Endosymbiosis Concept - Feedback Based on Developer Forums Thread "Endosymbiosis Theory"

Note that most of this is based on Buckly’s contributions on the Endosymbiosis Thread in the developer forums. I just added a bit of feedback and detail which could make the concept make even more sense.

There are two aspects to this concept:

1.) How the player gets from simple engulfment to endosymbiosis. Before there ever was a mitochondrion, there was an organism which evolved into the mitochondria; and before there were cells able to hold in an organism, there were cells that were able to just simply digest another organism via phagocytosis. Endosymbiosis occurred as a biological phenomenon only when an organism became capable of holding in a cell without digesting them, only when the morphology of the cells allowed them to functionally benefit each other. I think it would be unrealistic if the player was able to undergo endosymbiosis from the beginning of the game, and I feel that it would be a bit of a lost opportunity for depth and realism if the player could just create an endosymbiotic link in just one generation. Special caution must be held in not making progression feel pointless.

2.) How the player can edit each endosymbiotic organism. Once two organisms were fully integrated with each other, there of course was a period of evolutionary time in which the cellular component was refined and edited to what would be most advantageous to the cell as a whole. It would be rather unrealistic in Thrive for an integrated organelle to immediately become highly functioning upon achieving endosymbiosis.

How Progression via Endosymbiosis could work.

  1. The player initially starts out as a prokaryote with the ability to simply engulf and digest.

  2. After some progression, the player will be able to unlock their cell membrane to allow them the endosymbiotic function. The first “phase” of this function is the ability to momentarily hold a microbe in for twenty seconds. For this length of time, the ATP consumption rate of the player is reduced by .25 times (factor can be changed to make it better) the net amount of ATP produced by engulfed cell. So, for example, if the cell you ingested has a net ATP production of 4 ATP, you spend one less ATP maintaining your cell.

  • This makes it so that the player is not completely dependent on auto-evo to progress. A big problem that came up with previous endosymbiosis discussions is the concern that the environment wouldn’t have a suitable candidate for endosymbiosis, essentially handicapping the player’s progress and capabilities. By straying from realism just a bit in the beginning however so that the player can only benefit from endosymbiosis, the player is both incentivized to begin endosymbiosis and isn’t really held back by auto-evo.
  1. The next few upgrades focus on increasing both the factor by which the player gets bonus ATP and the length for how long the player can retain onto this organism. In the interest of time and fun, there doesn’t need to be a large amount of small upgrades. I think each upgrade level should simply be from (20 seconds, .25) to (20 seconds, .5) to (40 seconds, .5) to (60 seconds, .75); so only 3 big upgrades in this series. The final upgrade in this series should eliminate the time limit and change the factor to 1.

  2. Eventually, an upgrade will give you a slot, hereby allowing you to spawn into the world with an endosymbiotic organism inside of you. Also lets you edit these micro-organism components themselves. Basically, endosymbiosis is finished up in a fun and interesting way with a series of only 5 upgrades to the membrane.

Let’s stop and explain how the concept of the slot should be thought of. Most of this is based on Buckly’s concepts on the endosymbiosis discussion in the dev forum, the only thing I edited was simply adding designations.

A slot basically is a designated unit which lets the game discern what exactly it is that you are trying to build. Different slots account for different specialized organelle types; for example, there would be an energy producing slot (mitochondria), a photosynthetic slot (chloroplast), a poison slot (oxy-toxy), a bioluminescent slot, a rust slot, a chemo slot, and any other slot which pertains to the function of an analogous prokaryotic component.

A player only has a single slot for each designated function – that is, only one energy slot, per say – and the player can only designate one micro-organism to fill each slot. When you place an ingested organism into a slot, only the components corresponding to each slot can be placed on the micro-component; so for example, when you place an organism in your energy slot, you can only place/edit metabolosomes when editing that slot. Each different slot also has a progressive osmoregulation increase penalty after a certain size to help limit exploitation of making a cell giant by making organelles massive; for example, energy slot components with 3+ tiles being used have a 25% increase in osmoregulation cost for each component, with 4+ tiles have a 50% increase in osmoregulation cost for each component, etc. The bracket for applying this progressive osmoregulation cost increase can vary between components; for example. Penalties could also vary; with photosynthetic organelles for example, components with 4+ tiles could begin to have movement slowed down by a progressive factor. When the player is a prokaryote, they can only have an energy slot and one other chosen slot designation. The nucleus will allow you to have all other slots available.

Each slot has an advantage to it that incentivizes the player to develop organelles. Note that all components have the benefit of basically being a bargain for what they offer you. For example, an organelle with three metabolosome components priced 45 would be much better than buying three separate metabolosomes.


These are a few examples demonstrating what each slot could be.

Energy Slot
Each metabolosome within the organelle membrane generates 1.2 times more ATP with the same amount of glucose as an individual metabolosome would. Also unlock more upgrades for metabolosome components in organelle compartment.

Progressive Cut-Off Penalties = 4 tiles – 25% increase in osmoregulation cost of each component, 5 tiles – 50% increase in osmoregulation cost of each component, 6 tiles - 100% increase in osmoregulation cost of each component.

Photosynthetic Slot
Each thylakoid within the organelle membrane generates 1.3 times more glucose with the same amount of sunlight than an individual thylakoid would. Also unlocks more upgrades for thylakoid components in organelle compartment.

Progressive Cut-Off Penalties = 4 tiles – Movement slowed by 10%, 5 tiles – movement slowed by 25%, 6 tiles – movement slowed down by 50%, 7 tiles - movement slowed down by 100% percent.

Vacuole Slot
Cytoplasm will make up the inside of the membrane. Every unit of cytoplasm within this slot’s membrane will have twice the storage room of a regular cytoplasm unit outside of the component’s membrane.

Progressive Cut-Off Penalties = 4 tiles – osmoregulation cost for each component increased by 10%, 5 tiles- osmoregulation cost for each component increased by 25%, 6 tiles – osmoregulation cost for each component increased by 50%, 7 tiles, osmoregulation cost for each component increased by 100%.


These are the slots I have thought out as of now; I want to make sure the basic idea behind the concept is sound and reasonable before devoting more thought to it. I wrote this while being exhausted, so if there are any inconsistencies or odd omissions, please bring them up. I don’t know how Buckly came up with most of this: it’s ingenious yet exhausting to work out.

What are some thoughts concerning these ideas?

2 Likes

I mostly disagree. I think the player should be restricted by auto-evo. He must use whatever is available in his environment to thrive, just like in reality. If there is no proto-mitochondrion near him, then he must be creative and use what is available to solve his problem.

I greatly dislike this idea of “slots”. These are arbitrary groupings of organelles which the game must track. This also arbitrarily restricts the player. Why can each “slot” have only one organelle? What if the player wants to have both rusticyanin and mitochondria? If his organism can support it and the environment allows it, then he should not be kept from it. Imagine if KSP arbitrarily restricted the player to one engine type per rocket. Such a restriction would be needless and hamper both creativity and gameplay, as all player designs would be less versatile. These “slots” would make the game less of a sandbox and more of an arbitrary waiting game while the player seeks to fill all his “slots”. This would also make every organism feel too similar, as all organisms would have the same arbitrary “slots”. The game should be indifferent to whatever organelles the player (or AI) has. Evolution is blind, so the game should be blind too.

5 Likes

I also don’t like the slots concept. But the solution of not having useful microbes to make symbiosis with, I think is pretty good. My biggest problem with the endosymbiosis is that if we only have it as an option for the complex organelles, then a player whose planet has really useless other species is going to have a bad time. So instead of gradually being able to build up the organelles in your cell, you would be basically just praying to RNG jesus to let you proceed in the game.

But there is no need for RNG if a goal can be accomplished in many ways. If there is no proto-mitochondrion, then the player can use rusticyanin or thylakoids (or thermoplasts?) instead. That way, the player can reach the multicellular stage regardless of whether any particular organelle is available. If there are enough ways to generate energy for multicellular organisms, then the probability of being on a planet with no possibilities will be negligible. I don’t think he should have to wait for a particular organelle unless he has a specific plan.

2 Likes

That’s the problem, because I imagine that people will do a one “whatever goes” run of Thrive, and after that they will want to do something really specific. So randomly locking different paths through the game wouldn’t be good.

3 Likes

Perhaps a world generation setting that defines what endosymbionts are guaranteed to spawn in the world, for those who want a specific path.

2 Likes

To add more points against complete dependence on auto-evo; I’m pretty sure I saw discussions somewhere which basically stated that prokaryotic compounds were going to be nerfed a bit once upgrading is implemented, or atleast, eukaryotic organelles are eventually going to be more efficient than their prokaryotic counterparts. Right now, there isn’t a huge difference between the output of a metabolosome and a mitochondrion, so we must be mindful of that when planning a concept. And like hhyyrylainen said, I think it would be rather annoying if a player wanted to do something but couldn’t not because they weren’t skilled enough yet, but because their planet wouldn’t let them in the first place. Many restarts in frustration would be prevalent during the microbe stage.

With this feedback however, I think I might have made a worthwhile concept. Let’s begin by focusing on the slot concept.

The biggest reason why this interpretation of the slot mechanic isn’t very “Thrivey” seems to be because of the artificial designations - i.e. an energy slot, poison slot, etc. - limiting what the player can do instead of basing limitations on natural barriers the player would have to overcome. I sent this concept to @Buckly as well, and he mentioned this same exact thing; in a game about nature, it doesn’t make much sense to trap the player based on limits based on unnatural forces which the player has no real way of overcoming. In essence, why can’t a player place what they want on an energy producing organelle? Why must it only be a metabolosome? The artificial constraints placed on each caste of slot is the big turn off, it seems.

So, here’s my revision.

We keep the slots, but forget about having artificial restraints keeping you from placing what you want in the slot. Instead, there will be a mechanic which basically determines the output of an organelle based on what micro-components you designed it with and each component’s interactions with a neighboring component. This doesn’t really make sense without further explanation, so hear me out.

First, let me explain why I’m keeping the idea of the slot. I simply think that a slot the best way to answer a lot of the harder questions concerning endosymbiosis, like:

  • How can we designate which ingested organism we want as our base?
  • How do we organize endosymbiosis in a way so that a player could have multiple endosymbiotic organelles at once?
  • What would the point of a nucleus be?

The answers:

  • You simply choose one organism from a selection of say the last 5 engulfed organism to correspond to a slot.
  • We have multiple slots so the player can edit each independent organelle.
  • A nucleus lets you have multiple slots, and opens up an interesting organelle upgrade path for the nucleus. A prokaryote shall only have one slot, but the nucleus will give you two slots when you unlock it. Organelle upgrades for the nucleus can focus on adding more slots.

So basically, in this new concept, a slot can be thought of as an empty organelle button that can be filled with your created organelle.

Now, with that out of the way, how will creating your organelle actually work?

The endosymbiot editor is basically an editor where you can place the prokaryotic micro-components, and those micro-components would be encased in an organelle membrane and then would act as the part “unit”. For example, you can place a thylakoid, a metabolosome, and a nitrogen fixating thing (I forgot the name) in this editor, and when you go to your organism’s editor, that will be the “part” which you could place in your cell. There’s a flat bonus applied to each component which make it 1.3X more efficient than a free-floating micro-component.

Here’s where it gets interesting.

Along with the flat bonus applied, every micro-component has an adjacency bonus on any neighboring tile; that is, each micro-component within the organelle membrane has an effect on another component which can either strengthen or weaken the organelle as a whole. Each micro-component can have a unique interaction with another type of micro-component; one micro-component could have a negative interaction with another while having a positive interaction with yet another micro-component.

This is best illustrated with a few examples that can get across the point.

Let’s say that I go into an organelle editor trying to create an energy producing organelle. I place a metabolosome, which already has a 1.3X bonus simply due to being in a membrane, and place another one next to it. The adjacency bonus I receive as a result of that is a +20% production bonus in the two micro-components. I can add another metabolosome which can further compound this effect by having a bonus adjacency with the metabolosome right next to it. Satisfied, I exit the organelle editor, and place my energy creating organelle in the cell which consists of three metabolosomes.

Another example: I have an organelle with one metabolosome in it, and I want to try and experiment a bit. I add a thylakoid to the editor, and realize that while I get a 10% bonus in the production of glucose within my thylakoid from the metabolosome, there is also an adjacency interaction between the thylakoid and metabolosome which makes the organelle heavier by 30%, which slows my organism down. I realize that this isn’t really worth it for my playstyle, so I just decide to replace the metabolosome with another thylakoid, which simply compounds the glucose producing effect at less of an expense to speed.

We could combat a player creating supersized organelles by having the adjacency effect weaken after a while and have a progressively increasing osmoregulation cost.

Does this make sense? Think of it a bit like creating a city in Spore, where the happiness/productivity of a city depends on the layout of you city. Or think of Civ 6, where placing districts next to features or other districts could either damage or enhance the productivity of your city. If this makes sense, do you guys think it’s a better concept?

2 Likes

I don’t like this either. Your rework of “slots” made them synonymous with hexes. In Thrive’s current state, each hex is something of a “slot” which is filled with an organelle. As such, this idea of “slots” is moot, as it is already in the game.

I disagree with the idea of adjacency bonuses. Once again, this is arbitrary. Such a bonus made sense in Spore because cities are large and transporting things across them can be difficult. Cells, however, are tiny, and the distance between each side of the cell is miniscule (excluding abnormally large cells like stentors), particularly in prokaryotes. If something must be moved from one part of the cell to another, it can be done rapidly and with little effort. As such, no location in the cell is better than another (unless an organelle requires contact with the environment, in which case it must be on the cell’s edge). Adjacency bonuses would just be arbitrary and based on a two-dimensional hexagonal grid, which is not how real cells operate.

However, adjacency bonuses do make sense for larger organisms with complex organs. Blood, for example, must be transported around the body via a network of blood vessels. Proximity to these vessels is thus beneficial. So your idea isn’t completely unsuitable for Thrive, but it should not be implemented in the microbe stage.

I think you didn’t fully understand the idea of a slot.

A slot isn’t a tile; a slot is simply a part button in the editor that hasn’t been filled with anything yet; in other words, it’s a blank space on the parts screen that is waiting for a part to be created. I made a very rudimentary concept showing what I mean:


A slot is only there because it helps you assign an endosymbiot to be a part. Within the slot, there is an editor where the adjacency bonuses apply to micro-components within the organelle membrane; adjacency bonuses aren’t applied to entire organelles or the cell body itself.

What makes organelles so efficient and essential for the cell is that important things are compartmentalized to a single faculty within an organized micro-structure, so yes, placement does in fact matter within a cell and an organelle. Organelles are very potent for a cell because they contain within them all the essential proteins, enzymes, molecules, and structures that help facilitate a need of the cell. Let’s take mitochondria, for example. The mitochondria generates so much energy for the cell because of how each structure interacts with an adjacent structure within the membrane of the mitochondria. A vast amount of energy is generated because electrons get transferred along a series of structures and functions; if just one of these structures is misplaced, damaged, or otherwise incapacitated, the entire system goes down. If the molecule FADH, which helps transfer the source of mitochondria’s energy, is not based or found in the organelle, the entire system goes down. If something is screwed up, the entire thing is screwed up. This applies to chloroplasts too; if the light receptor proteins are down, if the components of the Calvin Cycle are down, if the membranes can’t facilitate well enough, then the productivity of the entire component is shut down. Everything within a micro-component – all the enzymes, proteins, structures – must interact positively with each other, or we would have a dead cell on our hands.

Would adjacency bonuses within a slot not represent this in a somewhat realistic and yet fun way?

2 Likes

Ah! I understand now. That image helped. I still don’t like it, though. Why should the player have a limited number of potential organelles? Why not allow him to have as many as he likes? Moreover, endosymbiosis will presumably be quite rare, so the player probably won’t be able to do it more than 2 or 3 times anyway. As such, limiting the number of endosymbiotic organelles to 2 or 3 is unnecessary.

I think either I misunderstood you about adjacency, or you misunderstood me. Adjacency is certainly important within an organelle, but I thought you meant there would be bonuses to organelles if they were placed next to each other. Yes, if the parts of a mitochondrion are separated and scattered around the cell, they certainly would not function as a proper mitochondrion. What I meant is that the location of mitochondria in a cell is irrelevant. The mitochondria can be clustered in one area or scattered throughout the cell, and hardly any difference is made. Of course, this depends on the cell. Some are particularly large or specialized and require specific arrangements, but for a typical cell (e.g. paramecium) this is not an issue (again excluding organelles that must contact the environment).

If the player designs endosymbiotic organelles with microcomponents, then he would require another editor within the microbe editor. I think that would be overcomplicated.

2 Likes

This is starting to look like what the multicellular editor would do regarding different cell types, in that you separately edit them, and then on a main view you would position the cell templates to make up your creature.

3 Likes

Is this meant as a bad thing? In my view, more similar mechanics between stages will simply serve to make the game as a whole seem more coherent.

3 Likes

I rather like the idea now that it has become more polished, Sure there might be some greivances with things such as limited slots for organelles but that’s hardly a big problem, I would like to think the nucleus would remove the limits to the number of slots while prokaryotes or perhaps proto-eukaryotes are still limited in some fashion.

Alternatively, some type of “soft” limit could be enacted for prokaryotes not unlike the dreaded complexity meter of Spore. This limiter could quantify the total amount of parts you are using for all your symbiotes regardless of number, and cap that, or perhaps it could merely be a limit to the size of your symbiotes. Or perhaps we could simply make prokaryotic organelles be less effecient, or require more resources for reproduction. Of course, these restraints could be broken by the evolution of a nucleus etc because nobody likes limiters like that.

Having an editor inside of an editor is certainly complicating, yet all the same it is a very appealing concept to me. We could perhaps have the game automatically cut your endosymbiotes down to something useful, so it would not be a requirment to use such an additional editor unless you wanted to really optimize.

Ultimately the concept of endosymbiosis as a whole is likely going to include a fair deal of trial and error. It’s a pretty interesting feature that has a great deal of potential, and I doubt we will get it right the first time. So by all means keep this up I love seeing what everyone thinks.

1 Like

I don’t understand why you would want to implement the arbitrary restrictions despite knowing that they are unfun and undesired, even if the restrictions are only for prokaryotes. Regardless, I want to see concept art for this editor-within-an-editor.

As we both know, in prokaryotes, genetic information isn’t very organized; it basically floats around loosely near the center of the cell. Thing is, organelles require a lot of genetic instruction to ensure that every component is properly functioning; there are also a lot of other factors and unique complex structures, such as transposable elements, that make complex structures very hard to genetically organize without some form of compartmentalization. This is what makes the nucleus so important; it holds everything that a complex organism would need in a single spot of the cell, whereas a prokaryotic organism would risk a huge list of potential complications that a nucleus-bearing organism wouldn’t really have to worry about because the genome is loose and free-floating within said prokaryote. The nucleus is very important for the existence of an organelle; we can’t just ignore that in-game.

So, restricting the amount of organelle slots a player is able to have not only can increase fun by increasing depth, but is also based on a well founded biological fact.


So, if everyone thinks the adjacency bonus idea sounds good so far, I could try and brainstorm how some interactions between microcomponents could work. Very special care must be taken in this process however, as we need to make sure that we don’t accidentally allow the creation of an extremely powerful organelle; if we make an interaction between a thylakoid and a metabolosome basically create an endless supply of energy without any drawbacks for example, all hell would break loose.

With this in mind, what are some possibly game-breaking interactions which you guys think should be accounted for in balancing interactions between micro-components?

1 Like

I don’t think I ever wrote that the adjacency bonus sounds good, but whatever. I accept your point about nuclei being needed for greater complexity. Now that you describe it that way, I also think that prokaryote complexity should be limited. Although this idea still seems a bit arbitrary to me, it is ultimately moot. The player will always become a eukaryote on the way to multicellularity, so restrictions on prokaryotes will only affect the player for a short time anyway.

In Thrive’s current state, chloroplasts and mitochondria can be combined in the tide pool’s 200% lux to create infinite free energy. Clearly this should change. If non-LAWK things like thermoplasts or audioplasts are added, they too will need to be reasonably restricted. I think rusticyanin might also lead to problems, as it bypasses glucose and thereby could make many cell processes unnecessary. If there were enough iron, a Thrive cell could use only that to power flagella, cilia, etc. This could be solved by limiting iron’s availability or reducing the effectiveness of rusticyanin’s ATP production (in accordance with its capability in reality, of course). Earth is not covered with large ferrovores, so there is obviously some major restriction which makes siderotrophy unviable. Perhaps this results from iron’s limited availability due to nearly all of it in the oceans having been oxidized by the photosynthesizers. I am curious about how a world without such photosynthesizers would work, though.

Game-breaking interactions are ones that allow endless energy, as you noted. I don’t understand cell biology well enough to know why plants should be incapable of locomoting like animals, but clearly the combination of chloroplast and mitochondrion does not eventually create an intelligent species. For whatever reason, a better strategy for them is to remain sedentary. As always, I recommend looking to nature for guidance. As Buckly noted, the first implementation will likely be poor, and exploitation by players will reveal its weaknesses. As such, I recommend implementing a simple and obviously insufficient system now that can be improved into a proper representation over time. Whatever gamebreaking combinations are possible in the first iteration can be fixed as the system is refined.

2 Likes

I have brainstormed a few ideas for how vacoules could interact with other microcomponents; as one could infer, their main effect relates to storage.


(And I just realized I was spelling vacuole wrong)

In essence, you could either choose to specifically and significantly expand storage for a single compound by attaching a vacuole to the component that uses said compound, or could expand general storage for all compounds by pairing vacuoles with other vacuoles, albeit at a less significant factor and with a hit to energy.

A big design concern with vacuole interactions is making non vacuole-to-vacuole interactions desirable. Why would a player want to only add space for a specific compound, for example, if they could just create a large vacuole to vacuole organelle that adds around the same storage for not only that specific compound, but every other compound? That concern is the reason behind my incorporation of the energy cost in the vacuole-to-vacuole interaction column. The amount of storage added for each component, as well as the energy cost factor, could be tinkered with to help alleviate this concern.

Another concern I have has to do with nitrogenase-vacuole interactions, which I notably left blank. Because adding storage to ammonia doesn’t really help the player, it’s a bit difficult for me to think of a benefit that is uniform and analogous with other vacuole interactions. I might need a bit of help with that. A question to consider: Does every micro-component need an interaction with another micro-component?

1 Like

I wouldn’t think so. Trying to get something to work just for the sake of everything being the same isn’t smart. Just having the vacuole interaction with certain organelles adds depth to the game.

1 Like

Alright, sounds good. I’ll roll on from now leaving some interactions empty, and if it becomes apparent that we do need there to be interactions among all micro-components, we could focus brainstorming on that.


In the interest of increasing simplicity while decreasing unnecessary complexity, I think it would make sense if the effect each micro-component has on another micro-component is consistent through out each interaction. So just like in vacuoles where the benefit to be gained always has to do with space, the benefit to be gained from attaching any other micro-component should always have to do with a single positive. Having each interaction be unique - that is, having a vacuole-thylakoid interaction be fundamentally different from a vacuole-metabolosome interaction - complexifies the game to an unnecessary extent, I feel; all those interactions would be very difficult to keep track of and account for. Simplifying each interaction eases accessibility while still increasing variety and depth. There’s also the fact that editing organelles isn’t going to be completely necessary to endosymbiosis, so I don’t feel there should be an extremely complex system under it.

So, with these ideals made clear, I have an idea as to what nitrogen-based micro-components could offer to a player.

Nitrogen is very important for life as we know it, as it serves an essential part in DNA and protein processes, and ultimately, growth. Because of this, I think that attaching a nitrogen-fixating micro-component within an organelle should reduce the phosphate/ammonia cost to growth that the organelle adds onto reproduction, hence reducing the need to reproduce a tiny-bit, maybe by around 5%. I feel that this effect is subtle enough to not warrant a unique drawback to nitrogen-fixating micro-component interactions, as the costs of an increased energy demand that comes from adding a micro-component and the opportunity cost of going for reduced growth needs instead of increasing the productivity of your organelle should be enough of a natural penalty.

It is also important to note that nitrogen is very important for the process of creating chlorophyll, so perhaps the growth reduction effect could be strengthened if the N micro-component is attached to a thylakoid, or perhaps there could be a reduced energy maintenance requirement added to the growth effect.

So, this is what we have so far.

Vacuoles - When attached to other vacuole, adds a small amount of space for all compounds, although at a cost of an increased rate of energy consumption.
When attached to other micro-components, adds a significant amount of space for a certain compound without that hit to energy consumption, although that space is only added for that individual compound.

Nitrogen-Fixating Component - When attached to self, increases the strength of ammonia generation at a given Nitrogen concentration.
When attached to other micro-components, reduces the added ammonia/phosphate reproduction cost of the organelle. Effect is magnified when paired with thylakoids.

1 Like