Revamping Compound Clouds

I’ve had one idea simmering in the back of my mind for the past few years. It’s probably one I’ve debated back and forth to myself the most. It was most recently triggered by reading one of our top Steam reviews mentioning that plants basically cannot exist with how compound clouds currently work, which sent me down a rabbit hole of reading similar feedback from others (such as Thim’s feedback here).

That got me thinking, remember when oxygen and carbon dioxide existed as clouds in the game? There used to be a time where there were two whole extra compound clouds floating around that needed to be swum through and collected (though granted Phosphate didn’t exist yet). And it was originally inconceivable to remove them as clouds. But then we slowly realized we had to, and look how much it improved gameplay. I think moving them over to background environmental variables removed a lot of frustration from the gameplay and was all around a really positive decision we made towards the progress of the game.

Which got me thinking… what if we did the same to Ammonia and Phosphate?

Since these are both required for growth, it’s basically impossible for autotrophs to be sessile and grow since they will virtually never get any of these (the way currents currently work). If we make them background variables, then a sessile autotroph is actually possible, either for the player or for the AI to evolve, and that’s crucial to have in any ecosystem. For that reason alone, I felt it was justified to reclassify Ammonia and Phosphate.

And then I realized, wait a minute, if we removed oxygen and carbon dioxide clouds and that improves the game, and then we remove ammonia and phosphate clouds to improve the game, why stop there? Why do we have this instinct to keep the clouds? Could it be possible that the game could be improved as a whole if we just removed all compound clouds altogether?

I know this is a crazy thought, but hear me out! I’ve really thought about this from many many angles (being one of the original proponents for compound clouds back in the day), and I kept coming back to the same conclusion with more certainty. I feel like I have developed a good case for it at this point.

My Pitch

We remove all compounds as coloured clouds swirling in the environment. Instead, all of them exist as background variables that are slowly absorbed over time (based on your size and membrane). Here are my reasons why.


  • The initial reasons we implemented them no longer apply. I remember the original discussions that led to us deciding to add compound clouds to the game. We initially implemented them because we thought that at that scale, compounds are concentrated in pockets and need to be swam to. But as time went by and we did our research we realized that’s actually not the case, the “clouds” in real life are muuuuch wider, and veeeery gradually increase in concentration as you work your way from the edge to the center. Most cells don’t need to actively swim in circles to get to them. Additionally, we figured that we could have huge clouds moved by currents; but not only would that be taxing on performance, that would make the game’s visuals a rainbow mess, and honestly at that point what’s the point of colouring them (the whole screen will just be covered with bright colours you won’t even be able to see the background). Additionally, compound clouds were implemented at a time when the team had a very incomplete picture of what the Microbe Stage would look like and how we could make it fun, so it was a way to give something to the player to do, but now that we’ve added in more cells, organelles, AI behaviour, prokaryotes, engulfment, agents, patch maps, and new combat types, we don’t need that anymore.

  • Sessile play is currently not possible . This is the biggest one for me, and it was inspired by the Steam review. At the moment, sessile play is not possible since even if you evolve a biosynthesis organelle (a chloroplast or chemoplast or thermoplast) you still need to hunt for ammonia and phosphate. Even if you evolve a Nitrogenase, you STILL have to hunt for Phosphate (Nature doesn’t have an organelle that can produce Phosphate). And this doesn’t even make sense since most autotrophs in real life do not have Nitrogenase. In real life, the reason marine autotrophs get past this is because nutrients are highly dilute in water and can continuously be passively absorbed. Terrestrial autotrophs use a similar trick by simply having their roots dig into the earth to passively absorb nutrients, especially if the soil is damp which facilitates absorption more easily.

  • It removes the RNG frustration of not being able to find clouds, and makes low concentrations of vital nutrients much less frustrating. Patches with low levels of compounds should be completely playable, just more difficult. However, patches with low levels of vital compounds are instead just extremely frustrating with a high probability of death due to bad luck with clouds. Free floating Nitrate and Phosphate are supposed to reach very low levels once life matures on the planet (the way glucose currently reduces). If the current cloud system is used for them, this means many playthroughs where the player won’t find of these important growth nutrients in time and die, which is extremely frustrating and RNG based. On the other hand, if they are a background nutrient, they will always be absorbed at a steady rate, so the worst that can happen is just that you take longer to grow because the amount in the patch is lower. Much less frustrating for the player experience. And that’s actually what happens in real life when nitrate and phosphate are low.

  • It balances out the extremes of compound collection. Currently, as a small prokaryote, if you get lucky you can literally reproduce within 30 seconds or less. This is a shame cause we don’t let the player appreciate being a prokaryote. On the flip side, as a really large eukaryote, you will have to swim through so so many clouds before you can finally accumulate enough to reproduce (unless you are an engulfer, but not all cells are). In a system where compounds can be absorbed passively from the environment, we can prevent growth being too slow or too fast.

  • Hunting clouds is an inherently boring interaction compared to other interactions in the game. Hunting cells is always going to be more fun, or running away from cells. Or chasing for a free floating organelle before another cell gets it. Or looking for other cells to bind to, or sexually reproduce with (when that’s implemented). These interactions will all become more common if the player doesn’t have to hunt for clouds anymore.

  • It will make the game look a LOT more realistic . We’ve made some huge strides in making the game look more realistic in the years and we should all be proud of that. But there’s only so much you can do to make a game look realistic when there are bright purple and yellow clouds floating through the water. It’s even more visually jarring when there are multiple overlapping clouds that create a multi-coloured mess. By removing them into background nutrients, we can keep the brightly coloured clouds which are a very anachronistic with the game’s realistic aiming visuals, and which are inconsistent with the rest of the game because we will not represent compounds using brightly coloured clouds at any other stage.

  • Will help us improve performance . This is also a big one. Also, think about it, the compound clouds are such a computationally intense system, and they will literally only exist in the Microbe Stage. There will literally never be another stage where we will use that code. So is it really worth adding so much weight to the CPU for 1-2 hours of compound cloud generation?

What would they be replaced with?

Compounds will instead exist as background environmental variables, that are passively absorbed by the cell’s membrane. This is a defining feature of every marine ecosystem. Nutrients and other substances spread out and dilute in the water. This is what makes it so easy for organisms to absorb nutrients in water with little effort, whereas on land you cannot simply sit on a field and absorb nutrients from the air. As such, we can set the player’s cell a Diffusion Rate / Absorption Rate dependent on their size and membrane type, which defines how quickly they absorb compounds over time. It will also depend on the concentration of the compound in that patch, so the benefit of nutrient rich patches is that you grow faster and gain glucose (or its equivalents) quicker. This also makes low compound patches less frustrating, since the only penalty is that you grow and gain energy slower, not that 9/10 times you will not find enough clouds and just die. Furthermore, the addition of a Diffusion Rate trait allows us to create specific organelles that boost your Diffusion Rate (like Microvilli), which can allow multicellular colonies to evolve specialized Diffusion/Absorptive Cells. Diffusion/Absorptive Cells would then be the precursor of respiratory tissue, gills, and respiration systems in more complex organisms.

It allows us to replace clouds with chunks (bubbles, grains, crystals, etc.) which are more realistic, more visually relatable, and more easily discernible. These could serve as an analogue to the clouds, where swimming near them allows you to absorb a sudden large burst of that compound (until they deplete and disappear). This rewards species who evolve mobility. We could even have bubble environmental chunks for the atmospheric gasses, where swimming near them grants a boost to any process using that gas compound. These would also add more terrain to the environment, as they could be attached to, pushed around, navigated through to escape predators, etc. I remember being one of the ones who argued against representing compounds with chunks way back when, since I mistakenly looked at a few scientific images and thought that no such particles exist at that scale. But since I’ve began rethinking compound clouds, I researched bubbles, crystals, grains, and particles again and I found that many of these actually do exist at microscopic scales. There are videos of plant cells slowly producing oxygen bubbles as they photosynthesize, before the bubble detaches and floats away. There are grains of salt and sugar as small as cells. There are crystals of different metals and minerals that are the scale of cells. I think these could serve as a perfect replacement for clouds, they look better, they’re more realistic, they are less computationally intensive, and they fulfill the same role of rewarding mobile species for exploration and giving something for mobile cells to race towards. I can make a separate thread to discuss this further.

For perspective, here's a quick mock-up of what the environment would look like with no coloured clouds, and instead with compound chunks. Shown are some sugar crystals which will yield glucose (and shrink) when swum near. Also shown are bubbles (which look thicker microscopically) of Hydrogen Sulfide gas.


If I was not able to convince you with my reasoning, there are some alternatives I can think of to the suggested changes:

  • Only remove Ammonia and Phosphate clouds. If we feel removing clouds is too radical, but do recognize they create some problems. This will still enable sessile play, but loses out on a lot of the benefits I think we can gain from completely removing clouds. Plus, at this point shouldn’t we just accept the trend and remove clouds altogether?

  • Do not implement background compounds, instead replace compound clouds only with the introduction of Chunks in the environment for all compounds. This removes the need for a Diffusion Rate / Absorption Rate trait, but again misses out on a lot of the potential benefits listed above, and does not solve sessile gameplay being impossible.


So what do you guys think? Have I convinced you that removing compound clouds will make Thrive a better game? I know it’s a radical idea, but I don’t think we should be married to compound clouds as a feature if it’s actively bogging us down in so many ways, just because it’s been around for so long. This is coming from one of the original proponents of the compound cloud system from years ago. We should be ready to cut a game feature at the first chance we get with no shame if we think it can make the game better, even if its a legacy feature.

In my fork I’ve added a passive ammonia and phosphate generation. It was specifically for sessile species, but I’ve found that it also helps as a nice “pity mechanic” for when you just can’t find that one compound. I did NOT try removing those compounds from the map, although you are welcome to download the zip and change those values to zero in the biomes.json

My hunch though is that you’d get a very dull, and less intuitive game. If you watch the full length lets plays of Thrive, and count up the total seconds spent doing different things, hunting compounds easily wins as the primary gameplay loop over all the more interesting things we coded in. The editor is a small portion of play time, and while interacting with AI can be exciting, the species in the early game aren’t much to interact with and even later on meaningful interactions aren’t common until the late game when the player inevitably builds a toxin-shooting plant predator and starts murdering everything in sight.

I would never have thought this when I joined Thrive, but I think that there’s actually some visual magic in our simple, unrealistic, and frankly broken compound cloud system. Look at this video for example. New player, still getting oriented, sees one blob on the right of the screen and, without even saying anything, thinks to interact with it. Always having some blob around to seek out is a constant task when nothing more important is happening, and I think there’s a satisfaction to watching the little clouds getting sucked up.

Now maybe small, auto-breaking chunks can do the same thing, or even provide other benefits. It would definitely change the look-and-feel of Thrive, but I’m not sure that’s a bad thing. I guess my conclusion is that I think that compound clouds are way more important than they seem, and I want to see what the replacement really looks like before pulling them out.

I went down a similar rabbit hole of thought a while ago, though my conclusions were not as extensive as yours are! I remember the specific Steam review it was inspired by.

I personally do think we should stick with the compound cloud system and I view it as an acceptable abstraction of how microbes get nutrition. The fact of the matter is that I just don’t see how we can meaningfully represent microbial gameplay that is very fun without giving players the task of seeking these concentrated resources. We would completely be relying on cellular interactions to generate gameplay, and for the most part, cells don’t have the complex interactions that macroscopic animals have. Plus, the fact that Thrive is structured around these clouds means things are pretty hard to change at this point regardless.

What I do think we should do eventually, however, is repurpose the function of phosphate and ammonia, which I think would cover most of the issues you have listed. I made a concept based on that thought train a while ago and posted it on the community forums.

Essentially, the amount of time your organism takes to reach the editor will be based on its morphology, representing a passive rate of growth. Ammonia is repurposed to be a growth bonus. Depending on the amount of ammonia you have consumed, the amount of time it takes to get to the editor can be reduced by up to 50%. Phosphate is repurposed to be important for various cellular processes, such as toxins and health regeneration, which regenerate much faster if you have stored up phosphate. Having less than maximum health will slow down the amount of time it takes you to get to the editor by increasing this duration by up to 50%. We can also attach an auto-evo bonus if you gather a lot of ammonia and phosphate.

I came up with these numbers:

Baseline = 10 seconds. So as LUCA, it would take you 10 seconds to reproduce with full health and no ammonia inputs.

Cytoplasm = 2 seconds added

Prokaryotic Parts = 5 seconds added

Nucleus = 20 seconds added

Organelles = 7 seconds added

This makes it so that phosphate and ammonia can reasonably be neglected at the beginning of the game and as a very simple prokaryote, but eventually makes it so that they become more essential as time goes on.

Prokaryote with 2 prokaryotic components = 20 second reproduction cost at spawn, with 10-30 second replication depending on performance. What I assume will be the average cell after a single trip to the editor.

Prokaryote with 10 prokaryotic components = 60 second reproduction cost at spawn, with 30-90 second reproduction cost depending on performance. What I assume will be the average prokaryote a few generations before the addition of the nucleus.

Eukaryote with a nucleus and 15 prokaryotic components = 105 second reproduction cost at spawn, with 52.5-157.5 second lifespan depending on performance. What I assume will be the average newly formed eukaryotic cell.

Eukaryote with 5 organelles and 10 prokaryotic components = 115 second reproduction cost at spawn, with 57.5 – 172.5 second lifespan depending on performance. A eukaryote beginning to take advantage of its evolutionary benefits.

Eukaryote with 10 organelles = 100 second reproduction cost at spawn, with 50-150 second lifespan depending on performance. A eukaryote that has comfortably evolved towards specialization.

I thought of this more to address the fact that I thought the priority we should be placing on the player shouldn’t be searching for phosphate and ammonia - which I felt wasn’t a realistic representation of life - and instead places the emphasis on finding food. I also thought it would help the gameplay loop, allowing us to control the pace at which a player goes through the game as you deemphasize the importance of finding phosphate and ammonia. But I feel like this approach addresses most of the problems you list out. It is somewhat more realistic, allows for sessile organisms, and removes a lot of the random factors behind a successful lifespan.

Here’s the original thread if you want to read through a lot of reasoning and background thought on a similar topic to this thread: Reproduction and the Gameplay Loop - Microbe Stage - Thrive Community Forum

@Thim : Interesting, I may try that fork out later to see what it plays like.

Without having test-played a cloud-free environment, I would agree that simply removing clouds for only background absorption would take away a major gameplay loop of the ones we currently have in-game. That’s why I think if we were to remove clouds, we ought to replace them with a combination of background absorption, AND chunks in the environment you can interact with that give you a bigger boost of that compound at once.

I really think that chunks can fully replace what clouds offer in terms of our core gameplay loop, and in fact even offer more potential on top of that.

They are more easily relatable. I can see a white glucose crystal and immediately deduce “oh that looks like sugar”. I can look at a bubble and think “oh that’s a gaseous compound”. Iron and sulfur would appear as small mineral particles which would also be immediately recognizable. They are also more easily discernible if you have several of them in close proximity to one another. They also offer more interactivity, since you can bump into them, swim through them to evade other cells or trap cells against them, or in the future attach to them like some cells do. They’re also less computationally intensive since they are fixed 3D models, not fluid clouds that need to flow back and forth to show the currents they’re riding. The way they would be absorbed is that, so long as you’re within a certain vicinity of them, they slowly diminish in scale as they are consumed, and fill your appropriate storage. Eventually, they will disappear once fully consumed. In the future we can animate “death” animations that play as they shrink, with each different part “melting” one at a time. I can’t find the video at the moment, it was a video of a microscopic crystal dissolving. Here is the closest thing I can find to what it was (Ice Melt 7010 (Long time-lapse version) - YouTube). Just imagine this ice crystal as a microscopic crystal of some compound slowly dissolving as you swim around it to absorb it. For now we could use a simple scaling-down animation of the entire model.

I strongly agree with the replies posted so far, but I’ll add a few more points that haven’t yet been mentioned.

I don’t think we should put this much stock into a few Steam reviews. True, people bring it up a fair amount in Discord, but I think there are other solutions, and I doubt jumping to such a massive change immediately without trying alternatives will end well.

Actually cloud performance is pretty far from a performance bottleneck. As @hhyyrylainen has said many times: unlike microbes or chunks, the compound plane requires the same processing capacity no matter how many clouds there are, so the performance hit is constant throughout the game. And it’s a tiny hit compared to everything else.

My personal reason for keeping clouds is that they’ve basically become one of Thrive’s unique selling points. What other game has a fluid simulation going on right in the game environment with direct gameplay impacts? It’s one of the reasons I like the current gameplay. It definitely can be frustrating, but that’s a product of the spawn system more than anything else, and there’s plenty that can be adjusted there without removing clouds.

1 Like

Yeah no I wasn’t persuaded immediately by it. I had thought about the role of compound clouds many times before, and then when I read it it re-triggered those thoughts. It was only after I continued dwelling on it for a long time before I started to formulate reasons for and against and then eventually post here.

Just to note, although I offered two alternatives to my pitch, my original pitch is to replace clouds with passive absorption AND chunks. Background absorption would solve the sessile problem and give added purpose to a Diffusion Trait, and the chunks would exist in the environment and be absorbed by cells the way clouds currently do and fill the current “Clouds” gameplay loop. I really encourage you guys to think a lot about it. At first I was really averse to removing clouds too, because clouds were just so familiar to me as a feature of the Microbe Stage. But I couldn’t argue against the reasons that kept coming back to me, that I listed above.

However, I understand it’s a very big change. Though I personally am very in favour of removing clouds since I think it’s the best outcome, I get that the team’s support may not be there for such a change. If the support for such a change isn’t there, would people be up for any/all of the following middle-ground options?

  • Only removing Phosphate and Ammonia as clouds. Instead implementing them as background compounds (like Oxygen and Carbon Dioxide) that are slowly, passively absorbed over time. Glucose, Hydrogen Sulfide, and Iron would remain as clouds.
  • Adding in chunks for more compounds. Bubbles can be used for gasses. For example swimming near oxygen bubbles could boost the throughput of all your oxygen using organelles. Clusters of algae could slowly produce and release bubbles of oxygen. Crystals and minerals can be used for the other compounds. It could help us feel the benefits of what such chunks would be like.
  • Replacing clouds with chunks for some of the compounds. For example, we remove Iron as fluid clouds, and make cells absorb from the iron chunks directly. The iron chunks would slowly shrink over time as they’re collected from. This would be a good demo for what replacing clouds with chunks would feel like, and if we like it we could continue further.

I think these alternatives help solve some of the problems, and can give an appetizer of being one step closer to a no compound clouds game. Perhaps then, after we’ve played around in that environment for a while, we can get a better idea if going the further step and completely removing the clouds is the best move to make.

Furthermore, perhaps a good next step on this front would be to create a fork where we replace clouds with chunks, add background absorption, and play-test it. Then we can see what feels better.

I think many of us have grown very familiar with the look of the compound clouds, considering they’ve been around for years and years, to the point it has a fondness to it like a childhood toy. But at the same time I think we should recognize that they are currently holding the game from looking as realistic and good as it could (considering everything else in the game has been designed to mimick real life as closely as possible), and that they are anachronistic with the rest of game since coloured clouds will not exist in the 3D stage. It’s ironic I now find myself arguing against them cause I still remember 10ish years ago writing that crystals and bubbles don’t exist at that scale, which is why large coloured fluid clouds should be what should be used to represent the elements. It can be hard to put ourselves back in the shoes of a player and what clouds look and play like for them. But I think even if we don’t go all the way, even the middle-ground options can help solve a lot of the problems clouds create.

This sounds like moving back to the 0.2.x gameplay with the random emitters scattered around the place…

In general I don’t think it is a good idea to remove compound clouds from the game. This is the gameplay design equivalent of changing the entire engine the game uses.

1 Like

I think the most legitimate point which you bring up is that with the current system it‘s not possible to play as a sessile species, and that it should be possible because irl that is a very viable strategy.

As far as I‘m aware this

has been the plan for a long time. Dynamic compound levels are one of the big milestones after randomly generated patch maps. So by generation 10 or so there would be virtually no free floating ammonium/phosphate clouds as most of these compounds would be bound in microbes.

Here‘s my pitch:

As ammonium and phosphate slowly dispappear as concentrated compound clouds, they partially become dissolved environmental compounds. LUCA is a hetrotroph anyways, and so will be his first few descendant generations. By the time truly non-locomotive autotrophs will evolve, ammonium and phosphate will already be partially tied up in the biomass and partially be dissolved as environmental compounds. That way they will have a means of reproducing without swimming around to look for compounds.

The thing is, this passive absorption will be quiet slow. That way locomotive species will still be heavily encouraged to gather ammonium and phosphate by hunting other microbes.

The cherry on top of this approach would be a timelapse button for non-locomotive organisms. That way a player with a non-locomotive species wouldn‘t have to wait for much longer in between editor sessions than players with locomotive species. So the non-locomotive gameplay loop would be:

speed up time as long as there is no danger in sight to gather compounds → set time speed to normal when a predator comes so you can time your defense mechanisms correctly → enable timelapse again once the danger is gone

Note that the timelapse feature is only a facultative complementary measure to this approach. The concentrated cloud compound to environmental compound transition of ammonium and phosphate would also work without this feature.


I think here is a classical case in Thrive development when some design has been come up with like 5 years ago, but never there’s been a programmer who has been able to implement the idea leading to a situation where design work gets way ahead of what’s implemented.


As much as I’m looking forward to currents, do you really think that they’ll solve these problems on their own?
Finding enough ammonium and phosphate to reproduce can be frustrating in ressource-poor patches as is, even as a mobile species. When free-floating ammonium and phosphate clouds will dynamically decrease with the growth of the biomass, this will only be made more extreme.

Now imagine you’re an immobile autotroph in a patch which has low ammonium/phosphate concentrations. Waiting for clouds to be drifted towards you by currents sounds really frustrating to be honest, as it’s entirely based on RNG luck and there isn’t even something you can do to increase your chances.
Granted, a timelapse feature could also make this approach potentially less jarring. And while getting unlucky can be frustrating, getting lucky can feel more exiting than a passive meter just slowly filling up regardless of circumstances.

Another question regarding the currents in this context: Wouldn’t they also move the microbes themselves? So when a current pushes a cloud towards a microbe, wouldn’t it also push that cell itself away from the cloud?

My main criticism with time related reproduction is that it basically makes Thrive just a waiting simulator, just a single step away from asking for money from people to skip the waiting. Though now that I’m reflecting on that, currents would kind of also make it so, but I imagined the player would have some control over which currents to hop into and slight course corrections while they are carried by the currents. So there would still be gameplay and it wouldn’t just be a waiting simulator.

That criticism is very valid indeed. Altough it would mostly apply to autotrophs, as the passive absorption rate would be so slow that heterotrophs would still be very strongly encouraged to find food. They need to hunt either way to get glucose to survive, and the ammonuim and phosphate they get as a byproduct from hunting would very strongly outweigh that which they absorbed passively from the environment.

Still, I completely get that you’re very hesitant about this out of principle. It would definitely make autotroph gameplay even more of a waiting simulator.
To still make a counterpoint, speeding up in-game time to absorb nutrients faster will also increase the frequency of AI cells coming to attack you (frequency in a real life time context, if you get what I mean). That way it could feel more like a defense/survival simulator rather than a waiting simulator.
Although this only applies to autotrophs which actually have active defense measures (like slime or toxin ejection where you have to press a button) as opposed to passive ones (like spikes or simply a though shell).
But again, I can see why you still wouldn’t want to go down this road. My points are merely a bit of food for thought.

1 Like

I’m just going to throw my own input on the pile here;

I don’t think a massive rework is really necessary.

Without compound clouds, many autotrophs motile or not just have absolutely nothing interesting to do but just kinda idle around. So instead of removing every compound, my suggestion is to, if anything, give players an idle phosphate generation.

Why just phosphates? Because out of all the compounds, they are the only necessary one that you need alot of, but cannot create yourself via any remotely plausible means.

Ammonia would be exempt because you can produce it if you really feel the need to.

1 Like

I’d personally be wary of just implementing phosphate generation because the gameplay loop is already so volatile as is. It would help with patches where you don’t bump into a cloud instantly, but for the issue of going straight to the editor a few seconds after a prior editor session because you bumped into a highly condensed cloud, that problem would be exasperated.

Though I’m not convinced by doing away with clouds entirely, I am heavily in favor of restructuring phosphate and ammonia mostly for the purpose of standardizing the gameplay loop. I don’t think it’s good design to have the two compounds be the sole determinant for a trip to the editor; it can unnaturally lengthen the experience of a tiny simple prokaryote with limited function, and can unnaturally shorten the experience of a large eukaryote with a lot of function. Though I wouldn’t say this is something to change for right now.


I thought I had mentioned this in my OP but I’ll elaborate more here. Currents are one alternative that can help address the issues. Though I don’t think currents should not be implemented (they should of course be implemented), I think there are some drawbacks to relying solely on them to solve the above issues.

  • Sessile and less active/mobile organisms would still be inviable in low current or no current biomes (note that this affects you even if you play super mobile since the AI will be less viable as sessile).
  • The amount of Phosphate and Ammonia needed to reproduce can get so high (for example for a eukaryotic plant cell or some other larger/complex cell) that it would take many many waves of ammonia and phosphate getting washed into you to be able to reproduce.
  • If clouds of ammonia and phosphate are few and far between, they will wash over you so infrequently that it would be quite frustrating. On the flip side, if they are large and frequent, the environment will literally be awash in bright purple and orange and really ruin a lot of the good visuals of the game that we have built.

Reclassifying Ammonia and Phosphate as Background Compounds

There are a lot of topics for conversation in the thread, namely: Do we replace all clouds with chunks and background absorption? Do we only replace Ammonia and Phosphate? Do we Replace clouds with chunks but no background absorption?

So in an effort to focus the conversation, it seems the least controversial idea is potentially changing Ammonia and Phosphate into background compounds. It seems to me that out of all the suggested changes, this is the smallest change with the biggest potential impact. Additionally, implementing such a change would allow us to then play with it, reassess, and then decide if we wanted to take the further step and move other clouds into background absorption as well. If they were moved to background compounds, this is what it could look like:


Phosphate and Ammonia exist as features of the patch like Oxygen and Carbon Dioxide. However, since they are liquids (and not gasses), they can be stored, and so upon first spawning you will start to see a steady trickle of them flow into your storage. They will exist in different concentration per patch, and will fluctuate in response to natural processes that produce them and how many organisms are consuming them.

Realism Justification

We can say that Phosphate and Ammonia exist as such small molecules that are so diluted in the oceans that they do not appear as specific clouds, as opposed to Glucose, Hydrogen Sulfide, and Iron, which are scarce and concentrated enough that we represent them with clouds.

Rate of Absorption

The rate at which you absorb Ammonia and Phosphate from the environment is determined by their concentration in the environment, by your membrane type, and by any specific mutations you may have to increase absorption rate (like microvilli). Moving can also increase your absorption rate while moving, since motion facilitates fluid flow across membranes.

Other Sources of Ammonia and Phosphate

Cells will still yield Ammonia and Phosphate when they die. As such, there is a large incentive to evolve predatory or scavenging behaviour, especially if no other species have yet, since the cells around you represent untapped ammonia and phosphate stockpiles.

Nitrogenase will still yield a rate of Ammonia over time, and so it would speed up the rate at which you accumulate Ammonia. However, Nitrogenase will not be required to be self-sufficient for Ammonia, which is good because most autotrophs never evolved Nitrogenase.

How will Different Niches Play

Heterotrophs will still need to hunt for glucose, hydrogen sulfide, or iron, as without them they will starve and their ammonia and phosphate will be useless. As such the constant pressure of the risk of starvation is still present.

Autotrophs will need to find a way of producing their own glucose, hydrogen sulfide, iron, or some other energy source to be fully self sufficient. For example, a photosynthesizer would use chloroplasts to produce their own glucose. Once an autotroph has that, they are fully self sufficient. They will gradually absorb Ammonia and Phosphate from the environment, and they will have some way of producing an energy yielding compound for themselves without needing to move.

Other Consequences of Such a Change

This would effectively make growth a passive process. The player will need to hunt for food to stay fed, and make sure they stay alive from predators and environmental hazards, so that they can ensure the passive process of growth continues uninterrupted. This is exactly the gameplay loop of 3D Multicellular, Aware, and Early Awakening. Hunt for food and stay fed, and meanwhile your organism will slowly and passively grow. It’s similar even in an indirect sense to the core gameplay loop of Late Awakening and onwards, you supply your people with food and resources and their population will passively grow on its own and their prosperity increases.

However, this doesn’t make the time it takes you to grow completely out of your control. Remember that your absorption rate is affected by several variables, so the following are steps you can take to increase your ammonia and phosphate absorption rate:

  • Migrate to patches with lots of natural production of ammonia and phosphate (vents, tidepools or ocean surface before the Great Oxygenation Event, patches with high rock erosion (which is where phosphate minerals enter the ecosystem)).
  • Migrate to remote or underpopulated patches where there are less other species consuming ammonia and phosphate
  • Evolve your membrane or other adaptations to increase your absorption rate
  • Become a very mobile and active species so that you swim around a lot, increasing your absorption rate
  • Evolve to become a predator or scavenger, so that you can harvest phosphate and ammonia from other cells
  • Evolve a nitrogenase to produce your own Ammonia
1 Like

The smallest impact idea was to add a background rate of phosphate collection. Though even that requires some explanation to the player because otherwise we’ll get a flood of “bug” reports after the next release.

In terms of not entirely breaking the game’s design, If I had to pick my poison, I’d select @Deus’s idea of reproduction being mostly a constant rate that can be sped up somewhat by acquiring ammonia and phosphate. That mechanic wouldn’t be too hard to implement, but redoing the tutorials about reproduction would need careful consideration.

1 Like

Well similar to what Deus said above, I don’t think it makes sense to separate Phosphate and Ammonia by making one absorb passively and not the other, both in terms of logical consistency, and since then it would not fully solve the problem of sessile or low mobility organism viability.

For example, in such a case, all sessile autotrophs would require a nitrogenase since ammonia would not absorb passively, but nitrogenase is actually a rare mutation not often found in autotrophs. If both Phosphate and Ammonia were absorbed passively, Nitrogenase would rather serve to improve, rather than enable, sessile or low mobility gameplay, which is much closer to what’s observed in nature.

And lastly, if we did decide to move these to passive absoprtion, they would of course have to be very clearly communicated in the patch notes and have a mention in the tutorial so that the players would understand what the changes were and how they play.

1 Like

Good news is that I don’t think that changing reproduction is urgently needed right now so we have a lot of time to consider tutorials and fully go over the shift. We can focus on filling the game with more content and then decide when best to standardize the gameplay loop; unless there is enough push to change everything now.

I think going with passive growth will help fix a lot of issues, but if we decide to go with that change and if there still is a problem, then we might consider doing away with phosphate and ammonia.

You raise a good point, I don’t intend that all these changes be implemented in 0.5.9 of course, I just wanted to see if there was agreement that this would be an ultimate best case to develop towards. If we did agree that some degree of these changes would be beneficial for the game, then we could decide at what rates to implement them.