The release is fullscreen but seemingly only 1080p, which messes with my desktop since I have a 1440p monitor. Is it possible to have it run in a window, or switch to the native resolution? I'd rather not mess with the source code myself, as I've had little luck compiling C-related code for whatever reason.
I have an idea for a possible part 6: the ability for colonies to branch out and create more queens and therefore other colonies. This will lay the foundations of "colony reproduction", a totally other aspect to ants. I would also like to add the idea of giving the ability for corpses to be food, suggested by notus42 to give better reward for war
@@Zedryx69 in nature, this behavior is dependent on species. Most ants fight with child colonize, some ants form global colonize who rage war over several continents.
Project is really going great- this was a great way to implement the soldier ants. Shows that perfect balance required between gathering resources and destroying the nearby colonies. Seems like location plays a big factor.
Yeah, from what i saw, there are a few factors in play, they are: 1. How close are the nearest resource points? 2. How many are there? 3. Are there any bordering colonies? 4. Do they have access to more resource points than you do? 5. Are you close to the middle / have a clear path? There are probably more, but in my opinion these are quite important, if not the most important.
@@MostScaryCat Yea, I noticed that proximity to more resources than the closest colony is a very strong advantage. Didn't think about the clear path but looking again it does make a significant difference.
@@MostScaryCat I found that those who branched out and discovered new resource piles earlier tended to win, especially those that established strong connection lines by the time fighting begins
From my experience I see that there are 3 types of ants: workers, soldiers and explorers. I wonder how will it be if you add this third type, they should be faster and release a signal of higher priority(almost the only one) for the correct path.
There is just workers and soldiers, it’s just that in this simulation ants are able to follow a path but have different genes compared to everyone else, making them take unique paths that happen to make some ants go off-course and end up finding other resources or areas.
this in my opinion would reduce the main advantage of the ant algorithm which is resiliency. The nice thing about only one priority is that the evolution of the colony along with the environment dictates where they go. Prioirties sort of arise from their interaction
As much as I love the RGB color space, when choosing colors for teams, I highly recommend using the YUV color space instead. With a constant Y value, all colors will be the same brightness. In your current video, blue ants are much much harder to see than yellow ants. If you used the YUV then you would be able to easily choose different distinguishable colors while keeping all teams equally visible.
@@Bloodlyshiva A while back, I wrote something similar to this to run on the GPU, the best ways of implementing player control would be. 1. Instruction pheromone - allow the player to paint pheremone directly - allowing them to "soft direct" ants at will (to attack other colonies, get a certain resource etc) 2. Population control, allow the user to dynamically set ant production rates for new nests vs worker ants - kinda boom vs rush style 3. Tech trees, allow you to improve speed, attack strength, or other stats which costs foods that the ants collect
"thanks for wathing" 8:53 xD Amazing vid, it is incredible to see how surprising are the outputs are differents depending on the map and the number of enemy colonies. I wonder what are the perfect parameters for a colony to position itself well in relation to its environment
This would make for an amazing mobile sandbox game. Just a simple UI where you can tab to place different assets such as food, walls, or different ant spawns and the ability to speed up or slow down time would make for a really simple but fun game. You could even put it for a small price to make some profit! Edit: a game similar to what I described above is called "Robot Colony"
@@onmomentum the only game I can think of is "Pocket Ants". Played it for a while, quite fun. You actually play as an "ant commander", so it's quite different from this but quite the same I guess
Great video. How about: Dead ants from the same colony could give of a death hormone. Workers who encounter these, will bring the information back to the colony. The more often the colony receives these (in a short timespan?), the higher the change it'll breed a soldiers to combat the threat?
You can compare strategies: one colony will consistently produce fighters, which means less workers but be more prepared once you are attacked. The other colony will only produce fighters when needed.
Next Video: C++ ant simulation, ants take over earth Later: C++ ant simulation, ants rule universe Little More Later: C++ ant simulation, ants become god
I am loving every single video of yours and this is probably the best follow up video ever. To pile on with the suggestions: it seems like positioning is the major factor for ants to win or loose here. How about some variation in the ants stats and priorities? Probably based on some square law plus constant. You can have a little evolution algorithm and tournament, hopefully producing several types of ants to excel at different types of levels. Again, love your work. Please keep them coming. Cheers
unbelivelble what you can do with programming ... everytime its surprises me somehow ... soon i will be able aswell ... well not thos thingis but none the less
I assume there is not limit for your innovative way of thinking. Impressive work and remarkable results. Thank you for inspiring the great community. Keep fingers for your goals.
these are great! It seems the winner is usually the nest spawning with the nearest food blob. I interpret it like those nests increase populations the soonest and population increases search space and coverage density, so the nest gets to make better 'decisions' sooner. Also, and perhaps mostly, the exponential increase in pop
I would agree. The sooner an ant can bring its food back the sooner it can go to collect more. Having closer food means more throughput, which means faster population growth, which translates to more efficient food gathering from farther food sources. A closer starting food source snowballs quite quickly. I wonder how the ants would do with chokepoints, infinite food sources, and imbalanced maps. For instance, could a colony with only a single infinite food source outlast a colony with many finite food sources if there's a choke point to defend?
@@Zebo12345678 That's pretty interesting. I think it would also be interesting to see the simulation run against those scenarios, but also the 'latent space' between them. So if there is a 1v1 map with a central food source, and I wonder what happens if one nest was further away or if there was an obstacle in the way or if the food had different shapes and cavities... Those scenarios may be interesting but if we call the map a picture and take a geometric 'morph' between the map of the base scenario and a modified scenario and we simulate on each of the maps generated by each step in the morph, that could also yield interesting results. Perhaps to watch something like that (in a timely manner) the videos would need to be layered over each other with each iteration using a slight brightness or hue shift or something. I just think it would be cool to have an idea of why a scenario plays out a certain way and watch to see if it's possible to determine, as a scenario changes little by little and morphs into the alternate scenario, when a result should change and why. Like, if there is a 1v1 open map except for a wall with a tiny gap between competing nests, and one nest has a nearby food source and the other a more distant food source, then if the gap is small enough, it could be difficult to press the advantage of faster growth. Seeing what happens as the gap opens wider each iteration until the advantage could be pressed and trying to predict when such a thing could happen, would be interesting.
Would you ever consider publishing and selling this somehow? The idea of generating maps with food and terrain for ant colonies to expand and fight over seems a reasonably compelling one for an indie game. I've rewatched this like 5 times now, just looking at all aspects of how the ants behave and how they react to their environment and how heavily that determines their success against a now opposing force. Especially interesting was the interaction between food signals from different factions, like how if one side had food signals laid down, past or present, they would draw workers there, often turning the tide in nearby battles against the opposing mainly-soldier force. All in all, super cool simulation, I love your creativity and the simple readability of your sim.
At first I came across this, was like, "yes. I like this." And then binged all the ant stuff that came before. Imma now go look at the rest of your content
I see, the denser food in the middle acts as a decisive resource point. Whomever gain's control of it first has a massive advantage because it guarantees stable population growth, and once they have the larger population they can fight off any attempts from other colonies to take that food. In the three colony battle this allowed the yellow colony to win from a disadvantaged starting position population wise.
Very true through the entire video I saw two main deciders for who wins and that is isolation & least distance to the most amount of food(usually this would be the mid). Although on the fourth blue won because it was able to take hold of some uncontested piles of food by taking out red and it had a pretty short path to the mid.
For the future: add a music option that can play music depending on the stats of colonies. If rapid decrease in ants, music gets more chase/action like. If increase, more happy music.
Lesson 2 : postponing combat allowing a long and late war to happen was a mistake made by cyan and green anthills. A long war meant a war with a new strong enemy that would take over the world. Ahah France vs Germany during WW1, a long trench war that would allow USA to effortlessly take over the first march of the podium. :p Now we can understand why Germany chose the blitzkrieg at the beginning of WW2 ;) All this thanks to your 4 vs 4 anthill simulation. Once again, thank you.
Really cool how the red ants just found another way to the food in the middle when the blue ants blocked one access point in the first simulation. Extremely basic AI making itself look smarter than it is. Good work on this.
The mechanics of this look similar to a Galactic Colonization game I'm currently preparing to make (GDD/Game Design Document right now)- once the documentation is near complete I'll finally start coding it- made a few games with the GDD being in my head and realized that I really need to plan more in advance so that's what I'm doing now. Very cool, nice work!
This an absolutely amazing simulation of irl ants, very accurate, though there are a phew things missing to make it nearly perfect. Full applause from me.
an idea I have for the next one is maybe inside the nest activities visualised? like ants storing larva/food, along with maybe expanding nest., another idea is once an ant nest has a certian number of ants in it it will create a new queen that will go out and found a new nest that might(or might not) be allied with the original
This calls for a competition. Contestants set certain parameters for their colony, like worker/soldier balance, sensitivity to the various pheromones, etc. Then pit them against each other on maps like these.
I love how in the second round blue and red were relentless on killing each other while yellow was jacking the last of the food and becoming unstoppable.
Cool simulation! I would suggest as @Joseph already said to add queens from the winning colonies that spawn a few new colonies. If you would give these queens specific genes, encoding their traits, and applying combination and/or mutation you could evolve the colonies, changing their traits depending on their environments. I know that real ants colonies do not attack colonies with the same characteristics, until they evolve to much, such that they don't recognise eachother anymore. You could also use that principle to spawn new battles.
When soldiers find enemy workers, they should turn the enemy worker into a food dot. When two soldiers fight the loser should get turned into a food dot. Dying ants should leave a "fear" marker that repels workers of any color.
IRL ants are able to utilize the corpses. We can include this process into the food chain: ant workers drags the dead ants to the hive, which gives them "food material" to born a new ones. [sorry for my bad english]
This is awesome! Just an idea for possible improvement: from 7:53, the cyan soldiers seem to be trapped, i wonder if there was a way to lead them home after they killed the yellows
From what i notice, some soldiers follow to enemy markers when the area is empty. Each time an ant dies the ant that killed it will make “no more enemy markers” which will allow the markers to dissipate faster
Part 6 idea terrain, builder ants can use dirt to create walls and bridges over water. water slows ant and potentially kills them. Dirt piles spawn and bridges and walls are destructible. Builder ants use to battle and to water scents to know where to put walls and bridges
It depends on the number of ants in the simulation, on an M1 mac book pro it runs at 75fps with 16K ants and 90fps on an I9 9900K. It is still single threaded though.
AntSim works strangely. When I click on M, the markers are not displayed. With a population of 4000 ants, time does not accelerate. I can't build and erase walls. Food is also not added when the right mouse button is clicked, although all other functions (displaying ants, pause, accelerated time (if there are less than 4000 ants)) work. Sorry for possible errors in the text, I use a translator.
This video raises all sorts of questions about ant diplomacy, what happens when two colonies meet? Is it always total war? Is there some mechanism by which hostile ant colonies gauge each other's strength and cease hostilities if the war proves too costly?
some ideas I have: one: non-ant enemies non-ant enemies can be from a wasp to an antlion and can either have their designated spawn areas or just spawn randomly. when an ant finds one of these non-ant enemies, it will do the same as when it finds an ant from another colony. two: different ant species: Argentine ants: these ants are able to create a new allied colony, these colonies can assist each other in wars, as well as a couple other things. carpenter ants: these ants can actually slowly break down a barrier (I would suggest making two types of barriers, one can be destroyed by carpenters while the other cannot) if ant figures out that they can shorten routes by doing so. Army ants: these ants workers are a hybrid between a worker and a soldier, and are stronger than most other soldiers (they would probably cost quite a bit more food to make a single ant, they can also completely move the nest if they feel that there are no longer any food sources that can be quickly accessed. Crazy ants: these ants are more likely to break from existing lines and can get a few food areas being harvested at the same time, they also make much wider trails and can make workers slightly faster, and soldiers slightly slower. there are many more ideas for other species, and you do need to make sure that everything is balanced (aside from the fact that army ants can completely annihilate entire ecosystems without much trouble in nature) Three: parasitic mites these mites may grab on to a VERY unlucky ant, this mite will reproduce at least once before killing the ant, this will drop all mites that were on the ant to the ground, where they can be picked up by more ants, these could easily DESTROY a colony, unless handled properly
Also just curious, do you have any plans of making this a public thing that people can play, instead of this being something done on only a youtube channel, making this a public simulation/top-down strategy game would certainly bring me and others joy :D
makes me want to learn C++, being an adult with little free time makes that difficult however. also, what kind of pathfinding are you using? is it grid based?
Published a release for windows here (with the code) github.com/johnBuffer/AntSimulator/releases/tag/v3.1
1000 + 1000 + 1000 + 1000 = 4000
it says its not good
The release is fullscreen but seemingly only 1080p, which messes with my desktop since I have a 1440p monitor. Is it possible to have it run in a window, or switch to the native resolution? I'd rather not mess with the source code myself, as I've had little luck compiling C-related code for whatever reason.
@@shuriken188 I've just updated the release link to add a conf file alongside the executable with configurable window size and mode
@@PezzzasWork Awesome
I have an idea for a possible part 6: the ability for colonies to branch out and create more queens and therefore other colonies.
This will lay the foundations of "colony reproduction", a totally other aspect to ants.
I would also like to add the idea of giving the ability for corpses to be food, suggested by notus42 to give better reward for war
And sometimes they either start to fight, or start to form 'supercolonies', which are colonies with multiple queens.
YES
@@Zedryx69 in nature, this behavior is dependent on species. Most ants fight with child colonize, some ants form global colonize who rage war over several continents.
Aother small possible addition: Corpses become food.
@@Zedryx69 Then you have team competitions not just FFAs
Finally, I was hoping for the project to evolve in this way, great job.
hehe "evolve"
Best notification I've received in a while
Hope you will like it :)
You're not the only one
Same
amen
Same
This brings me such a childish joy. Thanks for sharing this project!
me too hahah
Project is really going great- this was a great way to implement the soldier ants. Shows that perfect balance required between gathering resources and destroying the nearby colonies. Seems like location plays a big factor.
Yeah, from what i saw, there are a few factors in play, they are:
1. How close are the nearest resource points?
2. How many are there?
3. Are there any bordering colonies?
4. Do they have access to more resource points than you do?
5. Are you close to the middle / have a clear path?
There are probably more, but in my opinion these are quite important, if not the most important.
@@MostScaryCat Yea, I noticed that proximity to more resources than the closest colony is a very strong advantage.
Didn't think about the clear path but looking again it does make a significant difference.
@@MostScaryCat I found that those who branched out and discovered new resource piles earlier tended to win, especially those that established strong connection lines by the time fighting begins
From my experience I see that there are 3 types of ants: workers, soldiers and explorers. I wonder how will it be if you add this third type, they should be faster and release a signal of higher priority(almost the only one) for the correct path.
That's a good idea
@@mementomori7160 hmmm
There is just workers and soldiers, it’s just that in this simulation ants are able to follow a path but have different genes compared to everyone else, making them take unique paths that happen to make some ants go off-course and end up finding other resources or areas.
this in my opinion would reduce the main advantage of the ant algorithm which is resiliency. The nice thing about only one priority is that the evolution of the colony along with the environment dictates where they go. Prioirties sort of arise from their interaction
@@grantinlow2889 increased birth rate?
As much as I love the RGB color space, when choosing colors for teams, I highly recommend using the YUV color space instead. With a constant Y value, all colors will be the same brightness. In your current video, blue ants are much much harder to see than yellow ants. If you used the YUV then you would be able to easily choose different distinguishable colors while keeping all teams equally visible.
Thank you for this remark!
@@PezzzasWork I'm glad to be of assistance! Keep up the amazing work!
Now this is the top down strategy game I've been dreaming of.
With the main question being 'How can I influence this to get the victory?'.
@@Bloodlyshiva A while back, I wrote something similar to this to run on the GPU, the best ways of implementing player control would be.
1. Instruction pheromone - allow the player to paint pheremone directly - allowing them to "soft direct" ants at will (to attack other colonies, get a certain resource etc)
2. Population control, allow the user to dynamically set ant production rates for new nests vs worker ants - kinda boom vs rush style
3. Tech trees, allow you to improve speed, attack strength, or other stats which costs foods that the ants collect
Lord of the ants it's basically just lord of the Rings but with ants
just play starcraft and choose zerg
"thanks for wathing" 8:53 xD
Amazing vid, it is incredible to see how surprising are the outputs are differents depending on the map and the number of enemy colonies.
I wonder what are the perfect parameters for a colony to position itself well in relation to its environment
Sounds like it’s going to be a fun one! Love your ant projects keep up the good work.
This would make for an amazing mobile sandbox game. Just a simple UI where you can tab to place different assets such as food, walls, or different ant spawns and the ability to speed up or slow down time would make for a really simple but fun game. You could even put it for a small price to make some profit!
Edit: a game similar to what I described above is called "Robot Colony"
Since the project is open source, I guess someone can port it to android.
theres a game like it lol, but you can play it yourself and control things. mobile and online game
@@miniiogaming6954 Name please
@@miniiogaming6954 name please
@@onmomentum the only game I can think of is "Pocket Ants". Played it for a while, quite fun. You actually play as an "ant commander", so it's quite different from this but quite the same I guess
Great video.
How about: Dead ants from the same colony could give of a death hormone. Workers who encounter these, will bring the information back to the colony. The more often the colony receives these (in a short timespan?), the higher the change it'll breed a soldiers to combat the threat?
Dead ants to be used as food?
so kinda like people in a way?
@@vaibhavbv3409 I'm not an ant expert but i don't think ants eat dead ants
@@alansus586 they eat dead ant, but not the same colony one
Wow, it is fascinating how this project evolves; not just the ants, the visuals, the UI, controls (I'm assuming)!
Still, no machineguns on the ants...
hopefully next episode
soon, the workers will have pickaxes and the soldiers will have worker machine guns.
@@MostScaryCat and also soon the spies are going ninja stars and ranged soldiers are going to have shields
You can compare strategies: one colony will consistently produce fighters, which means less workers but be more prepared once you are attacked. The other colony will only produce fighters when needed.
Ant ideologies
@@Nugcon Fasticm can be one where they are fast
Next Video: C++ ant simulation, ants take over earth
Later: C++ ant simulation, ants rule universe
Little More Later: C++ ant simulation, ants become god
Ant understand how to produce food for them
C++ ants create create C++ humans simulation
Ants do photosynthesis
I am loving every single video of yours and this is probably the best follow up video ever.
To pile on with the suggestions: it seems like positioning is the major factor for ants to win or loose here. How about some variation in the ants stats and priorities? Probably based on some square law plus constant. You can have a little evolution algorithm and tournament, hopefully producing several types of ants to excel at different types of levels.
Again, love your work. Please keep them coming. Cheers
Great work so far, I’m loving how this project is turning out!
We've gone from walking to eating to helping to war just like mankind
I love that at the end you named the chapter "Thanks for wathing" due to your spelling mistake. Made me chuckle a little bit.
These videos keep getting more and more interesting. Keep up your amazing work my man
welp this has now turned into a self-playing realtime strategy game
love it
fully one of the best videos on youtube - this would be the king of all flash games with an easily editable environment
I love how simple instructions when applied on a large scale leads to complex outcomes.
This is increeible, good job man, nice to see this kind of project. You are great inspiration!
Incredible! So excited to see this in my subscriptions list!
"Thanks for wathing" 😂 This was fascinating! I just found your channel today really cool simulations! We need more!
unbelivelble what you can do with programming ... everytime its surprises me somehow ... soon i will be able aswell ... well not thos thingis but none the less
I've been waiting for this for ages, and i am not disappointed
I assume there is not limit for your innovative way of thinking. Impressive work and remarkable results. Thank you for inspiring the great community. Keep fingers for your goals.
these are great! It seems the winner is usually the nest spawning with the nearest food blob. I interpret it like those nests increase populations the soonest and population increases search space and coverage density, so the nest gets to make better 'decisions' sooner. Also, and perhaps mostly, the exponential increase in pop
I would agree. The sooner an ant can bring its food back the sooner it can go to collect more. Having closer food means more throughput, which means faster population growth, which translates to more efficient food gathering from farther food sources. A closer starting food source snowballs quite quickly.
I wonder how the ants would do with chokepoints, infinite food sources, and imbalanced maps. For instance, could a colony with only a single infinite food source outlast a colony with many finite food sources if there's a choke point to defend?
@@Zebo12345678 That's pretty interesting. I think it would also be interesting to see the simulation run against those scenarios, but also the 'latent space' between them.
So if there is a 1v1 map with a central food source, and I wonder what happens if one nest was further away or if there was an obstacle in the way or if the food had different shapes and cavities... Those scenarios may be interesting but if we call the map a picture and take a geometric 'morph' between the map of the base scenario and a modified scenario and we simulate on each of the maps generated by each step in the morph, that could also yield interesting results.
Perhaps to watch something like that (in a timely manner) the videos would need to be layered over each other with each iteration using a slight brightness or hue shift or something.
I just think it would be cool to have an idea of why a scenario plays out a certain way and watch to see if it's possible to determine, as a scenario changes little by little and morphs into the alternate scenario, when a result should change and why.
Like, if there is a 1v1 open map except for a wall with a tiny gap between competing nests, and one nest has a nearby food source and the other a more distant food source, then if the gap is small enough, it could be difficult to press the advantage of faster growth. Seeing what happens as the gap opens wider each iteration until the advantage could be pressed and trying to predict when such a thing could happen, would be interesting.
@@solerante That's a great idea too. These simulations are kind of addictive, huh?
Would you ever consider publishing and selling this somehow? The idea of generating maps with food and terrain for ant colonies to expand and fight over seems a reasonably compelling one for an indie game. I've rewatched this like 5 times now, just looking at all aspects of how the ants behave and how they react to their environment and how heavily that determines their success against a now opposing force. Especially interesting was the interaction between food signals from different factions, like how if one side had food signals laid down, past or present, they would draw workers there, often turning the tide in nearby battles against the opposing mainly-soldier force. All in all, super cool simulation, I love your creativity and the simple readability of your sim.
It's open source, you can check the description for a link
At first I came across this, was like, "yes. I like this." And then binged all the ant stuff that came before. Imma now go look at the rest of your content
I see, the denser food in the middle acts as a decisive resource point. Whomever gain's control of it first has a massive advantage because it guarantees stable population growth, and once they have the larger population they can fight off any attempts from other colonies to take that food. In the three colony battle this allowed the yellow colony to win from a disadvantaged starting position population wise.
Very true through the entire video I saw two main deciders for who wins and that is isolation & least distance to the most amount of food(usually this would be the mid). Although on the fourth blue won because it was able to take hold of some uncontested piles of food by taking out red and it had a pretty short path to the mid.
your welcome i loved wathing this video
For the future: add a music option that can play music depending on the stats of colonies. If rapid decrease in ants, music gets more chase/action like. If increase, more happy music.
Lesson 2 : postponing combat allowing a long and late war to happen was a mistake made by cyan and green anthills. A long war meant a war with a new strong enemy that would take over the world. Ahah France vs Germany during WW1, a long trench war that would allow USA to effortlessly take over the first march of the podium. :p
Now we can understand why Germany chose the blitzkrieg at the beginning of WW2 ;)
All this thanks to your 4 vs 4 anthill simulation. Once again, thank you.
Really cool how the red ants just found another way to the food in the middle when the blue ants blocked one access point in the first simulation. Extremely basic AI making itself look smarter than it is. Good work on this.
whoever has clearest access to the most food wins
the red markers could be used to mark "claimed territory" for each colony
Great work, always happy when you upload!
Wow when I saw this for the first time I thought it is a one time only. Please keep improving this. I’m looking forward to the next one.your awesome
First time viewer, this was surprisingly entertaining to watch. :D
The mechanics of this look similar to a Galactic Colonization game I'm currently preparing to make (GDD/Game Design Document right now)- once the documentation is near complete I'll finally start coding it- made a few games with the GDD being in my head and realized that I really need to plan more in advance so that's what I'm doing now.
Very cool, nice work!
This an absolutely amazing simulation of irl ants, very accurate, though there are a phew things missing to make it nearly perfect. Full applause from me.
really great work, the "thanks for watching" at the end is really cool !
You mean thanks for "wathing" :D
an idea I have for the next one is maybe inside the nest activities visualised? like ants storing larva/food, along with maybe expanding nest., another idea is once an ant nest has a certian number of ants in it it will create a new queen that will go out and found a new nest that might(or might not) be allied with the original
This calls for a competition. Contestants set certain parameters for their colony, like worker/soldier balance, sensitivity to the various pheromones, etc. Then pit them against each other on maps like these.
Yes please! I am hoping different stats will be better at different maps
The ending was really cool! Looks like whatever the ants en masse grabbed was like wind blowing it away.
Could the ants get "energy" from eating another ant-type's nest?
it already seems like destroying another colony is a big advantage when you get access to the resources, that doesn't seem like a super great idea
At this rate you're going to recreate real life ants! You're awesome!
I love how in the second round blue and red were relentless on killing each other while yellow was jacking the last of the food and becoming unstoppable.
I have just found everything that I would ever want in this one channel.
At this rate you'll have made a simulation of not only multiple colonies, but the actual ant nests! Now that is something id like to see!
Cool simulation! I would suggest as @Joseph already said to add queens from the winning colonies that spawn a few new colonies. If you would give these queens specific genes, encoding their traits, and applying combination and/or mutation you could evolve the colonies, changing their traits depending on their environments. I know that real ants colonies do not attack colonies with the same characteristics, until they evolve to much, such that they don't recognise eachother anymore. You could also use that principle to spawn new battles.
yessssss been waiting on this =)
Me cheering go yellow in my head The whole time!
When soldiers find enemy workers, they should turn the enemy worker into a food dot. When two soldiers fight the loser should get turned into a food dot. Dying ants should leave a "fear" marker that repels workers of any color.
how do u add more then 2 colonies?
Great video!
I'm always glad to wath (8:53)
With 3 or more colonies it should be a "capturing colony mechanic" and keep growing ants with its new team
Strong nostalgia for playing SimAnt on my family’s IBM Windows 3.2 back in the day
IRL ants are able to utilize the corpses. We can include this process into the food chain: ant workers drags the dead ants to the hive, which gives them "food material" to born a new ones.
[sorry for my bad english]
This is awesome! Just an idea for possible improvement: from 7:53, the cyan soldiers seem to be trapped, i wonder if there was a way to lead them home after they killed the yellows
From what i notice, some soldiers follow to enemy markers when the area is empty. Each time an ant dies the ant that killed it will make “no more enemy markers” which will allow the markers to dissipate faster
I could watch these all day. Love it
Part 6 idea terrain, builder ants can use dirt to create walls and bridges over water. water slows ant and potentially kills them. Dirt piles spawn and bridges and walls are destructible. Builder ants use to battle and to water scents to know where to put walls and bridges
There it is! The good video i was looking for!
Great job!
I think the sound you use in these experiments is always overshadowed. 🎵👍
I enjoyed wathing. I will wathe the next one. Glad I wathed it.
Wow! Simulations like this have come a super long way since the 80's.
what type of Pikmin will you fight for?? red, blue or yellow??????
Best notification I got in 2 years of UA-cam premium
Thanks for wathing?
this just shows that in real time strategy games, resources and economy is what wins games.
I was supporting dark blue! They made a pretty nice comeback
irk lol!
I definitely enjoyed "wathing" that!
8:50 "Thanks for wathing" is actually pretty funny, I didn't even notice at first untill I looked at the timestamps
Is this particularly taxing on the computer at all? Wondering how complex and large this can be made
It depends on the number of ants in the simulation, on an M1 mac book pro it runs at 75fps with 16K ants and 90fps on an I9 9900K. It is still single threaded though.
@@PezzzasWork That is pretty amazing. Pretty good testament of SFML's capabilities as well if you use their drawing API.
AntSim works strangely. When I click on M, the markers are not displayed. With a population of 4000 ants, time does not accelerate. I can't build and erase walls. Food is also not added when the right mouse button is clicked, although all other functions (displaying ants, pause, accelerated time (if there are less than 4000 ants)) work. Sorry for possible errors in the text, I use a translator.
Fight of attrition. Resource is key. What are the chances of the results being the same on multiple simulations on the same asymmetrical map?
what game/simulation do you use?
Thanks for wathing! Joke aside, you are an inspiration.
I enjoyed wathing your video! Very eduathional! ;-)
This video raises all sorts of questions about ant diplomacy, what happens when two colonies meet? Is it always total war? Is there some mechanism by which hostile ant colonies gauge each other's strength and cease hostilities if the war proves too costly?
You should try it with a food source that slowly regrows. I wonder if some kind of equilibrium would develop that allows multiple colonies to survive
Man this ant war thing is a great tool for checking out initial map balance in resource distribution.
I have never been so lost in my entire life that I decided to watch ants fights
and yet here I am
It just keeps getting better and better
So cool to see the soldier ants protect their suply lines
Very fun. Reminds me of the civ battle royale from a couple of years ago
thanks for wathing
Fr
Slang in my national language calls white noise 'ant war'...
Now this quite literally is the true definition of ant war.
applying RTS strategy to this was fun. the second a colony got stable supply and got it's population ahead of the others it was over.
some ideas I have:
one: non-ant enemies
non-ant enemies can be from a wasp to an antlion and can either have their designated spawn areas or just spawn randomly. when an ant finds one of these non-ant enemies, it will do the same as when it finds an ant from another colony.
two: different ant species:
Argentine ants: these ants are able to create a new allied colony, these colonies can assist each other in wars, as well as a couple other things.
carpenter ants: these ants can actually slowly break down a barrier (I would suggest making two types of barriers, one can be destroyed by carpenters while the other cannot) if ant figures out that they can shorten routes by doing so.
Army ants: these ants workers are a hybrid between a worker and a soldier, and are stronger than most other soldiers (they would probably cost quite a bit more food to make a single ant, they can also completely move the nest if they feel that there are no longer any food sources that can be quickly accessed.
Crazy ants: these ants are more likely to break from existing lines and can get a few food areas being harvested at the same time, they also make much wider trails and can make workers slightly faster, and soldiers slightly slower.
there are many more ideas for other species, and you do need to make sure that everything is balanced (aside from the fact that army ants can completely annihilate entire ecosystems without much trouble in nature)
Three: parasitic mites
these mites may grab on to a VERY unlucky ant, this mite will reproduce at least once before killing the ant, this will drop all mites that were on the ant to the ground, where they can be picked up by more ants, these could easily DESTROY a colony, unless handled properly
Also just curious, do you have any plans of making this a public thing that people can play, instead of this being something done on only a youtube channel, making this a public simulation/top-down strategy game would certainly bring me and others joy :D
I could watch this all day.
I want a strategy game based on this
It was very fun wathing!
This looks more and more like a game xD
Really cool!
can u make a tutorial on these ant simulations about how u get them to run so fast and smooth?
This would make the coolest wallpaper engine background
makes me want to learn C++, being an adult with little free time makes that difficult however.
also, what kind of pathfinding are you using? is it grid based?
There is no actual "pathfinding" algorithm here, ants find their way on there own using markers :)
@@PezzzasWork that's beautiful, its just like the real thing!
Choose your race
Player 1: *Zerg*
Player 2: *Zerg*
Player 3: *Zerg*