Final unifications

A place to talk about the simulation: dicuss how it works and what it does, and share what you've discovered.
Post Reply
User avatar
magmacube_tr
Posts: 934
Joined: Wed Jul 11, 2018 8:01 pm
Location: Now on the depths of the once-volcanic Hasan Mountain

Final unifications

Post by magmacube_tr » Fri Aug 02, 2019 10:51 pm

I have a question.

On all of my simulations after a long, long, long ingame time, all lifeforms seem to converge into one hyper-OP, unstoppable and almost zero diversity species. No matter how the simulation goes, how radical the fertility or tempature , how many extinctions happen or what type of the map is, just one species rule everything.

Why...

And more importantly...

Is this revealing something regarding the true nature of evolution :?:
Your everyday cube of molten rock. Also, a baby dragon.
Image

User avatar
counting
Posts: 185
Joined: Fri Apr 22, 2016 1:30 am

Re: Final unifications

Post by counting » Sat Aug 03, 2019 1:52 am

Are you being serious, I cannot seem to form a stable single ring species if I tried setting breeding range at 60 and have to go all the way to range 100 to keep them in one species.

How long is exact is a long long time ingame, how many generations is it in? And what is the setup of pop cap and fertility? Have you pause the game and do a general survey to compare the genetic distance of some geologically far away creatures with each other against nearby ones?
King Menander asked Master Nagasena: “I have questions.”
Master replied: “What questions?”
King: “I already asked.”
Master: “I already answered.”
King: “What is THE answer?”
Master: “What is THE question?”
King: “No question.”
Master: “No answer.”

User avatar
magmacube_tr
Posts: 934
Joined: Wed Jul 11, 2018 8:01 pm
Location: Now on the depths of the once-volcanic Hasan Mountain

Re: Final unifications

Post by magmacube_tr » Sat Aug 03, 2019 11:27 am

counting wrote:
Sat Aug 03, 2019 1:52 am
Are you being serious, I cannot seem to form a stable single ring species if I tried setting breeding range at 60 and have to go all the way to range 100 to keep them in one species.

How long is exact is a long long time ingame, how many generations is it in? And what is the setup of pop cap and fertility? Have you pause the game and do a general survey to compare the genetic distance of some geologically far away creatures with each other against nearby ones?
Wait a minute, making ring species in the simulation is easy as cake. You only have to set the fertility high and flat and after your world experiences its own cambrian explosion (like 5 to 85 in just in 10-20 minutes, no kidding), boom :!: One species outcompetes all the others and almost always stay as one gigantic ring species for the rest of eternity. breeding range was 10 btw. A second or even a third cambrian explosion can happen tho, so I can't really guarantee but it settles down to the one ring species after a while

And for the long long long time, it actually varies but I am saying that +50 days ingame time saves. Generation also varies but its usually higher than 2000. Pop cap 2000 and fertility high with tempature 30 gives the best results.

They outcrowd by pop. cap. or just plain competition and/or hunt other to species extinction. Creatures tend to settle to one omni-optimal strategy and stick to it 5ever once they found them. This way they also prevent new species or variations from emerging because there is simply no room for a new species and a creatures don't need to vary because they already found the optimal way to live, being different just means being outcompeted.

There is no definite strategy for being OP, they just do something it if it works.
Your everyday cube of molten rock. Also, a baby dragon.
Image

User avatar
counting
Posts: 185
Joined: Fri Apr 22, 2016 1:30 am

Re: Final unifications

Post by counting » Sat Aug 03, 2019 6:08 pm

I assume you didn't actually running for 50 days in realtime but x10 speed in game, right? I have long noticed that there are strange behaviors at x10 compared to running at x1 (or x3) speed. Creatures seem to skip some mating or feeding behaviors as well as some crucial behaviors that help them cope with extreme conditions (like curlup), this somewhat causes different kinds of creatures to be selected for reproduction at x10 speed. Just compare different running speeds in the same exact condition and the same in-game time several times, you will find the differences pretty significant and noticeable (especially with sexual reproduction, some might just go complete extinction at x10 speed no matter what but totally fine at x1 speed)

Also it has been confirmed by Quasar in the Steam forum that when constantly running at x10 speed for a very long time (regardless what breeding range is set), since speciation algorithm is running separately and independently, it tends to lag behind other functions. Hence you might just see the effect where the speciation algorithm just didn't have a chance to catch up and split them, where small splinter created by mutation just cannot sustained the one dominate group pressure. Have you paused the game and see if speciation happens during the pause after the long long simulation? And have you checked if the creatures actually further apart from each other have higher genetic distance than closer ones? (you can compare their genetic distance one-to-one in the genetic tab)

A very low breeding range I suspect also has an adversary effect in speciation, if a species don't have diversity built up and all the mutation speciation just died off, it stands for a reason that it will stay as one species where creatures are forced to evolved to become universal adapters. I actually think the default 20 is also too low, in many experiments I feel a range between 30 to 40 is more stable for sexual reproduction. Speciation will occur very slowly but they tend to split in more reasonable sizes and meaningful ways and stay in their niche, geographically or in different diet, (pure herbivore/carnivores are not very stable in very long run regardless, but I still managed to like hundreds to near a thousand generations where herbivores tend to have way higher generation numbers than carnivores), and for what I called light carnivores or light herbivores hybrids with enough genetic diversity within them tend to be pretty stable and can co-exist where they reach reasonable sizes. And paradoxically, they need pretty sparse food density (not fertility, the amount of energy a creature can get in an area and travel distance they need to reach another food source), and very low food efficiency, this way the omnivores don't have too much of an advantage, and diet of extreme value can have their niche (I think pure high diet value species are unstable in the long run for other reasons though, related to food targeting and food choices regarding inter/intra species behaviors, and their interlocked boom/bust cycles)

Although I have to say I also noticed the "curse of omnivores" phenomena. I think there are many conditions associated with that, such as unbalanced genetic distances between different head types and their types' diet range; or attack is way out of proportion to HP and defense, so creatures tend to one-shot each other (a side effect for their limbs/tips), hence aggression is heavily favored for cheap meat diet, and creatures all become sociopaths in the long run killing each other on constant basis (thus new small species just instantly got killed off); or population cap favors dominate species when pop cap is met for long period of time, the bigger group just reproduce more and squeeze out smaller species; or high fertility high food efficiency of rich food sources just make creatures with enough stamina and fast reproduction just don't required to adapt to any niche conditions, as long as they can produce many litters and eat siblings at birth, then give birth and die producing more food, thus creating perpetual alive group in one tight spot; or even the auto population control where food efficiency can have a run away effect where food types can have impossible values and tend to break the population balance (food efficiency tend to have higher auto adjust increment at larger value, but dropped way slower, I once observed one type of niche food efficiency run into trillions of trillions value and break the display of hotkey H food efficiency overlay). I certainly agree there are lots of balancing and debugging to do.
King Menander asked Master Nagasena: “I have questions.”
Master replied: “What questions?”
King: “I already asked.”
Master: “I already answered.”
King: “What is THE answer?”
Master: “What is THE question?”
King: “No question.”
Master: “No answer.”

User avatar
magmacube_tr
Posts: 934
Joined: Wed Jul 11, 2018 8:01 pm
Location: Now on the depths of the once-volcanic Hasan Mountain

Re: Final unifications

Post by magmacube_tr » Sat Aug 03, 2019 6:46 pm

counting wrote:
Sat Aug 03, 2019 6:08 pm
I assume you didn't actually running for 50 days in realtime but x10 speed in game, right? I have long noticed that there are strange behaviors at x10 compared to running at x1 (or x3) speed. Creatures seem to skip some mating or feeding behaviors as well as some crucial behaviors that help them cope with extreme conditions (like curlup), this somewhat causes different kinds of creatures to be selected for reproduction at x10 speed. Just compare different running speeds in the same exact condition and the same in-game time several times, you will find the differences pretty significant and noticeable (especially with sexual reproduction, some might just go complete extinction at x10 speed no matter what but totally fine at x1 speed)

Also it has been confirmed by Quasar in the Steam forum that when constantly running at x10 speed for a very long time (regardless what breeding range is set), since speciation algorithm is running separately and independently, it tends to lag behind other functions. Hence you might just see the effect where the speciation algorithm just didn't have a chance to catch up and split them, where small splinter created by mutation just cannot sustained the one dominate group pressure. Have you paused the game and see if speciation happens during the pause after the long long simulation? And have you checked if the creatures actually further apart from each other have higher genetic distance than closer ones? (you can compare their genetic distance one-to-one in the genetic tab)

A very low breeding range I suspect also has an adversary effect in speciation, if a species don't have diversity built up and all the mutation speciation just died off, it stands for a reason that it will stay as one species where creatures are forced to evolved to become universal adapters. I actually think the default 20 is also too low, in many experiments I feel a range between 30 to 40 is more stable for sexual reproduction. Speciation will occur very slowly but they tend to split in more reasonable sizes and meaningful ways and stay in their niche, geographically or in different diet, (pure herbivore/carnivores are not very stable in very long run regardless, but I still managed to like hundreds to near a thousand generations where herbivores tend to have way higher generation numbers than carnivores), and for what I called light carnivores or light herbivores hybrids with enough genetic diversity within them tend to be pretty stable and can co-exist where they reach reasonable sizes. And paradoxically, they need pretty sparse food density (not fertility, the amount of energy a creature can get in an area and travel distance they need to reach another food source), and very low food efficiency, this way the omnivores don't have too much of an advantage, and diet of extreme value can have their niche (I think pure high diet value species are unstable in the long run for other reasons though, related to food targeting and food choices regarding inter/intra species behaviors, and their interlocked boom/bust cycles)

Although I have to say I also noticed the "curse of omnivores" phenomena. I think there are many conditions associated with that, such as unbalanced genetic distances between different head types and their types' diet range; or attack is way out of proportion to HP and defense, so creatures tend to one-shot each other (a side effect for their limbs/tips), hence aggression is heavily favored for cheap meat diet, and creatures all become sociopaths in the long run killing each other on constant basis (thus new small species just instantly got killed off); or population cap favors dominate species when pop cap is met for long period of time, the bigger group just reproduce more and squeeze out smaller species; or high fertility high food efficiency of rich food sources just make creatures with enough stamina and fast reproduction just don't required to adapt to any niche conditions, as long as they can produce many litters and eat siblings at birth, then give birth and die producing more food, thus creating perpetual alive group in one tight spot; or even the auto population control where food efficiency can have a run away effect where food types can have impossible values and tend to break the population balance (food efficiency tend to have higher auto adjust increment at larger value, but dropped way slower, I once observed one type of niche food efficiency run into trillions of trillions value and break the display of hotkey H food efficiency overlay). I certainly agree there are lots of balancing and debugging to do.
1. Duh. I am not an idiot, who owuld wait for 50 days :?:

2. I did pause and waited a bit. But nothing came out except mutants.

Edit: Fixing typo NOOT NOOT
Your everyday cube of molten rock. Also, a baby dragon.
Image

User avatar
counting
Posts: 185
Joined: Fri Apr 22, 2016 1:30 am

Re: Final unifications

Post by counting » Sun Aug 04, 2019 9:16 am

I did run experiments for days to weeks at x1 though (have extra idle servers in the lab just put there and wait, if they don't crash). And the results are much stable and interesting though.

Lagging behind speciation might be a factor during the long long simulation where the speciation should happened didn't happen, and the internal groups might just able to exchange genes frequent enough to keep it in one species (did you also use sexual reproduction or parthenogenic?). I am curious what exactly is the genetic composition of your creatures and their clade diagram overtime? (Did you still have the save file?)

I've also tested with breeding range 10 for my ring species central lake oasis map, and it did sort of help. I used a very uniformed map, with flat fertility, and only grass (no tree, and 0 meat efficiency), running at x10 speed, the only variables are temperature gradient from left to right. When I kept the range at 10, they just spread and then broke apart, mostly into 3 to 5 major species sections around the lake shore, sometimes one of the species would become dominate slowly and took over half to 2/3 of the ring. But when they reached their peak, they broke apart again. So I came up with the plan of letting the game run for a while and waited for one of the species that spammed from hot to cold temperature and started to expand, then increased the breeding range up to 14 to 20 to keep them together. This usually end up with the major species able to spam further and didn't break apart for extended period of time. When they reach like 80% of the ring, I just manually kill the remaining "resistant species" with only the dominate ring species left, and let the ring connect.

Here are the save files
https://www.dropbox.com/s/c2o4uvu2xweb5 ... s.zip?dl=0
https://www.dropbox.com/s/crfe2veeaawp8 ... s.zip?dl=0

Interestingly, I have to drop the breeding range back down to 11 or 10 for the ring species to not break apart in major sections after I clear out the last resistant species. It would split out tiny groups to keep it together longer. If I keep at range 14 or even up to 20, they break apart faster and into bigger chucks.

When I exam their genetic makeup and distance of each other, I tend to find that in any sections there are several local clusters mixed together, like some kind of family clans. Within the clan the distance is usually below 10, where genetic distance to nearby other clan members range from 30 to 80. But further away the genetic distance can go as high as 200+. Also on the opposite side of the lake when 2 approaching ends meet, their genetic distance is not as further apart than to the warmer temperature section creatures. They seems to converge in evolution and adapt similar features when adapting to cold temperature, although they do have higher clan to clan genetic distance at the location where 2 ends is about to meet. I feel this kind of mixed clans internal structure is actually where the break apart happened over long period of time. Some local clans just grow and take over as majority in a sizable section, until they reach the narrow points of the shoreline, and then genetic drift occurred within their clan and not spread outside of this section. When a group reach certain size and able to used all the food constantly, they tend to be able to resist the invasion quite effectively if the connected section is narrow enough, few wondering creatures cannot survive or mate in the local section (even though they are technically still within the same species)

I really hope that when the map editor is out in next version, I can create smooth ring to further reduce this kind of geographically barriers with uniform ring width, and see what would happen.
King Menander asked Master Nagasena: “I have questions.”
Master replied: “What questions?”
King: “I already asked.”
Master: “I already answered.”
King: “What is THE answer?”
Master: “What is THE question?”
King: “No question.”
Master: “No answer.”

User avatar
magmacube_tr
Posts: 934
Joined: Wed Jul 11, 2018 8:01 pm
Location: Now on the depths of the once-volcanic Hasan Mountain

Re: Final unifications

Post by magmacube_tr » Sun Aug 04, 2019 12:27 pm

counting wrote:
Sun Aug 04, 2019 9:16 am
I did run experiments for days to weeks at x1 though (have extra idle servers in the lab just put there and wait, if they don't crash). And the results are much stable and interesting though.

Lagging behind speciation might be a factor during the long long simulation where the speciation should happened didn't happen, and the internal groups might just able to exchange genes frequent enough to keep it in one species (did you also use sexual reproduction or parthenogenic?). I am curious what exactly is the genetic composition of your creatures and their clade diagram overtime? (Did you still have the save file?)

I've also tested with breeding range 10 for my ring species central lake oasis map, and it did sort of help. I used a very uniformed map, with flat fertility, and only grass (no tree, and 0 meat efficiency), running at x10 speed, the only variables are temperature gradient from left to right. When I kept the range at 10, they just spread and then broke apart, mostly into 3 to 5 major species sections around the lake shore, sometimes one of the species would become dominate slowly and took over half to 2/3 of the ring. But when they reached their peak, they broke apart again. So I came up with the plan of letting the game run for a while and waited for one of the species that spammed from hot to cold temperature and started to expand, then increased the breeding range up to 14 to 20 to keep them together. This usually end up with the major species able to spam further and didn't break apart for extended period of time. When they reach like 80% of the ring, I just manually kill the remaining "resistant species" with only the dominate ring species left, and let the ring connect.

Here are the save files
https://www.dropbox.com/s/c2o4uvu2xweb5 ... s.zip?dl=0
https://www.dropbox.com/s/crfe2veeaawp8 ... s.zip?dl=0

Interestingly, I have to drop the breeding range back down to 11 or 10 for the ring species to not break apart in major sections after I clear out the last resistant species. It would split out tiny groups to keep it together longer. If I keep at range 14 or even up to 20, they break apart faster and into bigger chucks.

When I exam their genetic makeup and distance of each other, I tend to find that in any sections there are several local clusters mixed together, like some kind of family clans. Within the clan the distance is usually below 10, where genetic distance to nearby other clan members range from 30 to 80. But further away the genetic distance can go as high as 200+. Also on the opposite side of the lake when 2 approaching ends meet, their genetic distance is not as further apart than to the warmer temperature section creatures. They seems to converge in evolution and adapt similar features when adapting to cold temperature, although they do have higher clan to clan genetic distance at the location where 2 ends is about to meet. I feel this kind of mixed clans internal structure is actually where the break apart happened over long period of time. Some local clans just grow and take over as majority in a sizable section, until they reach the narrow points of the shoreline, and then genetic drift occurred within their clan and not spread outside of this section. When a group reach certain size and able to used all the food constantly, they tend to be able to resist the invasion quite effectively if the connected section is narrow enough, few wondering creatures cannot survive or mate in the local section (even though they are technically still within the same species)

I really hope that when the map editor is out in next version, I can create smooth ring to further reduce this kind of geographically barriers with uniform ring width, and see what would happen.
1) Parthogenic.

2) It crashed, so I deleted it.
Your everyday cube of molten rock. Also, a baby dragon.
Image

User avatar
counting
Posts: 185
Joined: Fri Apr 22, 2016 1:30 am

Re: Final unifications

Post by counting » Sun Aug 04, 2019 5:59 pm

Parthenogenic has the tendency for creatures to adapt asexual in the long run since its reproduction rate is slightly higher without the extra mating behaviors. It generally ended up with less clusters and gene exchange but mostly cloning.

Too bad the save files is corrupted (not even autosave can be reloaded?)

Still testing the extreme climate continent I will let it run at x1 speed for several days and see how they turned up. The windows auto update just reboot the server and have to reload from autosave. It seems to create some new balance after reloading.
King Menander asked Master Nagasena: “I have questions.”
Master replied: “What questions?”
King: “I already asked.”
Master: “I already answered.”
King: “What is THE answer?”
Master: “What is THE question?”
King: “No question.”
Master: “No answer.”

Post Reply

Who is online

Users browsing this forum: No registered users and 2 guests