Major Update for Ecosystem Demo Out Now

[MikeSlugDisco]

Welcome everyone on this fine Autumn day; a major update to the Ecosystem demo has just been released! This coincides with the Steam Autumn Game Festival kicking off later on but the update is live now, so let's talk about all the changes.



From @Tom Johnson :

"My overall goal with these changes is to make the game feel more like a living system so that every aspect of it is connected to all the others, giving the creatures a lot to work with when it comes to evolving adaptations to their environments. For example, in the previous demo, creating a nursery was artificial in the sense that any 'U' shape was as good as any other, but now a cove really does catch eggs and plant seeds and can be affected by how water is funneled through your landscape; each one is distinct and they are all connected to the overall system. Creatures are both more interesting and a lot faster, and now that plants spread and can be depleted in locations, foragers will need to migrate to follow them with predators following close behind. In this way, populations are more mobile and interact in varied ways, and a lot of work has gone into keeping a large amount of species diversity in the environment."

Changelog

Creature Physiology

• Creatures are now much more likely to evolve symmetric body shapes
• Creatures are able to swim dramatically more quickly
• Reworked soft-body simulation on creatures so they are a little less rigid

Vegetation

• It is now possible to make large kelp forests, and now corals will build reefs themselves over several generations
• Each species of plant or coral has an accurate representation of its tolerances for nitrogen and phosphorus as well as sunlight, floor substrate, and crowding
• Healthy plants now release seeds that are carried elsewhere on the current - unhealthy ones die off. Foragers now need to follow the food, leaving behind barren or depleted areas and seeking out fresh growths. In general, this results in much more species intermixing

Species Diversity and Trophic Stability

• It is now possible to maintain a variety of different species without them driving each other to extinction
• It is now possible to keep predator-prey populations steady, including up multiple trophic levels, ie apex predators who eat smaller predators who eat herbivores
• Population counts roughly fit the Lotka–Volterra equations used to model predation in nature

Gameplay

• Players no longer build fixed nurseries: creatures can spawn anywhere
• Players no longer build fixed mating grounds: creatures decide how to mate themselves
• Substantially increased life points and nutrient points

Fluid Simulation

• Used a Lattice Boltzmann method to simulate current flows around the terrain as the player shapes it. Currents transport plankton, eggs, nutrients, and seeds around the environment

Creature Vision

• Creatures must now explore their environment to learn about it rather than having all information available to them immediately
• To detect a mate, prey, or predator, creatures must pass a vision check based on light level, motion and a comparison of the creature’s skin color to whatever is behind it, allowing creatures to evolve to lurk in dark caves or camouflage amongst plants, corals, etc

Creature Behaviour

• Alternate mating strategies: pairing up into mutually-compatible couples, promiscuously seeking out the most attractive partner, or gathering en masse at a central mating grounds
• Creatures emote when they are happy, sad, or in love
• Creatures can lunge at a target in addition to swimming at a steady pace

Pathfinding

• Creatures can now make their way through cave systems and around nooks and crannies

Creature Editor

• Added a fully-functional editor that allows players to design a creature's body directly and let it loose in the environment; it will evolve a brain

Terrain Editor

• Added a fully-functional editor that allows adjusting large scale parameters of the landscape to easily and quickly create beaches, reefs, lakes, cliffs, or various alien-looking oceans

UI

• Reskinned menus
• Added menus for audio, graphics settings, and key bindings
• When the player selects a creature, they also get a small picture-in-picture view of what the creature is seeing
• Reworked tutorial to reflect changes to gameplay
• Reworked food source menu so that it's clear there are fifty plants and corals and not just twelv

Sound Design

• Finished score
• Added several sound effects

Bug Fixes

• Fix: selecting creatures with a mouse click would get thrown off due to the underwater refraction effect.
• Fix: creatures could block the camera from moving.
• Fix: blurriness in text that would appear over creatures.
• Many more

 

[Serafine]

This is amazing!
I really love the new current spore dispersal system, makes the game much more fun to see your coral reef grow on it's own.
I also think removing the mating grounds was a really good idea as especially in the early game many creatures struggled to actually get to those and died simply because they had bad movement or swam off to far.

Funny thing, I had pretty much the exact same idea with the stealth and visual mechanics.
(I haven't seen it in effect though, even after 6 hours of play all my creatures still always had 100% visibility)

Something that's not yet in the game I think would be cool is parasites that can cling to (or orbit) larger creatures and suck energy/health out of them (of course there should be a limit to their capabilities so they don't easily wipe out all their hosts). This would also open up a niche for "cleaner" predators that specialize on eating those parasites but no larger creatures.

I was also think about some sort of a toolbox trait system where creatures can acquire certain random abilities like armor, poison, ink spray defense (vastly increases camoflage to a point where it can easily break line of sight to a predator but has a long cooldown), tentacles (allows to grab and pull in objects), electric shock area damage and other things that are too special to be represented by an entirely gradual system.
These abilities could start at a very low strength level and have something like a permanent energy drain that corelates with their effect strength (better ability costs more energy/time) so they'd only evolve if they actually help the creature (aka are worth more than their cost), like in actual evolution - if creatures keep them long enough their costs could reduce over time to simulate some sort of streamlining effect (this could be done by randomly upping/downing the ability costs in offspring).

I already love the game (it reminds me of Swimbots but in 3D and much more complex), can't wait for future updates!


p.s. Something I'd really appreciate for the demo would be an option to continue the last game (not really a save game system, just keeping one game until you decide to start a new one), that would really help with testing how the game behaves long-term. In Swimbots it usually takes a few days of running time until creatures get really efficient at what they're doing and it seems in Ecosystem they're developing a lot slower overall.

 

[Tom Johnson]

This is amazing!
I really love the new current spore dispersal system, makes the game much more fun to see your coral reef grow on it's own.
I also think removing the mating grounds was a really good idea as especially in the early game many creatures struggled to actually get to those and died simply because they had bad movement or swam off to far.

Thank you, I'm really glad that you liked it! It's a little nerve-wracking to release something after working on it for such a long time so it really helps and is encouraging to read that you had such a good experience.

Funny thing, I had pretty much the exact same idea with the stealth and visual mechanics.
(I haven't seen it in effect though, even after 6 hours of play all my creatures still always had 100% visibility)

I've also noticed that camouflage is a bit rare. When I first implemented it, I ran some tests where a small percent of a population would blend in with the nearby environment and most wouldn't, and in 8/10 cases the camouflaged coloration did 'take over' once I introduced a predator, so at least in theory it works, but I think it may need a bit of tweaking somewhere to make it happen more frequently 'in the wild', as it were.

Something that's not yet in the game I think would be cool is parasites that can cling to (or orbit) larger creatures and suck energy/health out of them (of course there should be a limit to their capabilities so they don't easily wipe out all their hosts). This would also open up a niche for "cleaner" predators that specialize on eating those parasites but no larger creatures.

I like this idea and I always found that interaction with cleaner fish very interesting. It could also potentially link up with gameplay systems handling dead bodies and scavenging. I think it would be a good candidate for something to add while the game is in Early Access.

I was also think about some sort of a toolbox trait system where creatures can acquire certain random abilities like armor, poison, ink spray defense (vastly increases camoflage to a point where it can easily break line of sight to a predator but has a long cooldown), tentacles (allows to grab and pull in objects), electric shock area damage and other things that are too special to be represented by an entirely gradual system.
These abilities could start at a very low strength level and have something like a permanent energy drain that corelates with their effect strength (better ability costs more energy/time) so they'd only evolve if they actually help the creature (aka are worth more than their cost), like in actual evolution - if creatures keep them long enough their costs could reduce over time to simulate some sort of streamlining effect (this could be done by randomly upping/downing the ability costs in offspring).

I already love the game (it reminds me of Swimbots but in 3D and much more complex), can't wait for future updates!

Interestingly, this is very similar to what I was working on when it came time to start on the updated demo! There wasn't enough time to complete it so I put it all on hold (though evidently not perfectly, as someone on Facebook mentioned that one of their fish seemed to be firing electric arcs at others).

Thanks again! I appreciate you took the time to write.

 

[Serafine]

Thank you, I'm really glad that you liked it! It's a little nerve-wracking to release something after working on it for such a long time so it really helps and is encouraging to read that you had such a good experience.

I really enjoy it, can't wait for the early access release.

I've also noticed that camouflage is a bit rare. When I first implemented it, I ran some tests where a small percent of a population would blend in with the nearby environment and most wouldn't, and in 8/10 cases the camouflaged coloration did 'take over' once I introduced a predator, so at least in theory it works, but I think it may need a bit of tweaking somewhere to make it happen more frequently 'in the wild', as it were.

By now I actually got a few species with some individuals that developed camouflage (usually visibility at 25% when in vegetation). I don't know how the creatures are programmed (do they flee towards cover?) but a part of the problem might be them not seeing each other for mating (and as a result they have to leave their cover which makes them visible for predators and renders their camouflage useless).

Maybe giving species a bonus on recognizing their own kind could help a bit, also they should preferrably stay near their cover type and move towards it when threatened - this would also allow to further compartmentalize a map into different mini-biomes with species endemic to one biome type (that would be pretty cool).

I like this idea and I always found that interaction with cleaner fish very interesting. It could also potentially link up with gameplay systems handling dead bodies and scavenging. I think it would be a good candidate for something to add while the game is in Early Access.

I'm really looking forward to scavengers. Crustaceans would be perfect for that - they could form an adaptive baseline scavenger mechanic so bodies don't pile up if no scavenger fish species are present/when there are only few scavengers.

Interestingly, this is very similar to what I was working on when it came time to start on the updated demo! There wasn't enough time to complete it so I put it all on hold (though evidently not perfectly, as someone on Facebook mentioned that one of their fish seemed to be firing electric arcs at others).

I didn't see that yet but now that I know it I will definitely keep an eye out for it.

p.s. Will there be Jellyfish?
Particularly species that live most of the time as polyps on the ocean ground and release mobile Jellyfish like every 30-60 minutes (which then spawn further polyps in random places) would be really cool. This would allow the game to get some sort of "seasonal" special events (coral mating season, jellyfish spawn season, etc.) and make it feel even more like a real ecosystem governed by larger cycles.

 

[person111]

I think jellyfish would be a great source of food for foragers while being foragers themselves. They would eat only plankton that you spawn in and would generate plant points for you. Now I don't know how that would go into Ecosystem's mesh randomizer because I see some very deformed things so it'd be best to apply only 2-3 types of mesh to the jellyfish. First is the meduca with it's flat like bell and tentacles surrounding the bell. And then there would be things like sea nettles and stuff that have very long tentacles. And then there would be the mesh that looks like the shape of Japanese mushroom jellyfish, upside down jellyfish and blubber jellyfishes. And then the animation. They would probably have some neural network because it'd be weird for polyp shaped things just floating around the water. Not a critical ones where they can think about how to avoid predators or something. Just some neural cells for the jellyfish to passively pulse. And there's the movement, since I wanna match the ones in game and the ones In real life I would recommend you doing the jellyfish so that it can only go up and down and can only move in other directions by the current, which is something you have already implemented. It would be best for jellyfish to spawn in cliff areas so they have things to grip on, polyps would be long cones with many parts and stuff. And then applying the parasite recommendation from Serafine (thank you for the idea), you can make it so that things like worms can attach itself to the jellyfish, or have the weakest prey fish in the game surround the venomous jellyfish for shelter. Most predators would stay steer clear of them unless they are the weakest type of jellyfish which is usually Medusas. And for the camouflage abilities. It would be cool if there's a fish that can spasm along the sand to mimic the rock fish or flounder fish. Anyways I see lots of potential in this game so keep it up Tom. Also for the jellyfish bloom season, it would probably happen once every Super-long-duration-of-time which creates tonnes of jellyfish. To counter the game crashing because jellyfish numbers Would likely reach millions you could reduce the number drastically to about 50-60 jellyfish and make it smaller. Jellyfish would grow faster than most fish since they just eat whatever goes into their mouths. And following the bloom they could make a few fish species extinct Because they swarm in big numbers and hordes. They could also poop out using their mouths subsequently feeding the foragers and scavengers. About the coral mating season, there could be an extra percent chance for corals to have mating season on rocky maps so the spores can land on and grip onto the rough surface of it. Another season would be fish bloom season where each fish has passively one boost in its body suddenly making it spawn more eggs. I’d recommend creating it so we could edit how the fishes mate in the emotion panel thing that you have shown in the trailer, the option for it would be named mate in environment then the one below it would be something like “Open areas” “Dark areas” “Deep spaces” . And it would be cool if there was a mating dance for the fish when it mates, not exactly jumping around but just being close to each other and basically mimicking the mating dance in real life. Because usually the ones with children die and then the baby fish just spawn out of nowhere seemingly at random. And I’d also suggest worms that can go under the soil like miky asked a few months ago, it would be great source of food for predators. Also, I think some scavengers might feed on dead plant parts like worms, or the worms feed on dead plant parts and then in turn get eaten by the predators. This creates a food chain.

 

[person111]

And I'd like to make a suggestion where there's carnivorous things that are immobile. AKA hydras that can feed on baby fish and makes the ecosystem more balanced so all the plants wouldn't get eaten all at once. Since there's filter feeders I'd recommend you to add jawless fish that feeds on zooplankton and plankton like in the past millions of years ago. That would be interesting. And there would be an extra map called deep valley or something. Animals living under there could have dead snow drift on the seafloor. To create the dead snow there would be animals that die of genetic disease or something then have the corpse decompose and then gets carried away by the current. It would go down until it reaches the Abyss area where there's no light, (enough light for us to barely see at least). They would swim around to the flakes of dead meat and gobble them up. There would also be coral and fish feces in there. Now the best way to implement this would probably have the dead snow in the plants section and list all the minerals that the fish can get from the food.

 

[person111]

here's a fish that goes after its target, plants, in a very quick way. I deleted one of the fins and copy pasted the other one by doing Control+V on the fishes body. I also made the tail longer and thinner, the fins I made a bit wider and longer so it could really be fast. It looks like a smaller version of a seal that lives entirely underwater. I’ve been enjoying this update a lot so far and it’s really amazing to leave the laptop open overnight and see the creatures swimming super fast or learning how to turn its body faster. I also like the flow simulation because it creates many different spot with different types of fish instead of a single place with fish and all the other ones lifeless and empty. Plant seeds are also cool, I can definately see a seaweed outbreak happening if the Kelp forest keeps growing bigger than it already is.

 

[Tom Johnson]

I really enjoy it, can't wait for the early access release.
By now I actually got a few species with some individuals that developed camouflage (usually visibility at 25% when in vegetation). I don't know how the creatures are programmed (do they flee towards cover?) but a part of the problem might be them not seeing each other for mating (and as a result they have to leave their cover which makes them visible for predators and renders their camouflage useless).

Maybe giving species a bonus on recognizing their own kind could help a bit, also they should preferrably stay near their cover type and move towards it when threatened - this would also allow to further compartmentalize a map into different mini-biomes with species endemic to one biome type (that would be pretty cool).

I'm glad to hear that you are starting to see some camouflage! The prey don't specifically flee to cover but that is something I've been thinking about working in. One difficulty is that by the time prey are aware of a predator, they are often so close that just going the opposite direction is the dominant strategy. This can be tweaked but it often has knock-on effects to predator-prey population stability that need to be double-checked, and so on. But I really like the idea of creatures staying near places where they can blend in, and would definitely like to allow the map to have mini-biomes with their own specialized communities, and I think this could be a good way of making that happen.

It's possible that mating visibility is slowing the development of camouflage, but in this case I like the idea that there is a selection pressure that may lean one way or another depending on the environment. My goal would be that if there are few predators, creatures might specifically evolve to stand out, while if there were many predators or very tight competition over prey, they would match their surroundings (or evolve something like bioluminescence to be able to choose when to broadcast their presence). Hopefully it will be possible to get this to work with some tweaking, but it may be possible that it isn't quite workable.

p.s. Will there be Jellyfish?
Particularly species that live most of the time as polyps on the ocean ground and release mobile Jellyfish like every 30-60 minutes (which then spawn further polyps in random places) would be really cool. This would allow the game to get some sort of "seasonal" special events (coral mating season, jellyfish spawn season, etc.) and make it feel even more like a real ecosystem governed by larger cycles.

Maybe! I always liked jellyfish, and I think the reproduction mechanic that you outlined is extremely interesting. I would also be very keen to make the ecosystem be governed by larger cycles, so I think there is a lot in its favor. The way jellyfish move is a little hard to simulate with the way the engine currently works and the mesh generator would likely have to be adjusted to account for them. I may not get around to them until after the Early Access release (and probably after crustaceans) but I have definitely added it to my list and saved that diagram to my ideas folder.

I think jellyfish would be a great source of food for foragers while being foragers themselves. They would eat only plankton that you spawn in and would generate plant points for you. Now I don't know how that would go into Ecosystem's mesh randomizer because I see some very deformed things so it'd be best to apply only 2-3 types of mesh to the jellyfish. First is the meduca with it's flat like bell and tentacles surrounding the bell. And then there would be things like sea nettles and stuff that have very long tentacles. And then there would be the mesh that looks like the shape of Japanese mushroom jellyfish, upside down jellyfish and blubber jellyfishes. And then the animation. They would probably have some neural network because it'd be weird for polyp shaped things just floating around the water. Not a critical ones where they can think about how to avoid predators or something. Just some neural cells for the jellyfish to passively pulse. And there's the movement, since I wanna match the ones in game and the ones In real life I would recommend you doing the jellyfish so that it can only go up and down and can only move in other directions by the current, which is something you have already implemented. It would be best for jellyfish to spawn in cliff areas so they have things to grip on, polyps would be long cones with many parts and stuff. And then applying the parasite recommendation from Serafine (thank you for the idea), you can make it so that things like worms can attach itself to the jellyfish, or have the weakest prey fish in the game surround the venomous jellyfish for shelter. Most predators would stay steer clear of them unless they are the weakest type of jellyfish which is usually Medusas. And for the camouflage abilities. It would be cool if there's a fish that can spasm along the sand to mimic the rock fish or flounder fish. Anyways I see lots of potential in this game so keep it up Tom. Also for the jellyfish bloom season, it would probably happen once every Super-long-duration-of-time which creates tonnes of jellyfish. To counter the game crashing because jellyfish numbers Would likely reach millions you could reduce the number drastically to about 50-60 jellyfish and make it smaller. Jellyfish would grow faster than most fish since they just eat whatever goes into their mouths. And following the bloom they could make a few fish species extinct Because they swarm in big numbers and hordes. They could also poop out using their mouths subsequently feeding the foragers and scavengers. About the coral mating season, there could be an extra percent chance for corals to have mating season on rocky maps so the spores can land on and grip onto the rough surface of it. Another season would be fish bloom season where each fish has passively one boost in its body suddenly making it spawn more eggs. I’d recommend creating it so we could edit how the fishes mate in the emotion panel thing that you have shown in the trailer, the option for it would be named mate in environment then the one below it would be something like “Open areas” “Dark areas” “Deep spaces” . And it would be cool if there was a mating dance for the fish when it mates, not exactly jumping around but just being close to each other and basically mimicking the mating dance in real life. Because usually the ones with children die and then the baby fish just spawn out of nowhere seemingly at random. And I’d also suggest worms that can go under the soil like miky asked a few months ago, it would be great source of food for predators. Also, I think some scavengers might feed on dead plant parts like worms, or the worms feed on dead plant parts and then in turn get eaten by the predators. This creates a food chain.

I like the idea of prey hiding among something venomous; while it may be a little while before I will be ready to add jellyfish, it might be possible to support an interaction similar to that of the clownfish and anemone a little bit sooner. I think mating dances could be quite interesting too; in the course of researching the game, I spent a decent amount of time researching essentially 'What do fish do all day?', and trying to impress potential mates is definitely a big part of that. I am hoping to add marine snow to the game as well, and also hydrothermal vents with chemosynthetic bacteria to help make deep sea environment more viable.

here's a fish that goes after its target, plants, in a very quick way. I deleted one of the fins and copy pasted the other one by doing Control+V on the fishes body. I also made the tail longer and thinner, the fins I made a bit wider and longer so it could really be fast. It looks like a smaller version of a seal that lives entirely underwater. I’ve been enjoying this update a lot so far and it’s really amazing to leave the laptop open overnight and see the creatures swimming super fast or learning how to turn its body faster. I also like the flow simulation because it creates many different spot with different types of fish instead of a single place with fish and all the other ones lifeless and empty. Plant seeds are also cool, I can definately see a seaweed outbreak happening if the Kelp forest keeps growing bigger than it already is.

I'm really glad to hear that and I enjoyed watching the video. It's especially nice to know that the new features I added over the past year made the experience better.

Thanks to both of you for the feedback and ideas.