Simulating Natural Selection - Living Biomes!

If you want to support this channel financially, please feel free to join:
- Patreon: / eightlittlebears
- KZbin Membership: www.youtube.com/@EightLittleB...
Thanks so much for watching!
Let’s test the limits of artificial life in our evolution simulator!
Another step in the journey towards making a simulator game which allows us to watch natural selection in action. Learn (with me) about biology, ecology, and zoology!
#simulation
#simulator
#nature
Watch my original simulation! • I Simulated Evolution:...

[@EightLittleBears]

The raw footage for this simulation was over 9 hours - so I thought it might be fun to do a "Where's Waldo/Wally/Holger/Charlie/Walter/Any other that google didn't immediately show me..." style game where you point out animals I didn't catch! Have fun!

[@monochrosomnia]

with more detail being added over time i can definitely see this becoming one of my favorite artificial life sims, hopefully some sort of public access is released soon

[@Mo_lester123]

me to

[@EightLittleBears]

Thanks! I definitely want to release a playable version - the current issue is that it’s a bit of a house of cards so I just want to make sure all the core systems are there and stable before releasing anything.

[@moonrunesv2931]

Excited to see this! @@EightLittleBears

[@bensonbrotman1451]

@EightLittleBears will it be on steam

[@slwrabbits]

I would like a version where I can just have a rabbit population explosion and bask in a pile of bunnies, pretty please. (an unrealistic god mode or something)

[@phyose4793]

I'd love to see you give a mostly aquatic environment a go to challenge amphibious play styles, as well as shoreline ambushers.

[@EightLittleBears]

It’s on the todo list!

[@samueltrusik3251]

I still like the idea of having isolated island communities ocassionally started or disrupter by some sort of water crossing. Hovewer, there might be an another way to accomplish this. If the terrain itself changed, with new land appearing from the ocean, and vice versa, maybe creating mountains in the process, this would allow for very drastic changes when continents collide. Environment changes on a more fundamental level just seem fun.

[@EightLittleBears]

This is definitely something I want to pursue! Not sure when though as I would need to implement aquatic biomes as well (which don’t work with the Whittaker Biome System). The idea of continents colliding is interesting…

[@JerbilKonai]

@@EightLittleBears You would probably need to calculate where water would be heated up and cooled down, since water temperature is important for both biomes (e.g. coral reefs) and for currents (hot->cold and back) and therefore propagation of nutrients) Add to that that water temperature also influences the nearby land (e.g. there are deserts right at the coast, but due to the cold water the most water they see is a bit of mist) and just the addition of water ecosystems could change the very lay of the land.

[@Jorge71783]

Your voice paired with the intelligent, concise way you speak is so easy to listen to

[@EightLittleBears]

Thanks so much! I have to admit I always think it sounds terrible but that might just be the curse of recording your own voice…

[@art-eroflore]

as a biology student i LOVE artificial life simulators and I appreciate that this is one where you included some information about nutrient cycling! a couple of thoughts, I hope you're okay with me sharing these. Regards to the video watching experience, it would be nice to see it zoomed in more often so I can appreciate the little creatures. Additionally, maybe you could make some sort of color filter connected to some of the "invisible" stats in order to make it more obvious to the audience which populations are actually diverging? Even if there were like, a filter button you could press so it turned on a "green" filter that highlights herbivores, and a red one that highlights carnivores, and such in between. It would be a nice QOL change just for the video / data experience. And I gotta throw in some other ideas just in case you're looking for other ideas. I know you have a perception mechanic, I quite like it as representing predator/prey interactions (since most predators catch very few of the prey they chase). But I think that if you created some sort of "herding" behavior, where groups of herbivores had an increased perception stat without the energy penalty, that would really push the mechanic towards doing something obvious onscreen. There would be some solitary herbivores, like rabbits, small, speedy, stealthy and with high perception, and then the deer would have low perception on their own but increased when near others. This would allow the potential development of sort of "parasitic" behavior, where an herbivore who is not perceptive at all joins a group to take advantage of perceptive members of the group. The other thing is r-selection versus K-selection, which might be good tweaks to the reproductive system. r-selection species produce many babies and hope one will live to adulthood, whereas K-selected species produce fewer, "better" babies and take care of them. in the context of a simulation I suppose you could simply make the r-selected babies have a very wide variation in stats so some of them would end up with such bad stats they would die; the K-selected babies would have stats closer to their parents.

[@EightLittleBears]

Thanks for the detailed and thoughtful comment! I really like the idea of perception increases based on herding in particular! r-selection and k-selection are already in there to an extent (essentially what you have suggested with faster reproduction resulting in weaker offspring). I want to make this more detailed in the future, but there’s bigger things to focus on now!

[@janschmalz7980]

Wow, I really like this ideas and totally think they would benefit the content

[@apple_juice_person5595]

You should do this with a more scarce or energy filled environment like a desert or rainforest!

[@EightLittleBears]

That’s how it works! I use the Whittaker Biome System to calculate biomes based on temperature and precipitation on each tile. This video goes over that part kzbin.info/www/bejne/fmKxiXZqhsqHpc0 😊

[@scdijkens]

@@EightLittleBears I think he meant running a simulation with a world that is predominantly covered in that biome. Then making a video of it.

[@EightLittleBears]

@@scdijkens ah! In that case… coming soon! The next video will have a much larger map with lots of desert. And at some point I am going to do a sort of “versus” one with a scarce v abundant map

[@scdijkens]

@@EightLittleBears Sounds cool! I quite enjoy these vids. It reminds me of a hexagon based evolution program I made myself a couple of years ago but never got to continue on.

[@EightLittleBears]

@@scdijkens pick it up again! This is actually like my 10th go at making an interesting sim… think it’s more about consistency than anything else 😃

[@aidanwhite5631]

this is so cool. I would love to see this run for longer to really see how much more this has to offer

[@erinkarp]

I'm excited to see where this sim goes!

[@EightLittleBears]

Me too! It’s a journey!

[@TheGrinningViking]

I was just commenting something like this in another video! Deer are entirely herbivores and unsuited to eating meat, but if they are very low on salt they will crunch a rabbits head. Rabbits get more incidental minerals because of their digging and lack natural defenses that can affect a deer, so they seem to be the preferred target. It really only works because the rabbits just don't seem to expect it.

[@cloverring]

It’s incredible how much your simulations become more and more complicated at such a short time in between each video! Great work! Definitely subscribed after binge watching your content the last half hour-ish!

[@EightLittleBears]

Thanks so much!

[@toast699]

Just found these videos a few days ago, love your content

[@alexandrasmart1830]

Just found this video it's right up interests and it production was great! Awesome work! Subbed

[@vinestick]

Love the style of videos you do, it always feels like an animated documentary. Looking forward to mor from you, great content!

[@EightLittleBears]

Glad you enjoy it!

[@Essentially_Nobody]

Have you tried the simulation with larger maps? I wonder if more species could exist if the biomes were larger so that a generalist couldn't take over all areas as they would have to spend larger amounts of time in certain areas

[@EightLittleBears]

Yeah, I made a few changes recently to improve performance at scale. However some of that progress was undone by the fact that the new diet system supports more creatures per tile 🤷🏼‍♂️. I have a few ideas on the back burner about how to deal with this, but every update involves more calculations so eventually im just gonna need better hardware! The other option, I think, is to just let the sims run much longer. I am going to aim for 600k turns next time and see what happens. The problem I run into there is storage space for my video files ha. But there’s ways around this so shouldn’t be too much of an issue!

[@modulebot]

awesome video, I hope your channel gains more traction

[@DarkIce0290]

Just found ur channel, this is a really cool project, and I particularly enjoy how you layout and deliver. Looking forward to future videos

[@EightLittleBears]

Awesome thanks!

[@ladicusaiur1513]

Really cool video! I can't wait to see what you do next

[@EightLittleBears]

Thanks!

[@FQT_Keller-Ash]

Another great video thank you so much! I was so excited when I saw your video, and I was not disappointed!

[@EightLittleBears]

Awesome! Thank you!

[@yorifant]

this is amazing!

[@Ancusohm]

This is really neat! Thansk for making these videos.

[@EightLittleBears]

My pleasure!

[@chadgorosaurus4898]

This man needs more subscribers.

[@EightLittleBears]

Thanks!

[@RizzCentral-zx5rt]

Im an OG watching all of your videos since day 1. KEEP IT UP 🎉!!!!

[@fabiancosta340]

This is so much fun to watch and I just love it.

[@EightLittleBears]

Thank you!

[@karue7581]

Your content is amazing!

[@EightLittleBears]

Thank you! ❤️

[@tristanreejakobsen6157]

couldnt you also try to change their ai as this one roams more and this one stays closer to water. oh and love these videos oh and maybe also add seasons and flight and swimming.

[@EightLittleBears]

Haha - all on the list!

[@stealthyturtles]

Yo this stuff rules dude. I’d be interested in seeing some of the math used to calculate values or programming too, or general theoretical concepts expanded upon. Definitely earned a sub for me, super cool channel.

[@EightLittleBears]

Thanks! I’ll think about going into the math/code - it isn’t super organised at the moment though…😅

[@razenburn]

That outro convinced me to subscribe

[@adaml6703]

I think it would be interesting to make the calculation of resource expenditure per turn more complex than being proportional to individual traits like size. For example, it would be more expensive to be fast and large, large and stealthy, small and perceptive. ex. something like cost = size*speed + size*stealth + ((max_size-size)*perception) + ....

[@EightLittleBears]

It is based on something similar! I go through it at a high level in my new Q&A video if you are interested (around halfway through I think).

[@a.v.y8331]

I wonder how things would go over bigger maps (with more defined and separated biomes) and longer periods of time (although that might end up bricking your pc idk how computing intensive this program is). I feel like with every video you add in more and more real-life factors that make a significant difference in the operation of the simulation and I think it might be really interesting to introduce divergent evolution in the next one through these kinds of environmental factors.

[@EightLittleBears]

Bigger map is something I want to do but, as you have suggested, I am limited by computing power… I did some optimisations which roughly tripled the number of creatures the sun can support but, coincidentally, the new diet system also roughly tripled the amount of creatures my environment can support… lol. I have another solution which scales more or less infinitely, but at the cost of slowing down the simulation (essentially it just splits up the processing over multiple frames). So I’m theory I could run a much larger simulation now, but I’d have to run it for about a week to get to the same number of turns. This isn’t a problem necessarily, it just means I can’t use my computer for a week (although maybe i can treat it as a detox period ha). Anyway thanks for the comment!

[@torymiddlebrooks]

What an incredible display understanding and experimentation!

[@EightLittleBears]

Thanks!

[@morgan0]

something i’ve wanted to try making for ages but haven’t had the time/energy to work on is a distribution based evolution system, working on the distributions of traits in a species rather than individuals. could be a good way to simulate more with less cpu time, but also more complex to code. basically a bunch of trait distributions and optionally some way that each is mapped to the others, so when selecting for one, you know how it affects other traits, which would allow for species separation on multiple traits at once.

[@EightLittleBears]

Interesting concept.. you should definitely have a go at building it. Getting a basic sim is pretty quick to set up. It’s just that it becomes a never ending project as you keep thinking of more and more things!

[@morgan0]

@@EightLittleBears maybe sometime i’ll see if i can get single trait selection working, that might even be simple enough to make as a desmos graph lol

[@Ommartheegreat]

This is becoming scarily accurate Episode 49… the third galactic war

[@EightLittleBears]

Haha …an interesting concept…

[@richardjones8353]

So im an ecologist who appreciates games and ive been DYING to see something like this be developed. So much so I began teaching myself game design and started developing a concept strikingly similar to what you have here...i would LOVE to collaborate with you if youd be interested. If we could expand complexity while also maintaining a certain flexibility, i feel that this kind of work could really help to answer some pressing questions RE ecosystem mgmt, climate change, etc

[@richardjones8353]

I see from the comments that you're also thinking about aquatic systems too...i have a grad degree in marine ecology and currently with wetlands...just saying 😅 I'm glad you're doing this regardless

[@EightLittleBears]

Hi! Your degree sounds super interesting! I’m not looking to collaborate at the moment (this project is basically my whole hobby so I kinda just want to see where it goes) - thanks so much for the kind words and support though! It means a lot!

[@richardjones8353]

@EightLittleBears dang, oh well, that's understandable tho, i get that this is a lot of work and you wouldnt want it to be more!....I guess I'll go back to my drawing board for my idea then...I don't suppose you'd be willing to share your Unity code? (I would sign whatever NDA etc you'd like if you did). I dont have much time to focus on learning game design but have a lot of ecological models id like to test....regardless, ill be watching your progress with great interest!

[@SteursSarah]

I love evolution simulators, so I'll be following this for a long time to come!

[@EightLittleBears]

Glad to hear it!

[@DivineWerezwolf]

Be intresting to see aqatic added in with its own bioms

[@zane_sadauskis]

I dont fuller understand alot of what is talked about but its fun to see this sim progess and to learn a few new things about programming and biology on the way

[@EightLittleBears]

❤️

[@abhishekvchaudhari8181]

Good stuff

[@sicksock435446]

Interestingly the "large herbivore = Dense foliage" meta is somewhat reflected in real life with asian elephants, who essentially form entire biospheres around themselves by pushing down trees and encouraging grasses and reeds that they graze on.

[@EightLittleBears]

I really like this as an idea for the simulation in the long run. I think a few others, such as leaf-cutter ants, fall into a similar category of “altering the biome” so it would be cool to have this sort of effect!

[@DivineWerezwolf]

Would be intresting to see co effecienys in bioms say positive speed for savana but malic in rainforests.

[@aleafmusic1681]

This is pretty cool!

[@EightLittleBears]

Thanks!

[@rdmdude734]

Btw good Video you did great agien

[@thehoustmaestro]

The art style of these creatures is amazing, so cute! 🥰 When will you be selling these as plushies!?

[@EightLittleBears]

At some point!!

[@Lobotomistmankyeh]

Neat!

[@EightLittleBears]

Thanks!

[@edwardmighetto7327]

It would be cool to see some behavioral evolution as well. Maybe some social mechanics or pathfinding / biome preference?

[@EightLittleBears]

Agree! I am going to do “intelligence” at some point but not 100% sure how I want to tackle that. Will probably be some form of neural network but will see in time !

[@edwardmighetto7327]

@@EightLittleBears Interesting challenge to use neural networks for this. (I'm an AI researcher at UCSC). I would definitely worry about system speed with the population sizes you're seeing in this video's. How exactly you would make it work with the evolutionary algorithm is also interesting. I suppose the int stat could correspond to the size of the nn in some way, and then train it as the organism ages based on some objective measure like average hunger? You could also have the initializing weights of the nn be genetic, but codifying that as a stat or even interpreting it at all would be very difficult. The classic 'agent in environment' algorithm is q learning, but I also struggle to see how to codify that as intelligence or have it develop evolutionarily. An interesting implimentation I that expands on your current 'state based' ai that moves away from nn's would be a state system. Basically each models 'brain' is a set of states that altar the behavior. 'intelligence' could be codified as the number of reachable states. Events could swap the agent between sates, and the weight on a state change could be genetic, similar to the food preference seek plants or meat mechanics. The disadvantage with this is it would require some hard coding of what the states are, unless you want to make that a learnable parameter which would be difficult. Could be a good system for making cool videos though as it would allow for some specific tuning of what behaviors can arrise, I'd personally really like to see species that specialize for certain biomes.

[@EightLittleBears]

@@edwardmighetto7327 thanks for the the insightful comment! It’s one I will definitely come back to when I start implementing intelligence!

[@rhanb9920]

Thanks for posting something cool! You know, as a fantasy nerd, I kind of want to see how the introduction of a "Magic" stat would affect the simulation.

[@EightLittleBears]

Ha it’s funny you say this.. I was just chatting to someone about adding fire bears and stuff when I eventually gamify the system…

[@rhanb9920]

@@EightLittleBears that sounds awesome 😎

[@ryanmapping7944]

Ideas: 1.climate change: the "average temperature" could change, the sea level could rise and lower and the biomes could become different over time, a great speciation creator, however, it would possibly need to have limits. 2. Continental drift: complex, but would be an complement to idea 1.

[@ryanmapping7944]

3a: flight by speed: with enough speed, creatures could take off by being Very fast 3b: flight slider is not good enough 4: civilization: this would complement idea 1 again, due to antropic changes.

[@EightLittleBears]

All interesting ideas! Thanks!

[@isaiahcaswell4116]

Look into Regenerative Agriculture, more specifically Rotational Grazing. It's a real-world example of larger animals leaving healthier vegitation. There are also synergies with smaller omnivorous spreading waste and eating parasites.

[@EightLittleBears]

Great comment! This is partially simulated through nutrient cycling, which takes waste into account, but it’s definitely something I’ll be expanding further! I’d love to see the effect of returning bison to the American plains!

[@isaiahcaswell4116]

@@EightLittleBears Great Video! I'm looking forward and excited for what you'll be able to accomplish. I'm right there with you on reintroducing Bison! That would be a much better use of the land than the corn we have now.

[@ForestRaptor]

I wander if making a packet of these simulations into a exportable thing, and a couple of hundreds of us run them, if it would give you more data to kinda be able to see general trends or exceptional evolution lines. I know it would be a fun experiment ^^

[@EightLittleBears]

Interesting idea… I am hoping (without any promises) to release a playable,albeit buggy, version by the end of the year so maybe something like this can be built in!

[@DawnTyrantEo]

Perhaps considering specialisation would be an interesting direction to go in next? We do see it to a degree, but now there's a bit more focus on energy and nutrient efficiency, having different builds being able to waste resources due to being unspecialised (e.g hot area super-herbivores having less waste and more growth eating hot area plants) would be interesting.

[@EightLittleBears]

Yeah that’s a good idea! Need to implement temperature specialisations, but I’d like to have “territories” in first so creatures are a bit less random in their wandering ☺️

[@DawnTyrantEo]

​@@EightLittleBears Perhaps 'ant trails' as an option? E.g creatures mark their movement with a tag of something like 1-64, and if it has their number, they'll follow it.

[@joshuasteward6672]

Really cool. I have a couple of questions: What's the trade-off between eating meat and eating plants, as far as your simulation goes? It seems to me that the real-life tradeoffs are plants, less energy-dense but requires less energy to digest, and you don't have to hunt. Meat is more energy dense but requires more energy to digest, and then of course you have to hunt it. It seems that that most real-life carnivores are hyper-specialized to hunt down prey, whereas most omnivores lack hunting ability and sort of just scavenge for whatever they find. Maybe this means that eating meat is only worth it if you can get it without expending too much energy. How does your simulation handle expending energy to hunt? Anyway, that's just me rambling, I think this is super cool and can't wait to see more.

[@EightLittleBears]

Thanks for the question! I don’t have a very sophisticated solution to this, but for the moment the rate at which hunger builds scales with diet, so essentially an obligate carnivore will need to eat twice as often as a herbivore. It’s something I want to expand on in the future but for now I think this solution gives a reasonable approximation of the effect we are looking for.

[@diegolamanya345]

bro will end up creating dwarf fortress

[@douglas0828]

next step is probably simulating disasters on population and food suppy or diseases also maybe simulate health over time

[@EightLittleBears]

Interesting ideas!

[@alansmithee419]

I'm a simple man (I assume lol) I see "simulating natural selection," I click.

[@EightLittleBears]

I like your style!

[@callyfoot]

im shocked that with this easy to understand simple video which i understood you dont have more views yes scarcasm

[@EightLittleBears]

😢

[@DivineWerezwolf]

Preception might need to be broken down into sub catagories. For smell and sight

[@Cython35]

I just found your channel and have no coding nor ecology degree's but one thing stuck on my mind during this video. Can more than 1 animal exist on a tile? If they can't i think that would cause speed as a stat to be quite worthless since after a certain point it gets quite crowded.

[@EightLittleBears]

Yeah any number of creatures can be physically present on a tile, though obvious the more there are on a tile, the more quickly they will deplete its resources and kill each other!

[@rdmdude734]

I think i will win my Bet that you have 10k subs+ Till end of year

[@EightLittleBears]

Thanks so much! I remember that comment being very motivating!

[@tuures.5167]

At this rate, I wouldn't be surprised if the sub count hits 10k before the end of the _month_ . Glad to see a lot of people discovering this awesome content!

[@Schneemann-rx6jb]

me: reads the title: „YES! YES! GO PIGS GO!

[@LordNezghul]

In what programming language do you create your simulation?

[@EightLittleBears]

C# (using the Unity game engine)

[@tenma8797]

I think darker and lighter shades of “fur” would be interesting factor to put in for the animals. Regardless good video!

[@footballgameanimations]

what program do you use?

[@navidyaghmaei]

from where do u get the information about the biologio of the creatures and evolution itself. i am programming a simulation myself i have problems knowing how i should implement some feature like food digestion, how tempretures affect creatures and some other stuff.

[@EightLittleBears]

So the first thing is that I make it clear I am not a biologist, my simulation is an ever-improving project, and I am learning too 😊 But I am pretty good with research. What I tend to do is start with a concept that I see on KZbin (e.g. from Lindsay Nikole or Crash Course or some other educational content with at least reasonable credentials) or just read about it on Wikipedia - the goal here is to get basic familiarity. Once I have a rough understanding of the topic from those sources I seek out papers from more official sources to double check that I haven’t picked up any bad info - e.g. www.ncbi.nlm.nih.gov/pmc/articles/PMC6320072/ As a note - if you scroll to the sources section on Wikipedia, it can sometimes lead you to reputable sources as well! It also depends on the topic - some concepts are pretty simple and (like in this one) if I am only implementing a super basic level which will be built on later then I don’t need an advanced understanding. Best of luck with your sim and I hope that helps!

[@navidyaghmaei]

@@EightLittleBears wow. Thanks u veeerry much (and here is a idea u could start creating seasons where each season starts after x turns). Good luck with ur sim too.

[@FloridaMeng]

Nice visual calculator

[@EightLittleBears]

Dunno if this is a compliment but I like it either way 😂

[@Cerafem]

Do you have anywhere that explains the methodology, why those are the main characteristics simulated, how you have those affect each other, other approximations, etc?

[@Cerafem]

Scaling of parameters has a huge effect on the results, especially when you want to perform an analysis of the sim. Prey switching, food webs instead of food pyramids, whatever else. I've conceptualized all kinds of ways to go about similar things and would like to have an idea of what all is happening here, including phenomena you aren't interested in simulating.

[@Cerafem]

Side note: I do have a CS degree, no natural science background but I have spent a lot of time in topics like farming/biology/ecology/hydrology/etc, and have had opportunities to observe and affect some of these things around my rural home.

[@EightLittleBears]

I don’t have a methodology document if that’s what you mean. That said, a lot of people have asked me to go into more detail so I am planning on doing that in my upcoming videos - suppose they will become a sort of methodology! I have a couple more videos planned but will be working on more in depth reporting and data analysis in the next couple of months so hopefully that will provide some of the insights you are looking for! Thanks for the comments! Very interesting concepts!

[@Cerafem]

@@EightLittleBears yeah, any sort of breakdown that is something others can build on. I'm sure most of your audience aren't developers but plenty of people like looking under the hood. I assume this is mostly a cellular automata type simulation, look at neighboring cells and make a decision, consume and deposit biomass. I watched some of your other videos too, seems like at least some of the characteristics are opposite and have tradeoffs, but a good overview of the sim loop, actor properties, variable scaling, etc would be cool. I saw others suggest lotka-volterra, which is more of a continuous interpretation and yours is discrete. The big take away is predators population trails a quarter cycle behind their prey, growing at a rate proportional to the population of the prey. So when prey is at its highest, predator rate of increase is highest. When predators are at their highest, prey's rate of increase is negative. It's a result you'd expect to see over time in your simulation, as a sort of aggregate result. Might be worth tracking the rate of change in populations instead of just the populations themselves, usually in sims like this that is a really noisy value so I'd recommend a gaussian smoothing or something. Other discrete sims have shown that they do adhere to those equations, when performed at sufficient scale. In fluid sims, the lattice gas equation is a nice example of how simple sims can still show complex phenomena dictated by some nasty calculus. A really cool use of your sim would be to forcibly change some things and watch the effects, maybe even save the state so you can see how far it diverges. I'd love to see a population isolated on an island and see how different it becomes from the mainlanders.

[@Cerafem]

@@EightLittleBears the absolute nightmare tier version would be something accounting for... - trait changes necessary, like teeth and claws necessary to chew certain foods or climb to reach prey - detail on vegetation, allowing for different hiding strategies - barriers to switching prey quickly. A cow and a lion have huge differences in their digestion, both in design and microbes necessary. - overconsumption of vegetation leading to erosion Maybe you could have a laundry list of ideas to skim over and pick features you are interested in as you go?

[@ottosjachrani7821]

YESSSS

[@Tavg123]

👍

[@revimfadli4666]

Okay so did you, Pezzza, MakingFromScratch, and Simulife Hub conspire to launch evosim videos consecutively? :)

[@EightLittleBears]

Hahah no but I notice the exact same thing…

[@danieltabin6470]

What makes a creature a pig vs a hog? Is it a predetermined thing, or does it just get selected based on their "strategy" (size, affinity to meat, etc)

[@EightLittleBears]

The latter - so basically I have a csv with all the creature types and stat values for each, and when a new mutation happens it looks at that csv and find the closest match.

[@danieltabin6470]

@@EightLittleBears makes sense, thanks for the response

[@supercables251]

Source code?

[@erinkarp]

So are the species based on the stats, or are the stats based on the species?

[@EightLittleBears]

The stats are the base and then these run through a list of known species, compare all the stats, and pick the closest one. So when you see lots of the same animals what it means is essentially that lots of different game species are converging on what are the ideal stats at a given point in time.

[@erinkarp]

@@EightLittleBearsOh cool! I was hoping that was the case

[@asceticwolf]

I am interested in none of them developing a defense such as poisons or venoms.

[@peaceling9826]

❤

[@qoombert]

what happened: they evolved

[@TheQuyman]

Oh so this is the game that tierzoo plays

[@EightLittleBears]

Haha - I actually reached out to TierZoo about doing a series called “Testing TierZoo”, where I basically just test out things like his “which is the best bear” video. Didn’t get a response (understandably ofc given he probably gets flooded with messages all the time). But… which is really the best bear??

[@TitanLordofPizza]

Nifty

[@ClockworkGFX]

Giant Pygmy Hog is best boy

[@dukenussbaum3305]

The algorithm demands comments and I shall provide

[@EightLittleBears]

Thanks!

[@lexibyday9504]

There's so many lovely evolution games that are based on similar sytems but the creatures always feel like jokes. I think I'd apreciate a version set on earth or a human colony planet so that we can at least start with earth animals. The changes can then be based on morphing the base animals. Even tens of millions of years after dinosaurs, their chicken descendants still resemble them. But maybe there would have to be a limit using this method. After 100 milion years game time aliens come and wipe everything out?

[@EightLittleBears]

Sounds cool but I think I’m pretty far off that haha

[@redecterthomsonshadowascen5177]

These videos are quite short You could definitely make them longer by adding in more detail of what you’re doing.

[@EightLittleBears]

Hmm interesting, I did wonder what the optimum length might be and thought most people would like them less than 10 mins, but maybe not

[@redecterthomsonshadowascen5177]

@@EightLittleBears For what I'm assuming is your target audience. Probably 20 minutes

[@Mo_lester123]

how long is a turn

[@EightLittleBears]

They are essentially just a consistent mechanism for change rather than being an exact representation of time. At their starting values creatures die of old age at 1000 turns and take somewhere between 100 and 200 turns (can’t remember after the updates) to die of starvation and half the time to die of thirst, but all these numbers vary as creatures evolve!

[@Mo_lester123]

@@EightLittleBears ok

[@aleksszukovskis2074]

Make more

[@EightLittleBears]

Ha! The majority of time is spent actually building the sim, so that’s my bottleneck. That said, I still have the new diet system to talk through so I seas thinking I’ll probably make another video using this same build (or with only small updates), so that should hopefully come a bit sooner!

[@habarvaz3142]

THERE ARE NO LIMITS 😈😈😈

[@EightLittleBears]

🤯

[@miklemon2357]

Make the map 100X

[@EmperorPrinc3]

You hold the power of god in your hands

[@blackgold2589]

Comment for algorithm

[@chilldogs1881]

W

[@literallyagalaxy7789]

is it possible that freddy fazbear evolves

[@EightLittleBears]

Lol. Maybe..

[@qoombert]

1 hour ago, 12 comments

[@qoombert]

49 comments (50 now), 14 hours

[@qoombert]

7 days ago, 90 (91) comments

[@qoombert]

8 days ago 93 (94 comments)

[@Subcode]

Predesigning animals and calling them deer and pigmee hogs seems to defeat the purpose of simulation.

[@EightLittleBears]

The animals are not predesigned as the code for each is free to change and evolve in the whatever way suits the environment. The the avatars just indicate that the code for a given creature is close to a particular build. At the moment we are a bit limited because I have only drawn 29 avatars, but over time the plan is to add hundreds, so the visual representation of stats will get better and better.

[@handledav]

no

[@hotbit7327]

Accelerate your voice recording even more, I will understand even less.

[@EightLittleBears]

Does it sound accelerated?

[@hotbit7327]

@@EightLittleBears Is it not accelerated? You must know. Maybe you just speak very fast. I have no problem to follow Andrew Huberman, Brian Cox and many others. I was interested what is hard coded, what exactly do learn/decide those little brains you have coded. But I give up. Too many words per minute to listen with deeper understanding.

[@EightLittleBears]

Huh interesting, I thought I was speaking quite slowly but I guess it always sounds different to other people. Thanks for the notes!