I was skeptical of the bibites ability to develop behaviors more complicated than simply going after food - so the advancer herd dynamics and conditional behavior in this video really shocked me! Best of luck developing this - its some of the most advanced artificial life I have seen :)
Thanks a lot :) Yeah, I'm always impressed by what behaviors pops-up some times! Especially that kind of parenting behavior I've shown, can't wait to develop this further
@@TheBibitesDigitalLife Yea that parenting behavior - mind blowing. Super cool to see some interaction beyond predator-prey between bibites. Best of luck developing!
Its weird to me. I was running a simulation on 12 times speed. After like 24 of in game time almost nothing evolved, it was still just bibites going forward randomly. how long does it take for you to have your bibites evolve?
Depends a lot in the simulation settings you chose. Different settings will result in different adaptations. Some settings combination even make it too hard for life to develop so you're stuck with generations 0. If you just want to see a population stick you can use the "Helpful Initial Connections" setting to 2 or more, they should have an easier time sticking around
They’re handfed the ability to see, meanwhile you have the capability of interpreting and figure out arrangements of light without directly being told that it’s an object. You’re smarter.
I think it would be amazing if the pellets were also alive, as plants. They wouldn't have the ability to move, but would be able to spread seeds away. Some parameters for these plants can be an ideal tempeture for spreading (may be a way to implement biomes, with variable tempetures), spreading distance, or toxicity, which would make some species more dangerous than others. You can take the ideas that you think are good, if there are some. I think this project is incredible !
You should give them the ability to build dens where they store food! Would be interesting to maybe see them form colonies of multiple bibites sharing dens where they cooperate to search for food. An interesting behavior could emerge if they left a pheromone train when searching for food and following it back when they want to return home. Maybe they could leave a separate trail for others from their colony to help guide them to larger batches of food.
this is really cool and very impressive, i've been following evolution simulators on youtube for a while and this is possibly the most interesting one i've come across so far
I am so happy to finally be able to catch a stream! I have been working on a similar project inspired by the Bibites for a long while now. You have been a big inspiration for me throughout my undergrad! :)
Suggestion: add a graph that shows the population/infected population, and how sick someone is, and/or a family tree that you can open, I think that would be super cool
i've never heard of something like this before, and the concept of behaviors you can't predict naturally occurring (IE the babies following the parents) very unique
Hey, I love what you’re doing with the life simulation! You can for sure be more confident in the project :) I’m gonna support you in this. By the way, I believe that a simple environment simulation would be a good addition earlier on. Since evolution is a process of adaptation to survive and reproduce in the environment, that seems to have a lot of potential for the complexity that will evolve. That would be very fun! As well as more realistic Take care!
This is fascinating. The diminished fitness of herding behavior in the presence of viruses is quite relevant to the real world at the moment unfortunately. Instant subscribe. I am going to watch the rest of the series!
Thanks a lot 😁 Your words mean a lot. You'll see that my precious videos diminish in quality pretty quickly tho 😅 Hopefully I can start reaching at least the quality of this one from now on
Hi. I only discovered this project a few days ago, so this comment is late, but hopefully it's helpful. I don't thin explicitly implementing herding behavior as shown here is in keeping with the philosophy of the project as a whole. If herding is a beneficial behavior, then it should arise naturally through the structure of their network. You shouldn't have to have a specific "herding" output that is connected to code. It should be sufficient only to give the bibits the sensory inputs necessary to drive the herding rules, but the rules themselves should arise naturally as the emergent behavior of the neural networks. You shouldn't need extra code for it.
I just started making a project like this and thought I was doing something special... yours is so cool hahaa I hope one day to have a complex system like yours. Thanks so much for sharing so glad youtube made your original video pop up in my feed. Great work!
I've been wanting to do a project like this for such a long time, but never knew how to approach it!! Seeing your progress is very impressive and inspiring! Can't wait to see more!
I really enjoyed the "examples" segment. You should consider making a nature documentary style video where you explore the behaviour of specific species (after running the simulation for a while).
They're moving in herds...they do move in herds! Btw glad to see you're doing what you really enjoy in life. But don't ever feel like you can't do a part time job, or some sort of supplementary income. It'll take a while for this project to support you but that's ok. Can't wait to see what comes next!
This is a really awesome advancement. The hard coded behavior helps see and study interesting behaviors without waiting until (and if at all) they emerge on their own as would be the case if you made inputs and outputs more fundamental. I can't support financially right now, but I hope others will because this is pretty awesome.
I wood like to see you adding plant evolution and digestive evolution, and stuff like that For example Plants can have different vitamins And bibits would need difrent digestive systems in order to eat them
that'd be super neat! if i may add on to it, it could be interesting if some plants could evolve to be toxic, and some bibites could then evolve to tolerate or even repurpose that toxin for their own uses
@@mozarteanchaosinteresting ideea, If I can add some more to it : carnivore bibits would evolve to tolerate the poison too, and if a bibit evolves to tolerate poison and use it for himself, non poisonous food will not give them alot of nutrients , since they are getting used to the poisonous food, and they will lose there poison if they won't eat poisonous food for a while
New here - don’t know if it’s something you’ve considered already, but one of the driving forces behind evolution on earth is that between the eaten and the eater - plants evolving in responses to being eaten by animals, hosts evolving in response to their parasites, and prey evolving in response to their predators, and vice verca. I wonder if there were some behaviours or properties that plants could select for that would sythesise well with the Bibites? Maybe some plants would evolve in a manner to prioritise being eaten and spread, maybe some would prioritise never bring eaten at all. Then maybe some Bibites would specialise to pursue those more difficult ones and occupy a niche with little competition, or stick with the easy ones but actually end up adapting to best spread undigested seeds or similar. Now that’s an awful lot of complexity, but I’m sure you’d be able to distill out a couple of interesting parameters that could be played with
I am so in love with your work and also so Inspiring work... it's like the thing I always wanted to develope myself. truly great job and keep it going. I hope I can support this project in the future as soon as I am able to.
Thanks a lot 🥰 I would en outage you to still give it a go and try your hand at coding an Artificial Life simulation! It's extremely fun and rewarding 😁
I left this guys over night and at generatios 900 they reached a point when everyone is producing hormones that keep everyone else far and each one is farming pellets in their own zones
Hey, this was really cool! I'm happy to see your videos again. Are you considering going to the ALIFE conference again this year? I think people would be very excited about your idea of connecting the neuro-evolution to the herding parameters/behavior.
Do you know if the submission process is still open? And if yes until when? I thought I would maybe skip this one and present something amazing next year 😯
J'adore ton projet, Bravo! Pour augmenter le gameplay, on pourrait ajouter le pouvoir de contrôler les bibites. Durant qu'on le contrôle, la bibite apprendrait de notre comportement pour évoluer son cerveau. Je ne connais pas la complexité de ce que je propose, mais ce serait cool à implenter. :-)
I don't know if you're already doing this, but a good way to grow a channel is to put your shorts on tik tok. my friend gets the majority of his viewers from those tik tok videos, so it could help.
Viruses in reality, have so much more potential than people think they do. For instance, one could have most viruses be completely ineffectual for the bibites, while having a few bad ones, and a few viruses with surprisingly beneficial mutations.
I suggest that instead of the viruses just simy hurting the Bibites, they should instead be able to change their neural structures, be it for better or for worse. This would actually make it more accurate to real life, as there have been many cases of animals and viruses forming simbiotic relationships. The mechanic of viruses hurting Bibites wouldn't be lost either, as changes a virus produces in an individual that are beneficial to a species may actually hurt other species. This is why viruses even injure animals irl in the first place, viruses jump species and act in ways they weren't supposed to, so changing it would also have the added benefit of making it more realistic. I see however that you already have plans laid out for the future of the project, so I understand that changing it now may be an issue.
I don't know if you read ideas here or not, but what if they could trophallaxis. With herding, they could actually develop multi-bibite organisms, with them even having the possibility to have each part evolve differently! This could lead to so many differences, and if they developed this behavior in scarce food worlds just a theory, a few of them might develop to have themselves more productive is trophallaxis. And in a world with many predators, they could develop a shell even! This is just an idea, but it could lead to so many cool things.
je sais que je suis en retard pour le party, je ne fais qu'alimenter le monstre qu'est l'algorithme youtube. j'adore ton concept, continue comme ca,c'est hot en asti!
1 weeks ago, pro tip: be human when talking about those things........ the subscribing things. Just don't make a big thing out of it and don't bombard with positive (whatever the descriptive words in English are called) just say it naturally without going too far. Just, like: this is a big project and I really want to continue expanding it, but if nobody knows it's kind of a problem. You can subscribe if you want to and it will help the whole thing a bit. Also, if you have any ideas, please do share them in the comments because I want to account for things. So that was the video. If you want to, you can subscribe and all the other things. I will read and even try to answer some comments. That's it. The end of the video.
Eventually it's my plan, but for now I want to have fun with it a little more and experience the full development of the project. Open-sourcing it would mean becoming a manager more than a programmer and I'm not ready for that 😅
@@TheBibitesDigitalLife bibites 0.3 for linux do not work on my ubuntu 20.04 / nvidia, giving only grey screen (after splash screen) and nothing else. 00:40:47 j0 wormball@wormball-desktop:/media/3f16c67b-5e31-48bf-85c2-65bfd35e49f3/250gb/dima/alife/bibites$ ./The\ Bibites.x86_64 Set current directory to /media/3f16c67b-5e31-48bf-85c2-65bfd35e49f3/250gb/dima/alife/bibites Found path: /media/3f16c67b-5e31-48bf-85c2-65bfd35e49f3/250gb/dima/alife/bibites/The Bibites.x86_64 00:41:29 j0 wormball@wormball-desktop:/media/3f16c67b-5e31-48bf-85c2-65bfd35e49f3/250gb/dima/alife/bibites$ ./LinuxPlayer_s.debug Set current directory to /media/3f16c67b-5e31-48bf-85c2-65bfd35e49f3/250gb/dima/alife/bibites Found path: /media/3f16c67b-5e31-48bf-85c2-65bfd35e49f3/250gb/dima/alife/bibites/LinuxPlayer_s.debug no boot config - using default values
Hi again! Downloaded the game yesterday and ran it for about 50 in time hours, and I noticed that the bibites always let themselves have the virus and just deal with it, is there any way to change that so they actually will fight back the virus instead of just dealing with it?
Playing around with virus settings in the menu might impact the balance of the simulation and lead to different strategies! I encourage you to experiment with those 😁
I wonder how the behaviour would change if they could distinguish between bibits while calculating the herding vectors? Like perhaps distinguishing between smaller or bigger bibits and following the biggest, or shunning the infected, or all sorts of behaviours.
Cant wait until my eternally running bibites simulation evolves a society that is aware of higher beings Also cant wait until they develop into colony-based superorganisms Also also cant wait until the basic food source is more than just pellets that appear out of nowhere.
It's exciting to see Unity's DOTS migration on your roadmap. I've struggled to work in DOTS for 2D. Have you seen something that makes you think this is ready for your project?
Not really, and their development seems kind of rocky. Among other things they say that you have to stay on 2020.3.x for the coming year if you want to use DOTS. But in the end I already started playing around with it and I figure that some parts can easily be converted to DOTS gradually. That somewhat reduces the risk. But I don't think DOTS would be at the right point now to develop a full game from scratch
Can you try letting them evolve sexual reproduction? I think this would amount roughly to: - use a bunch of evolvable numbers (say, a vector of 10 dimensions) between -1 and 1 to determine compatibility - compatibility is determined by summing two individuals' vectors. If they sum to all 0s, they are perfectly compatible. Otherwise, a viable offspring only is generated with a certain chance. You can probably mess with the tolerances though: For instance, if they get to within 0.01 across all features the chance might still be 100% and if any of the features are out of phase by more than 0.5 or something, the chance is 0% And in between those values, you just interpolate chances. That alone could already be interesting, but you could also make that interact with pheromones or perception or something so they can learn to find other Bibites compatible with them so they aren't forced to just fruitlessly try when it won't actually come to anything. Alternatively, instead of summing to 0, you could also try using cosine similarity (so they have to point in the same direction for perfect compatibility) or even both. This should also potentially give rise to speciation, and it'd add more complexity to the herding behavior as they might potentially be tempted to stick close to known compatible partners, say.
Hey! It's me again, with another question hehe At 4:28 you said herding seems to me quite popular since it appeard in half of the simulations. That made me wonder: what do you mean my "appearing"? You programmed these herding rules, but what defines whether a bibite will follow them or not?
@@TheBibitesDigitalLife But how do you make an evolution of characteristics? I mean, I get evolving speed, size, etc since they are continuous values that can be slightly increased or decreased by chance in every generation. But how do you apply that do things that aren't numerical, like following these rules? The only way I can think of is all of them starting with a value of 0 for each of these rules and then (maybe) evolving to increase these values until they're fine-tuned to behave in a way that increase their survivability. How far am I? Hahah
@@TheBibitesDigitalLife Hahahah interesting! Do you have any recommendations for me to learn more about how this algorithm works? I appologize for asking a lot, is just that I'm really interested in this subject!
4:57 DUCK Bibits should be able to milk and communicate in ways other than feromones (how do they herd?), like via their color Also, how about change feromones from particles to giant grid(texture), where feromones just increase the rgb values of it? Or, if you need more colors, r32g32b32a32, and using bit shifts pack them... Also, wind.
@@connorhaley3190 Alpha. Though really, they only stand for the colors/transparency when in a very specific texture that is sent to the screen. Otherwise their meanings can be anything the developer desires.
I was skeptical of the bibites ability to develop behaviors more complicated than simply going after food - so the advancer herd dynamics and conditional behavior in this video really shocked me! Best of luck developing this - its some of the most advanced artificial life I have seen :)
Thanks a lot :)
Yeah, I'm always impressed by what behaviors pops-up some times!
Especially that kind of parenting behavior I've shown, can't wait to develop this further
@@TheBibitesDigitalLife Yea that parenting behavior - mind blowing. Super cool to see some interaction beyond predator-prey between bibites. Best of luck developing!
Its weird to me. I was running a simulation on 12 times speed. After like 24 of in game time almost nothing evolved, it was still just bibites going forward randomly. how long does it take for you to have your bibites evolve?
Depends a lot in the simulation settings you chose. Different settings will result in different adaptations. Some settings combination even make it too hard for life to develop so you're stuck with generations 0.
If you just want to see a population stick you can use the "Helpful Initial Connections" setting to 2 or more, they should have an easier time sticking around
@@Finnie1203 For me, my bibites are always competently getting food around the 50th generation.But that's just on default settings.
I have a suspicion that this project is gonna be super well-known in like a year.
Thanks 😁
I sure hope so
pshhh dont even need a year
That's completely right. I had interest in this kind of proyects and I don't know how I didn't discover this awesome project before!
This comment has aged very well :)
1 year milestone
The herding behaviour looks amazing! So many possibilities!
Aye
And its probably more advanced then most virtual creatures
I'm your 69th like
The infection pheromone evolution was super cool!
can’t wait for bibites to be smarter than me, not like that’s a huge advancement but still
69 likes lol..... I liked this comment like 2 weeks ago and here we are now
They can't write a comment yet so I guess you can feel superior right now
@@w花b 'right now'
They’re handfed the ability to see, meanwhile you have the capability of interpreting and figure out arrangements of light without directly being told that it’s an object. You’re smarter.
Its impossible to see that happen on this engine and this amount of brains put into the subject .
I think it would be amazing if the pellets were also alive, as plants. They wouldn't have the ability to move, but would be able to spread seeds away. Some parameters for these plants can be an ideal tempeture for spreading (may be a way to implement biomes, with variable tempetures), spreading distance, or toxicity, which would make some species more dangerous than others. You can take the ideas that you think are good, if there are some.
I think this project is incredible !
I think plant pellets are more similar to algae
You should give them the ability to build dens where they store food! Would be interesting to maybe see them form colonies of multiple bibites sharing dens where they cooperate to search for food.
An interesting behavior could emerge if they left a pheromone train when searching for food and following it back when they want to return home. Maybe they could leave a separate trail for others from their colony to help guide them to larger batches of food.
Good ideas all around!
Thats probably the most intresting life simulation project i v ever seen, gg
this is really cool and very impressive, i've been following evolution simulators on youtube for a while and this is possibly the most interesting one i've come across so far
Thanks! Means a lot 😁
Hope it'll be enough to get you subscribed 😏
I am so happy to finally be able to catch a stream! I have been working on a similar project inspired by the Bibites for a long while now. You have been a big inspiration for me throughout my undergrad! :)
Same here. However not for a long while.
Suggestion: add a graph that shows the population/infected population, and how sick someone is, and/or a family tree that you can open, I think that would be super cool
Most of those are planned at some point !
i've never heard of something like this before, and the concept of behaviors you can't predict naturally occurring (IE the babies following the parents) very unique
Honestly if you had to stop for a year I’d be happy to wait. The bibites are awesome.
I like that moment with bibbits rotating. Looks like a disco party!
Hey, I love what you’re doing with the life simulation! You can for sure be more confident in the project :) I’m gonna support you in this.
By the way, I believe that a simple environment simulation would be a good addition earlier on. Since evolution is a process of adaptation to survive and reproduce in the environment, that seems to have a lot of potential for the complexity that will evolve. That would be very fun! As well as more realistic
Take care!
AHH yes, i've waited so much for this video ! I hope you know how much means to some of us :D
Thanks a lot 😁
I put in a lot of efforts too, I hope I'll meet expectations
Evolutionary BOIDS is super interesting. I'm surprised this is the first time I see somebody attempt this. Makes a lot of sense!
This project is truly amazing.
Its great to see the Bibites still evolving and I wish you and the channel the best. Looking forward to what comes next for the Bibites.
How does this comment have zero likes
Digestion
@@nathanwi1147 I liked this comment just now.
This is fascinating. The diminished fitness of herding behavior in the presence of viruses is quite relevant to the real world at the moment unfortunately. Instant subscribe. I am going to watch the rest of the series!
Thanks a lot 😁
Your words mean a lot.
You'll see that my precious videos diminish in quality pretty quickly tho 😅
Hopefully I can start reaching at least the quality of this one from now on
This is one of the coolest projects I’ve ever seen.
Thanks a lot 😁
I hope to make it even cooler
@@TheBibitesDigitalLife you sure did, and you’ll surely will.
Cool ideas: day-night cycle, weather, temperature. Love the project, keep up with the incredible work!
Hi. I only discovered this project a few days ago, so this comment is late, but hopefully it's helpful.
I don't thin explicitly implementing herding behavior as shown here is in keeping with the philosophy of the project as a whole. If herding is a beneficial behavior, then it should arise naturally through the structure of their network. You shouldn't have to have a specific "herding" output that is connected to code. It should be sufficient only to give the bibits the sensory inputs necessary to drive the herding rules, but the rules themselves should arise naturally as the emergent behavior of the neural networks. You shouldn't need extra code for it.
I just started making a project like this and thought I was doing something special... yours is so cool hahaa I hope one day to have a complex system like yours. Thanks so much for sharing so glad youtube made your original video pop up in my feed. Great work!
I've been wanting to do a project like this for such a long time, but never knew how to approach it!! Seeing your progress is very impressive and inspiring! Can't wait to see more!
I really enjoyed the "examples" segment. You should consider making a nature documentary style video where you explore the behaviour of specific species (after running the simulation for a while).
They're moving in herds...they do move in herds!
Btw glad to see you're doing what you really enjoy in life. But don't ever feel like you can't do a part time job, or some sort of supplementary income. It'll take a while for this project to support you but that's ok. Can't wait to see what comes next!
Thanks a lot 😁
And yeah absolutely, I'll see how it goes
This is really cool! I've always been amazed by these types of video's and you nail it, keep it up!
This is a really awesome advancement. The hard coded behavior helps see and study interesting behaviors without waiting until (and if at all) they emerge on their own as would be the case if you made inputs and outputs more fundamental. I can't support financially right now, but I hope others will because this is pretty awesome.
...do you have a black eye? :o
No black eyes, just very big dark circles 😅
If one day, this is some amazing extremely complex AI system, I'd gladly pay for it on steam!
I wood like to see you adding plant evolution and digestive evolution, and stuff like that
For example
Plants can have different vitamins
And bibits would need difrent digestive systems in order to eat them
that'd be super neat! if i may add on to it, it could be interesting if some plants could evolve to be toxic, and some bibites could then evolve to tolerate or even repurpose that toxin for their own uses
@@mozarteanchaosinteresting ideea,
If I can add some more to it :
carnivore bibits would evolve to tolerate the poison too, and if a bibit evolves to tolerate poison and use it for himself, non poisonous food will not give them alot of nutrients , since they are getting used to the poisonous food,
and they will lose there poison if they won't eat poisonous food for a while
@@luciangg1553 oh, very cool ideas o:
i don't have a lot to add to that one, but it's definitely neat
Looking forward to seeing this project evolve!
All the best, I really like this project and I'd like to see it continue.
You might want to consider starting a Discord server as a hub
There is a reddit
He's got a discord server available to patrons as well.
Yey! More bibbits!
New here - don’t know if it’s something you’ve considered already, but one of the driving forces behind evolution on earth is that between the eaten and the eater - plants evolving in responses to being eaten by animals, hosts evolving in response to their parasites, and prey evolving in response to their predators, and vice verca. I wonder if there were some behaviours or properties that plants could select for that would sythesise well with the Bibites? Maybe some plants would evolve in a manner to prioritise being eaten and spread, maybe some would prioritise never bring eaten at all. Then maybe some Bibites would specialise to pursue those more difficult ones and occupy a niche with little competition, or stick with the easy ones but actually end up adapting to best spread undigested seeds or similar. Now that’s an awful lot of complexity, but I’m sure you’d be able to distill out a couple of interesting parameters that could be played with
i feel like the mindustry soundtrack would fit in perfectly with the bibites simulation
Love seeing this project evolve like the bibites !
I would like to see how these react in different environments. When I get home and have some spare time I'll look into helping support this project.
the amount of outfit changes is amazing
Thanks a lot, I am a true artist 🦸
absolutly amazing. i wanted to program something like this myself for years! ANd now it's here
I would encourage you to try it yourself as well 😁!
It's an extremely rewarding process to bring your own lifeforms to life
You must continue! Its great!
Ok so if this man is quitting his job in order to nurture this project i think we should support him.
6:10 hey!, that's actually really useful!, they're saying "Hey, i'm infected, be careful" and the other ones just go away
I'm so excited, this made my day
I just wanted to comment to help with the algorythm and say have a good day! :D
- The comment for support -
training gets you the set of genes that are required to survive in your conditions
It would be nice to see them evolve in 3D and reacting to more complex environmental constraints!
Sounds nice, but what about the theming?
The bibites shall rise
Wishing you the best of luck!
I am so in love with your work and also so Inspiring work... it's like the thing I always wanted to develope myself. truly great job and keep it going. I hope I can support this project in the future as soon as I am able to.
Thanks a lot 🥰
I would en outage you to still give it a go and try your hand at coding an Artificial Life simulation! It's extremely fun and rewarding 😁
I left this guys over night and at generatios 900 they reached a point when everyone is producing hormones that keep everyone else far and each one is farming pellets in their own zones
Hey, this was really cool! I'm happy to see your videos again.
Are you considering going to the ALIFE conference again this year? I think people would be very excited about your idea of connecting the neuro-evolution to the herding parameters/behavior.
Do you know if the submission process is still open? And if yes until when?
I thought I would maybe skip this one and present something amazing next year 😯
Comment for the algo after the utter decimation
Yeah TH-cam fucked up with this one
Nice, looking forward for the next video!
J'adore ton projet, Bravo! Pour augmenter le gameplay, on pourrait ajouter le pouvoir de contrôler les bibites. Durant qu'on le contrôle, la bibite apprendrait de notre comportement pour évoluer son cerveau. Je ne connais pas la complexité de ce que je propose, mais ce serait cool à implenter. :-)
Love your work man!
Thanks 😁!
I don't know if you're already doing this, but a good way to grow a channel is to put your shorts on tik tok. my friend gets the majority of his viewers from those tik tok videos, so it could help.
I'll do that!
Thanks for the idea :)
@@TheBibitesDigitalLife also a twitch stream for coding sessions would be nice ;)
I did that on my discord some times. But I figure its pretty boring 😅
Mostly me googling stuff ahahah
Viruses in reality, have so much more potential than people think they do. For instance, one could have most viruses be completely ineffectual for the bibites, while having a few bad ones, and a few viruses with surprisingly beneficial mutations.
The youngs following their parents is so cool!
Agreed 😁!
Hype hype hype
triple hype
Hype hype hype 2.0
Hype hype hype 3.0
I suggest that instead of the viruses just simy hurting the Bibites, they should instead be able to change their neural structures, be it for better or for worse.
This would actually make it more accurate to real life, as there have been many cases of animals and viruses forming simbiotic relationships.
The mechanic of viruses hurting Bibites wouldn't be lost either, as changes a virus produces in an individual that are beneficial to a species may actually hurt other species.
This is why viruses even injure animals irl in the first place, viruses jump species and act in ways they weren't supposed to, so changing it would also have the added benefit of making it more realistic.
I see however that you already have plans laid out for the future of the project, so I understand that changing it now may be an issue.
what a cool bibite thing cool! I wish there were more bibite thing.
There's at least going to be more in the future than so far in the past 😁
I don't know if you read ideas here or not, but what if they could trophallaxis. With herding, they could actually develop multi-bibite organisms, with them even having the possibility to have each part evolve differently! This could lead to so many differences, and if they developed this behavior in scarce food worlds just a theory, a few of them might develop to have themselves more productive is trophallaxis. And in a world with many predators, they could develop a shell even! This is just an idea, but it could lead to so many cool things.
It’s a very cool project, man 🤞
I love this!
Thanks again for the tunes 😁❤️
It's funny how 2 of the most controversial features were added in the same update
I think it's super cool they developed 'parenting' behaviour (though in this case it is the child which initiates the behaviour)
Keep up the work.
je sais que je suis en retard pour le party, je ne fais qu'alimenter le monstre qu'est l'algorithme youtube. j'adore ton concept, continue comme ca,c'est hot en asti!
1 weeks ago, pro tip: be human when talking about those things........ the subscribing things. Just don't make a big thing out of it and don't bombard with positive (whatever the descriptive words in English are called) just say it naturally without going too far. Just, like: this is a big project and I really want to continue expanding it, but if nobody knows it's kind of a problem. You can subscribe if you want to and it will help the whole thing a bit. Also, if you have any ideas, please do share them in the comments because I want to account for things. So that was the video. If you want to, you can subscribe and all the other things. I will read and even try to answer some comments. That's it. The end of the video.
Also try to make videos more of something that is complete on it's own. I know it's not 2016 but people watching older videos is still a thing.
Very cool ideas! It would be interesting to see some kind of queen (hard coded maybe?) to simulate hive mentality like ants or bees!
Let’s gooo 🍾🍾
3:19 "violent tendencies" Yeah I think that is an understatement here.
Also, seeing the herd tear itself apart made me lose my composure.
Good luck and thank you
How wild would it be if the Singularity accidentally came from the Bibites
I would feel bad ahah,
But hey, it would probably be better than if it came from Microsoft or some other company's AI 😅
looking great
What about adding water to give the bibbits something to drink?
Also add mountain , so the bibits can use the landscape to hunt herbivores and to hide from carnivores
Do you consider moving your project to open source? Maybe community will be able to help you in development.
Eventually it's my plan, but for now I want to have fun with it a little more and experience the full development of the project. Open-sourcing it would mean becoming a manager more than a programmer and I'm not ready for that 😅
@@TheBibitesDigitalLife bibites 0.3 for linux do not work on my ubuntu 20.04 / nvidia, giving only grey screen (after splash screen) and nothing else.
00:40:47 j0 wormball@wormball-desktop:/media/3f16c67b-5e31-48bf-85c2-65bfd35e49f3/250gb/dima/alife/bibites$ ./The\ Bibites.x86_64
Set current directory to /media/3f16c67b-5e31-48bf-85c2-65bfd35e49f3/250gb/dima/alife/bibites
Found path: /media/3f16c67b-5e31-48bf-85c2-65bfd35e49f3/250gb/dima/alife/bibites/The Bibites.x86_64
00:41:29 j0 wormball@wormball-desktop:/media/3f16c67b-5e31-48bf-85c2-65bfd35e49f3/250gb/dima/alife/bibites$ ./LinuxPlayer_s.debug
Set current directory to /media/3f16c67b-5e31-48bf-85c2-65bfd35e49f3/250gb/dima/alife/bibites
Found path: /media/3f16c67b-5e31-48bf-85c2-65bfd35e49f3/250gb/dima/alife/bibites/LinuxPlayer_s.debug
no boot config - using default values
Hope this projects stil inspires you, looking forward to the next update!
It really much does 😁
Still actively working on it!
Welcome back !
Thanks a lot, it's a pleasure !
I wish you the best of luck!
Thanks!
Hi again! Downloaded the game yesterday and ran it for about 50 in time hours, and I noticed that the bibites always let themselves have the virus and just deal with it, is there any way to change that so they actually will fight back the virus instead of just dealing with it?
Playing around with virus settings in the menu might impact the balance of the simulation and lead to different strategies!
I encourage you to experiment with those 😁
Awesome work!
I wonder how the behaviour would change if they could distinguish between bibits while calculating the herding vectors? Like perhaps distinguishing between smaller or bigger bibits and following the biggest, or shunning the infected, or all sorts of behaviours.
This channel is so cool!
Thanks!
Finally!!!
stunning....really. i downloaded it and wow.....cant wait for more. subed and shared
Thanks so much ☺️
Next video is coming with the next week
@@TheBibitesDigitalLife im gonna donate at some point when I can. This is clever stuff. Congrats!
Cant wait until my eternally running bibites simulation evolves a society that is aware of higher beings
Also cant wait until they develop into colony-based superorganisms
Also also cant wait until the basic food source is more than just pellets that appear out of nowhere.
It's exciting to see Unity's DOTS migration on your roadmap. I've struggled to work in DOTS for 2D. Have you seen something that makes you think this is ready for your project?
Not really, and their development seems kind of rocky. Among other things they say that you have to stay on 2020.3.x for the coming year if you want to use DOTS.
But in the end I already started playing around with it and I figure that some parts can easily be converted to DOTS gradually. That somewhat reduces the risk.
But I don't think DOTS would be at the right point now to develop a full game from scratch
Can you try letting them evolve sexual reproduction?
I think this would amount roughly to:
- use a bunch of evolvable numbers (say, a vector of 10 dimensions) between -1 and 1 to determine compatibility
- compatibility is determined by summing two individuals' vectors. If they sum to all 0s, they are perfectly compatible. Otherwise, a viable offspring only is generated with a certain chance. You can probably mess with the tolerances though:
For instance, if they get to within 0.01 across all features the chance might still be 100% and if any of the features are out of phase by more than 0.5 or something, the chance is 0%
And in between those values, you just interpolate chances.
That alone could already be interesting, but you could also make that interact with pheromones or perception or something so they can learn to find other Bibites compatible with them so they aren't forced to just fruitlessly try when it won't actually come to anything.
Alternatively, instead of summing to 0, you could also try using cosine similarity (so they have to point in the same direction for perfect compatibility) or even both.
This should also potentially give rise to speciation, and it'd add more complexity to the herding behavior as they might potentially be tempted to stick close to known compatible partners, say.
Yo, you spitting mad fact
Good luck my man
Hey! It's me again, with another question hehe
At 4:28 you said herding seems to me quite popular since it appeard in half of the simulations. That made me wonder: what do you mean my "appearing"? You programmed these herding rules, but what defines whether a bibite will follow them or not?
I just programed the systems and possibilities, it's up to them to evolve and use them or not.
@@TheBibitesDigitalLife But how do you make an evolution of characteristics? I mean, I get evolving speed, size, etc since they are continuous values that can be slightly increased or decreased by chance in every generation. But how do you apply that do things that aren't numerical, like following these rules?
The only way I can think of is all of them starting with a value of 0 for each of these rules and then (maybe) evolving to increase these values until they're fine-tuned to behave in a way that increase their survivability. How far am I? Hahah
Very far 😅 ahhah
That's what brain evolution algorithm are for! Things like NEAT and others 😁
@@TheBibitesDigitalLife Hahahah interesting! Do you have any recommendations for me to learn more about how this algorithm works? I appologize for asking a lot, is just that I'm really interested in this subject!
The algorithm in the bibites is custom, but the initial NEAT paper would be a good place to start!
nn.cs.utexas.edu/downloads/papers/stanley.ec02.pdf
I even got the classic green color. :)
At what point will we see this man get the attention he deserves, these need to be implemented in minecraft mobs and shit. Villagers.
4:57 DUCK
Bibits should be able to milk and communicate in ways other than feromones (how do they herd?), like via their color
Also, how about change feromones from particles to giant grid(texture), where feromones just increase the rgb values of it? Or, if you need more colors, r32g32b32a32, and using bit shifts pack them... Also, wind.
First of, currents not wind. Second, what does a stand for, I’m pretty sure b, g, & r=blue, green & red, but a?
@@connorhaley3190 Alpha. Though really, they only stand for the colors/transparency when in a very specific texture that is sent to the screen. Otherwise their meanings can be anything the developer desires.
Pure genius
Interesting...
For the algorithm