When the population of ants increases too much, some groups leave the main center and nest elsewhere. Over time, these ants differentiate and even become enemies with the original colony. It would be nice if you add this situation to your software.
@@All_Natural_Soda I think ants can swarm the same way bees swarm, so I think you're wrong about them being in the same colony forever. How would new colonies be created in the first place?
Some ideas that just came to my mind: - Procedural maps for the ants to explore - Different types of ants, (workers, fighters, queen etc.) - Different colonies somehow competing (for food etc.) - Collision of ants, so that there could be jams at busy junctions
I'm always impressed at just how effective random walking is at exploring spaces, and the random variation on following the return path causes optimization.
It would be a pretty bad screensaver, since parts of it are basically constantly lit up. Not that anyone uses CRT monitors anymore, but if you are to have a screensaver it should be effective.
@@davidgustavsson4000 Well some screensavers implemented by microsoft aren't really saving anything so they are probably here just to make some pretty animations :)
Yessssssssss I absolutely love these and this is crazy seeing another one came out 2 minutes ago! I love your work so much, thank you so much for posting such interesting content!!!!!
Mate, your work is AWESOME. Pleasepleasepleaseplease make either a really long video, or a playlist that is hundreds of videos long. Make it anything...I just want to throw it on in the background and let it run ❤❤❤ Beautiful to watch.
@@caseymcvay4727 Let also please implement RNG for ant love. Colony crossed lovers must face the wrath of their colonies and risk all their food for each other XD
And each colony having different behavior / brain / marker strength. Then let them fight for resources against each other. Slightly modifying brains each round. Genetic programming!
Love seeing these get more creative and I'm curious about when you start adding in the soldier ants and creating hazards that the colony would fight off
Gorgeous work once again! I think a nice change might be making the fog of war visualization transparent so that we can see what the map looks like even though the ants can't
Maybe next time instead of a map already made for them you can experiment with making them dig their own tunnels? It'd be cool to see how they form, especially if you add rules to how they dig and get a bunch of really crazy results. Maybe eventually we could get into different types of ants with their own rules and program some kind of nest behavior? There's a lot you can do here!
@@Daro-Wolfe exactly! It could be very interesting to simulate larvae placement and food storage, or even look at the queen's production of winged ants to create new colonies. Lots of possibilities 🙂
Heyy! This is starting to look less like an experiment (a fun one, nonetheless), and gradually much more like something that could easily be turned into a game. As a child, I once tried to do something even mildly akin to SimAnt. It was one of my first ventures into the world of C and C++. Of course, I but miserably failed, as I had bit more than I could chew, and abandoned the project, left to bitrot on GitHub. Nowadays, I might just be able to, and this video is exciting inspiration to do so! I'm working on a PR to add IPv6 support to Chocolate Doom, but once I'm done with that I'm excited to try a new thing like this. Thank you Pezza! :D
This is really cool! I have an idea for another new pheromone: "food too far" It gets released when an ant has found food but takes too long to come home, so it eats the food, restoring (some?) lifetime. It could be used to determine where an new colony should be started, if that's something you also would wanna implement. Alternatively the ant could just die and drop the food to be found by another ant.
I love this series! What if you coded in two colonies racing against each other? Maybe even fighting and killing each other? It'd be cool to see that as a long-term simulation, maybe with renewable food sources 🤔
Okay so now, create the maze procedurally, add food at random intervals in random places, make the maze infinitely scalable and run 1x speed. Best screen saver EVER. personal virtual ant farm.
Wow.. This just keeps getting better and better! Love watching it solve the maze! ;) I've actually been working on my own ant pheromone simulation, written in python.. And it's fun, and sorta works, but nowhere near as function as yours.. I think my rules and evap time need adjusting, as well as wall avoidance.. heh
Maybe add an incentive for ants to explore areas in the fog of war. Also you can mark explored areas with no food as non-interesting, that way ants won't waste time going to places they already been to. And lastly, maybe add other types of ants for specializations (like explorer, gatherer etc...). Really cool and fun to watch simulation!
There should be special areas which regularly spawns some food, because the logical conclusion of what happens to the colony in these simulations after food runs out is that they all die, so a food source should have a special color of itself so ants know to occasionally check it (say every few trips or so)
We're very similar. I have a ants build of a unity game I made many years ago on my PC. I actually kind of wish now that I did more with it. I was trying to make it 3D so ants could walk on walls. I got it working in 2d to the level you did on your first approach. This was about 5 years ago
Do foraging ants have any reaction to idle trail/home signals? because personally, if I was out foraging, an easy way I would try to find new places unvisited before would just be to head directly away from the greatest concentration of trail I could sense.
it would be cool if you get them to dig for food. Like a material that an ant can randomly decide to "eat" through, but it's not food. I wonder what kind of tunnels they would make.
Hey Pezza, how do you prevent the Antmill of death? when two ants cross each other and going the opposite direction. Why isnt one following the other? Since both output the pheromone the other is looking for? One going towards home and the other going to the food source.
hey pezza i downloaded the simulation from your github. how do you make the more complex wall shapes and larges batches of food? as far as i can see in the version on github only single pixels can be added. Thanks for the amazing simulation!!
