedication and complexity of this project definitely makes it one of the best evolution sims you can find for free, not to mention the style of the editing and pacing of the video is very easily digestible. Overall, well done, can't wait to see this project get bigger and more popular!

[The Bibites: Digital Life]

Thanks a lot 🥰

[partner555]

I think it’s great how you managed to recreate the struggles of a palaeontologist by complete accident

[Ewan lee]

A helpful feature would probably be an auto saver. Set a 1, 5 or custom hour timer and don’t have to worry about missing stuff. Rozoculi hostis will definitely conquer the island. Love the project. Keep it up 👍

[Alacer]

Man, I don't know about other people, but this content is so great that I could wait an age for more. I'll always be around to catch more of this amazing series. Thank you.

[Alacer]

@The Bibites: Digital Life Good! You deserve it for giving me and countless others a wonderful thing to watch, be entertained by, and engage with. :D

[The Bibites: Digital Life]

Wow, thanks so much for the kind words! You made my day 💪💪💪

[Cretarius]

While creating actual automatic taxonomic classifications would be downright impossible, have you considered creating a UI which allows you to click on a bibite and view all of its ancestors? Since bibites reproduce asexually this would just be a single sliding scale through time. Issues with storage may arise, which is why I would also implement a slider that lets the user decide how many generations back they want it to be saved.

[Giovane Dias]

if instead of tracing all individuals forever, doing a little roll back and resuming full logs that lost their space in the current ecosystem every x iterations to just a snapshot, and saving the ancestry from the surviving lineages for more detailed investigation.

[sajeu cette fois tu ne vas pas me voler le compte]

"Issues with storage may arise" what happens generally is that bibites spawn and die again and again then one of them laids an egg and then the whole species goes extinct and then an other one laids an egg and a new rule specie appears and take over the whole map and it ,repeats so no real storage issue but a lot of "write/delete" which isn't a problem when well made

[Orthodox American]

You are the first evolution project people I've seen actually release it to the public for people to play with it themselves Thats pretty cool of you

[Happy Poop]

@The Bibites: Digital Life i hope you implement clade diagram to ease ancestry :) love your project very much

[The Bibites: Digital Life]

I find the project too cool not to share 😳 Initially the actual simulation was only available to my Patreon, but I figured that everyone deserved to play around with the full version

[JDoawp]

An autosave feature might be a good idea. Also as some others have mentioned being able to see any individuals family tree (or a shortened version of it where it only shows creatures of any significant changes) would be real cool.

[Andrew900460]

I remember seeing your old version of the simulation a long time ago. Glad to see you're still developing it! It looks really great too! There are so many different creature sprites for a lot of variety. And a lot of different genes that affect behavior. I also have a suggestion, I feel like from a scientific/experimental perspective. It's important not to tamper with a simulation after it has started. So I would suggest making it so eye colors mutate more slowly. Or make it where eye color is attached to certain genes. And those attachments are also randomly selected, but they don't mutate often. So the eye colors only noticeably change when a creature makes a large stride in mutation.

[Jackson Davies]

I thought it was a bit mean destroying the red eyes, they might have been an important branch of the genus that has been eliminated simply because the eye colour wasn't preferred.

[Jacob]

It'd be neat if you could have some kind of barrier or object that blocks out bibites of a certain size. This could mimick areas like thickets where smaller animals are dominant, allowing for one ecosystem to have multiple niches that interact with eachother. The bibites could learn to flee to these areas in case of predators, or to spawn their young there because there's less competition.

[Bob Jhon]

You should make a program that takes makes a life chart showing every mutation and who that mutation came from. Both in terms of evolution tree and where it is and where their ancestors primarily lived. This way you can keep it running without missing anything.

[Ziano1192]

You know you've suceeded in creating a good simulation when even things that you didn't intend to happen (like bibites crossing the void and migrating) somewhat follow real world patterns Amazing job!!

[Fireballme]

Love the naming ideas! It's like I'm watching a real documentary about an alien planet.

[manfred97232]

Great video ! Maybe keeping some basic info of the family tree (coord of birth, parent genes (or a name based on a region? done manually or randomly through evolution)) for each bibite could help you figure out which species are responsible for big changes and wipeouts in an area. You can keep maybe 2/3/or more ancestors worth of info to avoid performance issues and still have a kind of archaeologic path to follow. Combined with the random creatures saving mentioned by @nacoran, it could give you 100 hours worth of work :D to study the history of an 100h simulation and narrate its story to us, story that only you experienced.

[S1m0ne]

You should add a field to your bibites that documents their ancestry. This would be valuable for you to track changes of lineages, which seems to be a thing you like to do. Either that or export each generation into a separate database containing its lineage and genes so you can restore any given generation at-will.

[Herodotus Von 8428]

One suggestion I have to make ancestry identification easier would be to pick like 20 or so “base colors” and then make it so that the color can only be used by one initial spawn at a time. As long as early individuals die at a rate faster than new initial spawns are added, this should work. Then, make it so that color mutates incredibly slowly. Like perhaps the value of the color gene is divided by a factor of a fifty so that it requires 50 successive mutations for the color change to be visible. Otherwise the best option would be to store all the ancestry data and make it possible to search individuals in the ancestry tree and determine relationships that way (though you probably chose not to do this because it is very data intensive)

[Person]

I don't think it would be too hard. Have each initial spawn have a randomly generated, 5 long string (like "eny8x"). Every single one of it's children gets (the same) random letter added to its tag. This gives a way to follow lineages and shouldn't take too much memory

[Flunkee]

14:45 I don't believe that's a bug! I think the bibites evolved a habit that displays this emergent organization.

[Flunkee]

​@Jappards You are correct, I didn't think about it like that. Perhaps I got a bit excited.

[Jappards]

I think that is emergent, but not how you think. There are more pellets than the creatures can digest, so they end up clumping things together randomly. Think "Sensorless Robots"

[JustMondo]

@Serih Linte thanks

[Serih Linte]

@JustMondo Size of the simulation: 2000+ Pellet spawn number power: 1 Pellet spawn size factor: 0,3 (Or 0,2 for small islands and 0,6 for big islands, but sometimes they may intersect) Pellet growth concentrarion: 2,25 +/- (this one u can choose by urself) I learned it from a post in the subreddit, check it there if u want more info

[JustMondo]

@Serih Linte what settings did you use to make the islands? I cant seem to make them like he did in the video🤔

[Joseph ballin]

I love the bibites art style and the complexity, I’ve never seen anything like this, it has so much charm to it!

[Kermalis]

Honestly this is becoming one of my favorite channels, I really love what you're doing and how you are presenting it, and can't wait for more

[smaug131]

While it is fun to see the detective work done to fiure out where the new species come from, I think this shows that your simulation would really profit from having trackers implemented: some sort of family tree of the species, and the ability to draw a graph of the population over time of a branch of the tree, of or the amount of bibites that fit in a parameter set (two imperfect ways to define a species). This way you could see the development of a species much more closely!

[Erin Karp]

Oh my gosh, I love this analysis of the genes and brains of the bibites to try to figure out ancestry. Fantastic job!

[Riz Khan]

You should definitely make more videos like this during dry spells for content, I can't imagine this taking more work than an update. Also it is a good way to show the cumulative features from new updates.

[Aeternaliter]

This is pretty cool, you should make some kind of automated species timeline generation recording system and the option to observe and possibly even copy different bibites from this automatically-generated timeline to see their characteristics and possibly place them in the world, this timeline would greatly help with finding the complete timeline for a run

[Parmesan Zero]

Suggestion: Trace mutations with numerical/hex codes. Basically creating DNA. That way, if two creatures create the same behavior with convergent evolution, you can do what biologists do and analyze the genome to see where they actually evolved from. Regarding the useless brain connection - there’s no pressure against such things. Do more connections cost more energy?

[GHOST2RM]

Love this series man, I think two things you could do to improve the quality of life in the sim would be some form of a family tree tied to each bibit, and an auto saver with a controlled timer. It would be so cool to have a base world you saved previously that you could then tweak with new params to see how the bibits there would survive. Also, since predation seems to take forever to evolve, maybe add "bleeding" to the bibits when they're attacked, that way predator bibits could tracks the trails like pheromones. This could also be it's own food source, which might lead some bibits to take on a "hit and run" approach to getting food from prey :)

[Vitor Lopes]

Sugestion in Portuguese: Implemente um número de identificação para cada indivíduo. Um número duplo. Um número que representa a ordem de nascimento do próprio indivíduo, e outro herdadado do seu progenitor. Assim os os indivíduos originais de uma simulação podem ser representados assim: (0, 1) (0,2) (0,3) por exemplo, na hipótese de cada um ter duas crias, seus números de identificação seriam por exemplo: (1,4) (1,7) (2,5) (2,8) (3,6) (3,9), assim fica fácil de perceber, (até mesmo a olho nú e mais facilmente para o caso de muitos indivíduos com uma função simples), qual é o padrão de descendência entre os indivíduos da simulação, mantendo uma lista dos indivíduos que nasceram. :) Isso deve resolve e facilitar a identificação. E pode funcionar como uma prova para vários aspectos à serem abordados nos próximos vídeos. :)

[Aaron Theobald]

I think you should add an optional auto save/snapshot feature. Make it an option with a time parameter, that way you could just set it to five hours and then not have to worry about missing key evolutionary events, like the pinkeye migration. Loving this!!

[TrueKristian]

If you want to make it easier for you, you could implement an autosave feature (switchable in options maybe) to save at least every hour If you want to make it _even more_ easy for you, you could also implement an internal naming scheme based on the irl naming scheme you used here which automatically generates a first "full name" on creation and changes the species name if too many internal values don't match up anymore. Could even consider natural evolution so it doesn't create a new species every few minutes Just throwing that in as some ideas :)

[Lion's Garden]

Love your format of videos! Really enjoy the family tree feature and it was very clever to create different habitable locations!

[Diana Puna]

I'm just ordinary stupid person with no knowledge of programing, but this is amazing. Saw few videos on this topic, but no one had such comedian soul like you, love the sketches and all the work u put in even without the programing itself.

[Anderson Stridder]

Your project has an enormous capacity to succeed, you and your team are doing an awesome work! :D I'm kinda curious about one little thing, it's about how the omnivorous creature system would work

[jelleludolf]

Keep doing this man, I followed you for a long time - this kind of stuff is going to be fundamental to our future

[Starrsystem]

this is by far the best evolution simulation I've seen!! I'm absolutely in love with the graphics, the way the colors indicate family more than traits is cool! Their shape and size, too.. it's so great! I'd wait forever for more of this!

[Keith Goddard]

This is freaking cool! I'd never heard of an evolution simulator before, and I'm not entirely sure why KZbin decided to suggest this to me.... but damn, I'm glad it did! Such an interesting project to work on :)

[Grant Rogers]

I’m not sure if you remember me but a few years ago I saw one of your videos about this project, I left comment about how interested in the game I was. I just got the game this morning and I have to say I’m very impressed, I will say the saving and loading worlds could be a little better. Also having a chart/family tree for each species would really help those who want to see how they changed overtime. All said and done I hope more people discover the game and give you the support you deserve, allowing you to make more incredible stuff.

[Vander Karl]

While it's very interesting to see how you can infer the development of species, it seems like a good idea to add an automatic species tracking system like so many other evolution sims have.

[Erik Failing]

This is such a cool simulation man! It already has so much depth its crazy! I'm really excited to see where you take it :)

[pgmp]

I love to see how the AI software EVOLVES too. This is starting to get very interesting.

[GamePhysics]

This is really cool! Would be cool to make a 3D environment version of this, just even more variables. I guess the goal would be to simulate billions of generation in a very complex, everchanging environment to see if you could evolve intelliget life. This is already really impressive work!

[AidanThird]

I have quite a few ideas but firstly: this is incredible. I discovered you recently and every episode has left me astonished. I can't wait to see what else you add! Here are my ideas now: (i should preface that i havent yet played the game myself, so i have no idea what im talking about most likely) Firstly, **camouflage.** This could be a game changer. (wait is that a pun NOO) Camouflage could allow for some species to escape from predation, or perhaps sneak up if the predator was camouflaged. However, it should definitely cost energy to be camouflaged, or have some other downside, such as being bumped into or something. Instead (or maybe along with) camouflage, **patterns.** The aesthetic that the species have could be used in the brain structures. Kind of like the flashy patterns that some poisonous species' have. Maybe it's what kind of wing or arm or whatever is there, or maybe how red a species is. Or, some kind of stripe, dots, or even horns. Maybe even some harmless species will adapt these patterns! My next idea works alongside camouflage. Some kind of **sleep system.** When you transitioned from a white screen to a black screen I thought about a day/night cycle, and sleep works perfectly with that. It could either be to gain energy or prevent the loss of energy or some other weird effect. This next idea is about **the herds.** A few episodes back (so probably years at this point) you created herd behavior. Now that's all fine and dandy for the herbivores, but what about the carnivores? Could predators hunt in packs? Or, maybe like bees and wasps, certain species would be able to defend themselves by using herd behavior (I think you mentioned this before, but i cannot remember) Here is something you definitely mentioned. **Evolving plantlife.** I don't know if you're working on it right now or if you will in three years, but here are some ideas: Thorns. Some kind of gene that tries to prevent itself from being eaten. Or maybe some kind of pheromone. If a pheromone that is released by a plant to signify "hey, dont eat me" is the same pheromone that can be released by other species, then some kind of very interesting deception tactic will develop. You mentioned a long time ago something about **waste.** Now, I don't know if you're still doing that or if it's in the game already, but it will be a very interesting dynamic. Perhaps some species are only able to eat the waste, and that is why plants can form. Or maybe, it has viruses in the waste. Or even parasites! And here's a big one. Other than Bibites, what species' are there roaming around? What adaptations do they start with, or are able to adapt at all? I'm proposing, along with biomes and stuff, certain **species specific adaptations.** Kind of like how certain species can only consume certain foods. So, how would birds work in this simulation? They can pick up prey and kill them in a variety of ways, or perhaps dive down to get plants. But they need a spot to lay eggs, they can't just do it anywhere. It takes both energy AND time. Or maybe worms that can be either parasitic or decomposers. Also, to add to the biomes. Some kind of dark area, or cave or something that goes along with both the sleep system and vision. I know someone else also mentioned water patches. Sadly, I do not know who. But I can say that is a great idea. I don't know if you're gonna see this, or add any of these features, but I hope you do. I'm going to post this comment on Reddit also, just to try and improve the chances that possibly you will see it. I just hope you add at least a few of what is mentioned here. edit: i dont know how bold works lmao

[Tom Davies]

This is such a cool idea! Just grabbed it there and I am fascinated! I have 0 idea how you got the 3 small islands though, seems really cool and I wanna try it but I can't get it to work! Can I ask what parameters you used to get them so nicely separated, and just a few of them?

[Missing Name]

Really liked the video. I agree a few generations of family tree would be cool (like show after selecting one). You can also store cordinates of where it was born and that can help with mapping where it (and its ancestors) comes from. You might consider an autosave function, like every hour or 30 minutes (or configurable). I'm not sure how I'd consider things a different species (not a biologist myself), but you can try autogenning names (or taking your input for those two fields, i liked the story you told) and it auto updating the smaller one when its sufficient different. like a percent difference, maybe 3%? then you can go back in and set it to your 'offical name'. This would also help with tracking.

[ArEyeses]

You should add an autosave timer, maybe something like every hour Also, are you able to add save triggers, that save when something enters them? Then you could put these around your islands, and see where crossings happen

[DaOneTrueCupcake]

A useful tool for this kind of thing would be fossils. They would easily allow you to see what had happened between intervals, but may also be detrimental to preformance.

[pendragon]

Well this is a computer simulation. He could just take a snapshot of the whole simulation at regular intervals

[ThighErda]

maybe make fossils be rare? like only one in ten drop a fossil. that may help performance

[Reno Fox]

This is amazing! Most evolution simulators have a set amount of neurons, but this system really lets them grow in complexity by themselves. And the digestion system is just brilliant in all the new options it provides!

[Ordinary Knife]

as a separate simulation, you should have 3 islands: 2 of them alternate between producing a lot of food and producing only a little, and while one produces a lot, the other produces a little, like they're seasonal on opposite hemispheres. the third island stays fairly consistent but doesn't produce that much food, like, a bit less than the average of each of the other 2 islands.

[Low IQ Watchdog]

Really looking forward to the plant update. I think it's really going to be a game changer, especially if you put some more environment stuff.

[Т1000 Youtube]

It's good to see the improvements in graphics, as atmosphere is a very important part of these slow-paced spectator type games. Simple mainstream videos like this are also the right direction for attracting more people

[StingrayNinetySeven]

I got inspired from your videos and made my own version (kind of) in scratch and I found a bug where the "creatures" would randomly accidentally evolve to have negative speed, so they would get stuck against the walls and actually gain energy from not moving so I actually left it because it reminds me of a clam or muscle which stay still and just get energy from not moving

[Aaron Speedy]

You should add a feature that saves the simulation after a certain amount of time periodically

[Adam Hawkins]

There is an auto save feature already. But by default it saves every 10 mins and erases older saves when you get to 25 saves. So if you want data from more than 250 minutes in the past, you need to manually save at some point or change the auto save settings. Also, I wonder if he intentionally didn't save for an extended period of time as a way to demonstrate how the real world has gaps in information that makes analysis difficult for scientists.

[Keliuszel]

This would be cool

[Gvoya]

Great video as always, it's so interesting to see how strategies are formed over time. I love the research you did on the topic it actually makes the video so educational(and the naming scheme for the bibits was a top notch). Keep up the good work. 👍

[Vito Micallef]

As absolutely incredible this is as software, your editing and style of presentation in this video is fantastic my dude. Wonderful job

[Andrew Alekseyev]

I have downloaded the Bibites and am thoroughly enjoying it. I just want to know how you make 3 individual islands like you did in the video. I'm messing around in the settings but I can't find anything that might control the number of plant spawners or the distance between them. Anyway great video!

[Andrew Alekseyev]

@JustMondo I don't know the specifics but I turned simulation area to 4000. "Pellet spawn number power" to 1. "Pellet spawn size factor" to 0.5. "Pellet growth concentration" to 3. This seemed to achieve a similar result. Adjust as you see fit and pray it doesn't lag.

[JustMondo]

Wait how dis you make the 3 islands im trying to do it but cant digure what settings make this happen?

[Andrew Alekseyev]

Finally did it! No need to reply. Still think those two settings I mentioned might be helpful but it's probably because I'm a bit slow. :)

[Tipsy Jaguar]

Hey, I think you should add an air composition system. Different organisms would breathe and emit different compounds. The more of a certain compound is in the air, the healthier the affected bibites would be. They'd benefit from augmented lifespan, size, speed, etc I've always loved evolution simulators and I think its awesome we're getting one from Québec!

[Khâmul the Easterling]

This project is insanely cool and I love it a lot. PS: Not being a biologist makes your insights more important to me as I will have some bias. PSPS: Also you are learning first hand evolutionary concepts which is really cool and I cant wait to show your project to other interested people!!! PSPSPS: I think an evolutionary armsrace will happen between the Rozoculi species with eventually the infensus as the winner since they probably wont lose their generalism.

[DV Gen]

Kinesis for the win! It is really interesting watching how this played out, and to see the struggle identifying a lineage. There has to be a good solution for that, like tagging the bibites with some arbitrary ID that cannot mutate. I wish I could say that I have a solution that I recently implemented and suggest you do the same, but I'm really at this point, I'm not sure what the ideal solution would be.

[Revi M Fadli]

The One, the developer of LSES(Large Scale Evolution Simulation) did almost exactly that, but with mutation and crossover

[AlejoLab]

I recently implemented something along those lines 😅 and I suggest... to watch my next video 🤷 and use System.Guid.NewGuid();

[hobrin]

@The Bibites: Digital Life Yeah I added like useless ID genes which can mutate randomly to my little evo sim. So like 3 ID genes for RGB. But the more ID genes the better ofc.

[The Bibites: Digital Life]

Absolutely! In fact I had already written the script, but watching your video I had to go back and change a few things to fit that in there! Couldn't let this pass 💪

[wartech]

I think the reason the green giants are different from the ones in the western island is because you have the pink selection thing there. It would actually be cool if you could code another type of kill node based on certain behaviors that would kinda simulate different environments.

[Ginger Sasquatch]

I think I have a good idea to implement into the game at some point. You could add a random name generator to name virgin birth species and after some number of mutations or generations a new species name or genus name is generated. Not only does this let us track the divergence of species more easily but the names could be used by the species as a way of identifying relatives and a gene could control levels of aggression towards them, so predators have the ability to choose not to eat their babies, or encourage species to herd as a family group.

[kraft matik]

I run into this in my "by hand" manual evolution process 13:00 I call it template stages or bubbles. You evolve 2 different creatures over time and get completely different appearances, yet completely identical behaviors between species. It's a really awesome phenomenon, also helps me sleep easier when thinking about our first encounters with extra terrestrials.

[Like Zee]

I am avid Game Designer. I have been working on board and role-playing game systems since I was in 2nd grade, it made me fall in love with the concept of "rules". I got into designing computer games only recently and I have been working on an evolution system VERY similar to this one. Finding this was a gold mine, and I really appreciate your work. I'm in college rn and don't know any of the next steps. If you happen to somehow see this comment, what did you do to feel successful? What makes you tick man???

[zji]

So cool! What fun. I really like the time when you didn't keep track for 20 hrs and so got to be like a real paleontologist, trying to piece together what may have happened without all the data. I also Love how you name each species scientifically! It's amazing how the creatures vary, some even crossing the void to another colony. Gonna have to try this for myself! thanks for making such a fun evolution game/tool!

[Mands The Red Panda Pirate]

"He recently implemented something along those lines and suggested I do the same" "I recently implemented something along those lines and I suggest you do the same" I don't know why this was so funny to me Reminds me of Kung Pow Enter the Fist: "A good friend of mine told me there will be a chosen one" "There will be a chosen one" "And then he told me of its significance" "It will be significant"

[Gigantopithecus]

@The Bibites: Digital Life Bibites x kung fu panda

[The Bibites: Digital Life]

😂😂😂

[Mands The Red Panda Pirate]

@The Bibites: Digital Life "And then he told me Kung Pow is actually a great source of inspiration" "Kungpow is actually a great source of inspiration"

[The Bibites: Digital Life]

Ahahaha, kungpow is actually a great source of inspiration 😉

[AlejoLab]

I recently said something along those lines and also don't know why this was so funny to me

[naoki steward]

Super funny editing, it is one thing to be able to make that kind of complex simulator but it is another to be able to express the result of your work in a funny and interesting way. Please continue your work ! (salut de la part d'un ingénieur normand;) )

[The Bibites: Digital Life]

Merci! Je fais de mon mieux 💪

[Keliuszel]

Hi! I really like your evolution simulation running wildly! It has its own unique graphic style so its fun to watch, and also it has one of the most builded evo sim game i could possibly find out! Congratulations!

[The Bibites: Digital Life]

Thanks 💪!

[Shad Amethyst]

Interesting video, I love the paleontology work. As others have said, you could've avoided doing that work if you programmed in ancestry tracking. I wonder, have you planned on implementing sexuated reproduction? There are a lot of techniques and parameters that can be included for this. Also, at 18:55, I'd have loved to see the integral of the growth curve too.

[Sam Rodriguez]

This project is so so cool and has inspired me to get started on a project of my own! Any good resources you know of to get started on the practical elements of coding something like this out? Once again, SO cool. Keep up the great work man!

[Jacob Laney]

Just found your channel and loving the cute lil guys. I like that you’re sticking with this one simulation and just improving it over time I’ll be following their growth for sure :)

[Octo_Bill]

I want to try a similar thing and think the Islands are a fun idea, but everytime I start the simulation the spawn areas are inside eachother or only a bit away

[j7]

Yes! How can you seperate them?

[Christopher Gray]

This game/simulation is amazing, I've been messing around with it a bit. Could you please provide the settings for getting the 3 islands or even the saved world parameter file. I've been trying but the plant food blobs are spawning all over the map so keeping the different bibites on their "planet" to separate their evolution has been difficult.

[DD]

You probably thought of this but what about logging the x,y coordinates on spawn, as well as the parents details. This would help for migration tracking. I do love this channel, working on my own version at the moment.

[AhNoWiC]

This stuff is so cool, love that you keep showcasing it. Been following for a few videos, have been fascinated by the results. Keep up the good work dude!

[The Bibites: Digital Life]

Thanks! I will 💪

[t1sk1jukka]

Finally some new Bibites content! Thank you for this project, really interesting to see how things evolve. Idk has this been addressed before but could some sort of sexual selection work in this kind of environment or would it be un-doable?

[The Bibites: Digital Life]

In the work for medium term!

[Jemmerl Last]

Perhaps a tracker value could be useful too? Assign a random hex value to every *original* species, which descendants would all inherent. It wouldn't give a full picture, but it would be helpful in those "which population did this species come from" cases and shouldn't affect processing speed afaik

[Brain in a pickle jar]

Welp, guess I'm staying up for 20 hours •-•

[Warbrain]

@The Bibites: Digital Life i think adding a function for a tree would be useful. Like you can name a creature/assign it a species and then you can check if a creature has it as "ancestry"

[Gigantopithecus]

@The Bibites: Digital Life go sleep you deserve a rest

[The Bibites: Digital Life]

same, barely slept this night

[Ping]

Same

[Thelaretus]

As a Latinist I must say your attempts at giving them Latin names are hilariously ultrageous; they sound like you just wrote them on Google Translate (which is essentially useless for Latin). Keep on the good work.

[The Bibites: Digital Life]

Thanks a lot! And yeah... Mostly Google translating the attribute words I wanted until I found a Latin words that sounded nice enough 🥲😅😂

[Kai Doust]

This is probably the coolest evolution sim series on yt I have found. I'm binging all of your videos right now.

[AlienXtream]

coming from part 2 to get myself a refresher and i just had an idea. in future releases being able to have a once click "quick start" option that takes some predefined starting Bibites (with significant mutation applied) akin to the complexities shown here (as well as default recommended world variables). this would mean that people that dont have (or dont want to wait) hundreds of hours can get straight to the interesting part after the initial false starts. additionally, having a time-warp method to skip forward a few generations/timesteps as fast as the system can process it via a simplified smaller sample of the environment and doing some emulation of things (like for example instead of simulating the actual movement you estimate the probability of a creature getting food each tick and use that). obviously this would sacrafice some of the "legitimacy" of the simulation but it would be an advanced option and possibly useful for testing later, more complex, life forms for you too (as opposed to doing this for hundreds of hours). another cool feature that would blend nicely with that is (assuming it comes to steam eventually) workshop/creature-pack support. that is, downloading either preset configurations, Bibetes, etc. this would really be useful once the Bibete editor is done, allowing users to create, share and download curated creatures to spawn in and let loose in crafted scenarios. i mean, what good is a life simulator if you cant play god a bit and test your creations personally :P

[Alchemical_Moon]

The only problem with this channel is that there isn't more videos, I've binged all these video multiple times, really good stuff

[The Bibites: Digital Life]

Thanks! And yeah I know 😭 it's just so long for me to make a video

[Derik Duley]

Great video! The analysis and comparisons are so interesting. Would love to see more of these!

[Tuxu HDS]

"Cleanse the members that difference too much from our Eye-deals" *Presses the Thanos button.*

[Alex]

I really loved this video, I think it's your best yet! I especially loved the genealogy aspect and showing the species and how they diverged.

[The Bibites: Digital Life]

Thanks! I really loved it too 😁

[Yossi Zinger]

I'm new here, and very impressed! Would love to see some sort of built-in "evolution tree" diagrams, maybe with automated species naming along the way

[plants&stuff]

Great video! I'd watch an entire series of this!

[Arthur Boloch]

Here are some suggestions for your game: The taxonomy would be nice to see implemented (every x generation it's a new species or comparing how different the bibit is compared to his ancestors and the number of bibits similar) If there is species, some sort of list to track how well they are doing and their general characteristics A way to add a bibits ad favorite so that you can find them A way to load a save without having to recreate a simulation (if there is I haven't found it) Overall, i find it a fun project and really hope it will go well

[CrjaseMechaEngr]

This is underated, would be cool to see a school introduce this in science or something.

[TheBook Less]

This is literally a national geographic documentary, what the hell??

[TheBook Less]

@The Bibites: Digital Life you're definetly honouring him with this content, keep it up!

[The Bibites: Digital Life]

I should put on my David Attenborough voice

[Эдуард Сухарев]

You definitely need ancestry tracking and graphing system with a semi-random taxonomy generator, which should take those properties into account and name species respectfully but random enough to not get same namings

[Milokoj Jones]

Let me just say, that I really love the sort of little skits of yourself that you put into the video, they make it more interresting and enjoyable ( and they are quite funny ). As for which of the two is going to survive - I think that both will, but the Rozoculi Infensus will imo develop some sort of carnivore gene and start hunting the Rozoculi Rigicutis, which will than try to adapt - perhaps by growing more armour, which could lead to the Rozoculi Infensus hunting in packs in order to kill them effectively. This whole sort of battle could also lead to evolution of carnivores which eat the remains of the battles of these two.

[Lapis Liozuli]

I automated progressive time steps of an artificial life sim by running a batch file step by step. The downside is that at later steps, the memory usage soars and more time is needed to run the sim. Data storage for each organism's genome and lifespan also presented some issues which I've partially resolved.