They said make the simulation bigger...so I did!
ฝัง
- เผยแพร่เมื่อ 12 พ.ย. 2023
- visit brilliant.org/EightLittleBears/ to try everything Brilliant has to offer-free-for a full 30 days. The first 200 of you will get 20% off Brilliant's annual premium subscription.
Patreon: / eightlittlebears
TH-cam Membership: www.youtube.com/@EightLittleB...
Evolution simulator! Simulating natural selection! Artificial Life Simulation!
To stop carnivores from wiping out populations so quickly and to keep them viable for longer couldn’t meat be able to sustain a creature for longer similar to real life? (Apologies if this is already the case). This could lead to more diverse builds in the future.
I think carnivores also sleep longer than herbivores.
Probably ya. But it would just slow the inevitable wave of predators....maybe.
The wave of carnivores is probably inevitable without some of the smaller, more minute complexities of the ecosystem.
Because man, is the ecosystem incredibly delicate at times in real life.
I mean, it's been shown that one of the big causes of invasive species being so bad is the fact that parasites in the new ecosystem haven't adapted to them, and parasites help moderate life in a way.
This is already the case to an extent as we scale hunger with diet (more means = less hungry) - I am going to overhaul the energy system at some point to make this a bit better. Probably when I introduce insects and things like that. Thanks for the comment!
@@EightLittleBears Have you considered using eggs?
I made a simple plant, herbivore, carnivore sim a little while back which always suffered from over predation.
I added eggs that the herbivores would lay and hide / bury under ground away from predators.
After that If the predators swept through the population and then moved on or starved there were still eggs that could hatch and repopulate afterwards.
I'm not sure if that can work in your case as my predators would starve quite fast and also would often quickly move away from an area without food.
But worth considering the possibly of adding an adaptation for an alternate reproduction method.
I do love Big Hog Island
1. I absolutely love the art for the animals in this simulation
2. I was so happy to see my favorite animals, cheetahs, as one of the possible evolutions and even more excited when I saw that they popped up a couple times
3. I’m excited for when you add even more detail to individual tiles as I can see it as the next step towards many more exciting things. I.e. rain changing growth speed of plants and water levels leading to an increased chance of migration
Glad you liked it!
This is cool. I'm liking the increasingly detailed simulation. You're doing great work here!
Glad you enjoy it!
so as a note regarding the northern Oasis, Distance might just be the biggest reason why it was inhabited so much earlier then the southern one.
the smallest distance the Northern oasis and the Eastern forest is about 5 tiles meaning that the chance a creature migrating there is a lot bigger then the 15 to 20 tile needed for the southern Oasis compared to the Western forests
Good point!
Would it be possible to simulate biomes changing due to things like continental drift, ice ages, volcanic activity, meteor impacts and ocean acidification? I have no clue how you would do this but I have a lot of respect for what you've already achieved.
Anything is possible!
In the same vein as this, more subtle environmental changes, like gradual but consistent rise in temperature could be very interesting to see how creatures adapt
For diet, it would be interesting that instead of simply shuffling around the stats with the same sum, each diet type could increase/decrease on its own. The tradeoff would be that higher digestion efficiency would result in higher passive energy cost, maintaining digestive organs isn't cheap. That way the digestive efficiencies adding up to a fixed sum would become an emergent property instead of part of the system, though not always, since some environments may still support creatures with multiple high digestive efficiencies. This along with some general refinement of calorie density would open up a lot more avenues for evolution.
Interesting idea! I like the idea of implementing the cost of maintaining organs (this could be used for brainpower as well)
@@EightLittleBears God knows how much energy it takes for some of us to maintain some semblance of brainpower. It's a real struggle, sometimes. 😩
It would be nice if you could select some region and accumulate statistics for that region. With that you could select regions at the start of simulation and compare them later, instead of looking only at the whole world statistics and guess what they mean.
Yeah I agree - I think that is the next bit of reporting I will put in.
I was thinking for trying to extract data maybe adding a purely cosmetic hue shift to the evolution could help make it easier to differentiate between the different population easier. Both visually at a glance for the viewer but also as a point of data extraction so if you want to check on specific groups you can sort them by their hue shift for their grouping
I had a go at that a while back but it just kinda looked awful so was thinking of a different solution. I might do something like adding an outline that changes colour or a little disc at their feet or something. Thanks for the feedback!
@@EightLittleBears another idea for grouping could be an invisible long variable which is just an exact average of the parents with high generational variance. As the groups split but still have internal cross breeding they should quickly average away to different mean values within non crossbreeding groups even if they have a common ancestor.
How amazing would it be if you also added plant evolution. But that would be a lot of work and what you are already doing is already amazing!
I deffo want to add this, just need to figure out the best way to do it mechanically.
@@EightLittleBears Some parameters you could use for the plant species are the height and the presence and type of leaves (flat and needle but also spines). To replace the food requirement, plants would need some amount of nutriments (aka biomass accumulated in soil) and water. The bigger the plant, the greater the requirements.
@@EightLittleBearsholy fudge, that is something to look forward too! Algorithm brought me here
It's amazing the leap in complexity each iteration brings!
I love this simulation. I kinda wish the animals were dinosaurs.
That is in the plan 😀
OOOO so a mix of extict and alive animals!?! that would be amazing!@@EightLittleBears
I like the game of "spot the bear" whenever the camera zooms in.
I love these videos, and I really enjoy seeing your project grow.
You probably thought about it already, but I'm certain if you made this simulation easy to run and extensible, there would be quite the crowd wanting to buy it as they would other evolution games.
Looking forward to see more and all the best!
Nicely done, you really sound like a someone who wold narrate a documentary, that adds even more to the video !
Haha thanks!
Clearly the best strategy was to be on the small isolated island
The animals are so CUUUUUUUTE OMG. This simulation is next level, love the improvement. The migration theory is pretty good, kinda similar to what happened in real life with people living in the easier, more hospitable areas being able to run around everywhere else because they had the means to. Lookin forward to your next one!
One thing to note is that, deserts are definitely not void of herbivores or plant life, they are just as wild and diverse as the rest of the Earth. Go watch the livestream of the Namibia desert and you'll see so many different types of animals and insects coming by to drink, its incredible! Deserts definitely act as buffer zones between populations, the same way oceans, jungles, and other such terrain that is unpassable, except for the highly specialized animals that live in it and can't go anywhere else.
You should be able to have species be named after patrons. I think that'd be a fun way to encourage more people to visit it and give a little personality for each species in the simulation.
Can’t wait for more evolution!
Exited!!!
Good video!
I love the white bear at 22:40
Loved it! But when are you planning to introduce water as a habitat? Seing some oceans and lakes with creatures would be fun.
Thanks! Probably next year I think. I need to do some research on aquatic biomes first. The other thing I find intriguing about aquatic biomes is that creatures can pretty much all move in three dimensions (whereas on land this is pretty limited in comparison), which makes it a lot more dangerous, so I want something to take this into account ideally.
To solve the mystery extinctions without carnivores present - look at the food strategies available.
You let them have different vegetarian diets, and those extinctions happened in leafy biomes.
I bet you had a situation like: the species focuses on Sap for energy, and did well. But then they ate all the sap producing plants and starved because they were too specialized for sap and couldn't transition to anything else.
It’s a good theory - I had a look at the diet data I couldn’t see anything that confirms this so I left that part out. It doesn’t mean it isn’t right, just means I need to improve my data reporting!
This is great! I’ve loving all the little details, and the camera zoom you put in was an excellent and welcome addition!
Thank you very much!
The food change is wonderful!
You've made impressive progress on this. I cannot wait to see what you add in the future. I could give a billion suggestions, but I'm sure you've got more than enough to do already.
so glad for another video from you! awesome!
Glad you like them!
i love your work! its great to see these simulations.
Glad you like them!
Very cool. Enjoyed it a lot!
Awesome, thank you!
Great job!
The amount of effort you put into this is amazing honestly 👏
The efficiency carnivore seems something like a tasmanian devil. Its not particularly stealthy or large or fast and doesnt breed quickly.
This is absolutly facinating.
Fantastic video, this was extremely interesting to watch
Glad you enjoyed it!
Amazing video!!!!!!
This is amazing. This is the kind of thing I could only dream of after seeing spore for the first time and wondering: "Wait that's not how evolution works"
Found this channel a few months ago. Cant wait to see the evolution progress further! One of the most underrated projects imho. Do you have plans to release it?
Yes eventually - not sure when though!
have you thought about making the world temperature fluctuate? A big part of evolution is adapting to new challenges as they arise and it would be a cool way to incorporate them. maybe have tiles change up or down depending on the temperature at the time? You could have water turn to ice or swamps to form land bridges as well
Easy for me to say, im not the one that would have to code it I know. Awesome work!
Next patch has the basics of seasonal and daily temperature changes. It won’t have thermal regulation (which will come later) in creatures but affects the environment and, by proxy, creatures.
woot
Not me taking a ton of time whenever he says East or West to figure where he is talking about 😅
no clue about the back end, but i would imagine the easiest way to view regional data would be to manually define them beforehand instead of making the simulation itself identify unique booms to track. Just as an example, i cant fully tell if the narration is hindsight and to what extent, but you estimated having "outposts" at the north and south oasis; define them as such. You start an isolated rain forest which had to repopulate; define it as a population. The island was a closed system off to the side until you integrate the ability to cross water; define it as another region. Heck, this last one is less of a way to track population, but instead the hardiness and willingness to travel, but you could also define the large central desert so as to track exactly when attempts at great migration (and population collision/reaction) occur.
I lightly touched on it beforehand, but it looks like the single biggest factor yet to be addressed is aquatic life. Anphibians can hide in the water at tough times but repopulate the land with groups to mutate. Fish provide a food source or food boon to adjacent tiles, and plenty of other unforeseen ramifications. It should be easier to cross a desert along the banks of a river, even if the biome is too hostile to support plant life to attract prey, heck, the nile river is famous for a reason.
As the simulation gets closer and closer to mimicking humanity, changing the value of tiles would also be interesting. Maybe plant life cant survive in the desert, but that changes once humanity starts irrigating crops. From there, some prey might form a symbiotic relationship, "stealing" crops in return for becoming a source of meat, and from there, attracting predators. That said, humanity is probably still a very long ways off and requires evolution and mutation to bestow certain powers and privileges that change the way populations interact with the world.
Just found your channel and it’s amazing!
Have you ever thought about changing amounts of meat on a tile for the size of the dead animal killed on that tile?
Thanks! And yes the amount of meat dropped on a tile currently scales with size 😃
It would be cool to see what would happen if climate shifted with time too
In the pipeline - there’s quite a few bits to put in place for that though so not sure how long it will take!
I'd love to see something like global Ice Ages affecting the tiles. Or introducing seasons occurring over regular intervals as an evolutionary incentive.
Ice ages are a while away, but looking into seasons now so hopefully they will be ready in the near-ish future, at least in a simplified form!
You should add another stat called intelligence and basically the higher the stat the better choices they make, if to fight another one for food or teaming to jump another animal.
could you share a tutorial for something less advanced (how you started the simulation and what programs did you use)?
hey a quick question
what language do you use to program it all?
C# (via Unity)
Man I really like these videos Mr youtube algorithm pls recommend it to people
I really love this content. FYI your Patreon link is not hyper linked
Thanks! Did not realise that!
You should add decomposers, like mushrooms. And scavengers too!
Really cool simulation! I would have loved if you would have gone a bit deeper into the different food types. How are they different? Does algae only grow next to water tiles?
Glad you liked it! At the moment there is no real difference in food types (apart from meat), but they do grow at different rates in different biomes. So grasslands have more grass, rainforests have more leaves and fruit, etc. algae is pretty rare everywhere since we don’t yet have aquatic biomes.
Desert looks like a terrier
just wanted to post to say this is really cool to watch, personnaly i would prob buy it even just as a sim, like 5-10 bucks
it's really interesting to see how the population evolve over time
it seems that life have a really hard time enduring in environnements that are not easy, maybe is it due to the "small" map/ time period compared to earth?
Thanks -that’s amazing! I am planning on making it playable at some point and releasing on steam or something. Not sure when though.
With the map I totally agree - I think the with more time/area we would start seeing certain specialisations arise in all of the environments.
Best evolution sim
Weird how TH-cam sent me a notification for a privated video on your channel right now about the winter or temperature
I got hit with a copyright claim for some background music that was supposed to be safe to use. A new version of the video with that music removed is processing now. Sorry for the mix up!
I'd love to see a simulation with a shifting/changing environment. Such as water making new routes and spawning more habitable tiles alongside. Your content is fantastic 😁
Thanks! These are all things I want to implement at some point. Wind patterns too!
@@EightLittleBears I can't wait to see it when you do!
Do creatures have any type of defence mechanism? Aside from running and hiding?
Does the simulation have rivers, or ways to track erosion. Like what happens when there's too many grass eaters in an area? If no, and you do this again those might be good factors to add.
It feels like maybe herbivores should prioritize their favorite foods. Like, order the list of foods by how effective that food is for them, and then looking for the "best" ones first, and as they run low on foods, start upping the odds of trying to eat the less effective foods, as they get desperate.
cool stuff
add insulation with temperature next
This is on the todo list - probably not next but pretty soon I think, along with seasons.
Ok now make it so that your code is a 100% accurate simulation of nature. If it’s not exactly perfect, I will make it my life’s mission to prevent your success as a content creator.
(Seriously though, this project is awesome. Don’t stress yourself about making it perfect or accurate, just make it in a way that you’re proud of
Hahaha thanks!
I would like to see like the total number of creatures and the different types
oh my days you're running this on a laptop, no wonder it's so hard to do these. truly commendable that you're still soldiering on, these are awesome
oh, by the way, are there any plans for necessary hydration levels for various species as well? as well as the ability to gain certain amounts of hydration from food sources? it could lead to supporting certain desert populations just like how we have in the real world.
Hey I just like making them 😀
With hydration, for the moment there is just a very simple system in place where thirst builds 2 (or possibly 3.. can’t remember) times faster than hunger, so it is essentially a secondary “energy” requirement. I want to add hydration from food eventually though as I think this would open up opportunities in desert biomes!
@@EightLittleBears ah, got it! that makes a lot of sense for most animals I think, but yeah I think eventually it'd make sense to vary levels of hydration required for different animals and for moisture to be obtainable from different sources, it'd probably give some very interesting results!
« the world change because they changed it and once it was changed, they were doomed, but the world moved on. » 15:07
The CHILLS dude
Is the gaint eland supposed to have over one thousand in size?
Nope! Good catch!
What happens if you randomize every tile?
You should add hand placed radiation/mutagenic fields to simulate all the ways our natural world enhance mutation speed.
That field would have a base chance for any creature thats on it to randomly die while on it. A (higher) chance of a create that is on it to randomly mutate and both a higher chance for births on the field to mutate and a higher change factor if those birth mutations occur.
It could also have a modifier that enhances effectiveness of mushroom recycling ability.
That is an interesting idea!
"The world was changed because they changed it, and once it changed, they were doomed"
exactly
What if you build a base, like 5 more tiles in each hexagon with a smaller ecologically like insects, use it as a factor to change plant growth. No clue if even doable but your an amazing content creator and this is super cool, keep it up.
this was really good maybe make the map even bigger?
I was wondering why the Eastern population developed stealth so early when there weren't any carnivores present for most of history.
Even though there aren’t carnivores, they were all omnivores so predation would still be happening.
Stealth is pretty powerful in that scenario and, if they go a build that doesn’t rely on speed, then it doesn’t really cost very much.
Thanks for the comment!
it's strange that carnivores are so similar to enviromental nukes...the fact that they can't stabilize the population with the other ones and reach an equilibrium is strange....the code for the carnivores migth need to be tweaked
Yeah, I think some of it is about the size of the map. There were some instances of carnivores not destroying the environment, but they eventually faded out of existence themselves. One thing that is worth noting is that most real animals are omnivores (even like foxes and things we might usually think of as carnivores). It might also be that life is just hard in a world where we don’t have big herds of big herbivores or something like that. Guess we will see over time as new stuff gets added 😄
@@EightLittleBears one thing that migth be interferring with this is a lack of specialization of the predators, or it seems like there is not, maybe allowing them to focus on specific types of prey could be helpful...or maybe increasing the food values of meat, so they can kill less.
also i don't know if predation has a chance of straigth up failing or resulting in the predator death...
Cheers mate thank you for such an educational and entertaining video! My take away is that if we dont take care of the world we live in we can expirence an extinction event sooner rather than later.
What are you using to explore the data out of your simulations?
I use different things from sim to sim. It exports everything to a CSV so I have a few options. Sometimes just simple spreadsheet software (cuz easy) and in some I use some custom chart software I made. As things get a bit more complex I might switch over to python but I’m thinking it’s probably better to just build a library in Unity (otherwise when I eventually release the game it will be a hassle).
@@EightLittleBears I really love your videos. I’m a data analyst and I work with PowerBI. It’s a powerful and relatively simple tool you can use for free for single use. It’s really powerful in exploring data, if you want you can share some results in csv and I can create for you a demo report to try and dig in your data
interesting thing about irl herbivores is that very few of them will always turn down meat or carcasses, especially during lean times. Deer in particular are known for scavenging bones for calcium for their antlers! But that might be a bit granular for your simulation lol
regarding 16:56, how come the energy-efficient carnivores were small then?
Can desert sustain any population or only creatures we see there are those that wander too far?
I’ve not seen it sustain a unique population yet, but I have seen certain large animals do well because they can wander between oasis points. In theory it should be possible for very small creatures to make it, as deserts do have (limited) resources
Going to need a quantum computer soon
Wow loved the simulation, something that could make it intresting as well woud be enviorment changes trowout time sound hard to code so maeby not
Totally agree - it’s something I want to do eventually!
@@EightLittleBears meaby coud me like tiles change overtime if condition a or b is next to it
Yay!
make the taiger a little gray or somthing like that, so it looks more hostile
My prediction is that the smaller more efficient carnivore is more successful. Less stealth to deal with.
maybe implement mitochondria. Only given from mothers. Mutations are only slight amd insigmificant. Great to track population migration by just looking at (colot coded?) mitochondria traveling. A random letter/number string that mutates one single letter every 5 generations should work for that (maybd use te RGP colors as leters and let them 'evolve' just slightly among all colot axes)
Ooh, a thought, mayber very rarely (Like once every 100k turns) a very small population of a randomly generated animal appears at a map edge, simulating invasive /, migrating species?
And I think it would help if you also add small groups of carnivores
What are your thoughts on how to create a world where a Desert Niche can begin forming?
I think the desert region probably needs to be a bit bigger to encourage that kind of socialisation. There’s basically two strategies that I know of for creatures in real life for extreme deserts:
- be a lizard, which basically needs very little energy since cold blooded animals are capable of surviving on incredibly small amounts of food and water.
- be a camel - i.e. an animal large enough to travel hundreds of miles, with a special adaption to store and carry water/energy(in the form of fat) until you get to the next oasis.
Wouldn’t it be cool if the world wrapped around so east teleports them to west and west to east
Also, fish.
Ha - I did think about this but I thought it might make things a little more difficult to track visually. Plus the edges make it a bit easier to get some isolated populations so the map is sort of a win win.
Fish someday!
To keep carnivores more stable, maybe give them a success rate depending on the density of prey in the hunting grounds?
I don't know if it is within the scope of the simulation... but plantlife also evolves depending on what gets eaten less/more, and if their reproduction is aided, or hindered, by getting eaten.
If plantlife also varied, meaning "biomes" become more varied, perhaps there will be more diversification of species?
It is in scope eventually! Not sure when yet, as I would need to re-architect the system a bit. But at some point!
make it real big
I wonder, would animals have a "Fearfulness" stat? To simulate the way some animals are very paranoid of others, while others arent?
I e, some island populations dont tend to react to introduced potential carnivoree, i.e. humans, while in other places animals are very quick to frighten.
It could probably affect the ease at which animals can utilize their speed, though at some sort of detriment? Maybe a higher stat incentivizes them to move away from areas with higher animal counts, or animals that are different to themselves if they are a herbivore, while also using more energy or lowering life expectancy (idk?). Though at a lower fear stat means rhey wont properly avoid predation.
It coulr cause predators to spike from time to time, as a low predator environment would trend towards a more placind population thats easier to hunt, while an actively hunted one gains a better ability to avoid predators
Idk, coule be an interesting thint to consider
Wow. This actually reminds me a lot of Dwarf Fortress but instead of DND world generation we get spore 2.0 here lol
you should add birds
So like... take the size of this map and think of it as a tile... then make the make 5x5 tiles large... please.
I think adding rotting meat would be good.
You should make animals able to evolve to go in water or just cross bodies of water in general. There could be a ‘water mobility’ trait.
you forgot youtube likes comments too (btw video was nice love your conntent)
In real life carnivores help stop over grazing of environments- look up Yellowstone wolf reintroduction for an example. Would be very cool to see over grazing affecting plant productivity.
I agree! We have this in a very minor way, but I am going to look into a better plant life system next year which should better simulate stuff like this. Thanks for commenting!
What is this game called? I’d love to play it