suggestion Items Buoyancy

[Maris]

I have a great idea!

Although it's not new, there was even a mod for this (Floatation), but that mod hasn't been updated since 2022 and no longer works.

The idea is to give items buoyancy! For example, Wood should float in Water; solid Igneous Rock should float in Molten Gold; Ice should float in Water (just like in real life), etc. This would be an amazing addition to the game within the familiar PHYSICS framework.

The game already simulates density-based layering for liquids - lighter liquids rise above heavier ones. Extending this principle to solid debris feels like a natural next step. You could design schemes for "transporting" items from bottom to top or side to side. I think this would be a great addition to the game!

Motivational picture:

[gigamoi]

While possibly fun, I fear adding such a mechanic at this point in the life of the game may break many existing designs (and ruin running play-throughs as a result).

[Maris]

1 hour ago, gigamoi said:

may break many existing designs

That's a fair concern. However, most existing designs don't rely on debris staying at the bottom of a liquid pool as a core mechanic. It's usually just an unintended nuisance.

That said, if breaking existing setups is a real concern, this could be introduced as part of a DLC rather than a base game change, that way it remains opt-in and doesn't affect anyone's current playthroughs.

Could you give a couple of examples of designs that would break?

[gigamoi]

2 hours ago, Maris said:

Could you give a couple of examples of designs that would break?

Anything involving a weight plate submerged in liquid with now floating debris. I frequently submerge setups meant to handle stuff that may off-gas like oxylite. I may do so with some of the densest liquids like oil.

[asurendra]

Hm... What this should change? Like, debris floating. Cool. What next? How exactly we can use it?

[Maris]

1 hour ago, asurendra said:

Hm... What this should change? Like, debris floating. Cool. What next? How exactly we can use it?

Here are a few use cases that come to my mind:

1) Passive item transport - debris floating on the surface could be carried by liquid flow, both vertically (bottom to top) and horizontally. A new way to move resources without conveyors.
2) Density-based filtering! Heavy items sink in a given liquid while light ones float. You could separate mixed debris by choosing the right liquid.
3) If floating debris could get caught inside a Mesh Tile from below, it would enable automatic resource harvesting from liquid surfaces.

I want to emphasize, I'm not suggesting making the game EASIER. I'm suggesting making it more INTERESTING. This means new problems and minor inconveniences that players would need to figure out how to deal with.

 

1 hour ago, gigamoi said:

I frequently submerge setups meant to handle stuff that may off-gas like oxylite. I may do so with some of the densest liquids like oil.

Thanks, that's a good example! I think there are two possible game design approaches here:

1) Make sublimating resources never float (prioritizing backward compatibility).
2) Require the player to choose the right liquid (prioritizing interesting gameplay!) - not oil, but water or something even less dense.

Either way, the player gets to choose whether to enable a new DLC or not. And you can't enable it on an existing save anyway.

Edited February 19 by Maris

[LeandroBraz]

5 hours ago, gigamoi said:

While possibly fun, I fear adding such a mechanic at this point in the life of the game may break many existing designs (and ruin running play-throughs as a result).

I think that's a bad reason to avoid implementing something. Coming up with these designs is part of the fun, so having to rethink them to accommodate a change shouldn't be seen as a problem. If anything, it'll force people to rethink how they do things and lead to new designs. 

Ruining runs, as long as it doesn't happen often or without warning, shouldn't also be an obstacle, mostly when it's possible to keep previous versions available through steam. 

 

With that said, I don't think buoyancy would have a huge impact, unless Klei adds some buildings that make use of it. Not much would change on how things are done. It seems more like a cool detail than a meaningful addition. It might not be worth the headache.  

[KittenIsAGeek]

7 hours ago, Maris said:

Could you give a couple of examples of designs that would break?

ONI physics isn't quite the same as real physics.  

Somehow the dupes managed to compress 10 tons of ice into an area smaller than 1 ton of water will fit.  So I imagine that unless the code evaluates the total mass of the object and its material density, more physics will break. =^.^=

[Maris]

Right, ONI physics has its quirks. But buoyancy should logically be based on element density, not MaxMass per tile. Those are different things. And if some edge cases don't work out, Klei can always hardcode exceptions. It's not like every interaction needs to be perfectly realistic, it just needs to be consistent and useful enough.

On 2/19/2026 at 4:01 PM, LeandroBraz said:

unless Klei adds some buildings that make use of it

Wow, great idea! I imagine floating buildings, maybe boats or something similar. Maybe floating ice tiles, like icebergs.

I don't believe ALL of this will make it into the game, but I hope so. Either way, it's not up to me to write a GDD, Klei will do it their way.

[ThePyro]

Cool idea! You could have a full-on river running horizontally across the entire map (or even diagonally!) and send a stream of solid resources down it. It would act as a pipe for solids - like a conveyor belt, but for massive bulk quantities. In real life, the heaviest cargo is shipped by water, so why not use a similar mechanic in the game?

There was a discussion above about the CPU load, and that worries me too. What if we limit the travel speed of floating resources but remove the mass dependency? So a ton of ice floats in a kilogram of water - so what? For example, resources could sink at 1 tile per second, float up at 0.5 tiles per second, and drift horizontally at 0.1 tiles per second. This would encourage players to send large volumes at once, rather than small, frequent shipments.

[CrutchPunk]

Honestly, I'm all for breaking existing setups if the addition benefits the overall game. It keeps the challenge alive and the game fresh. 
I absolutely loved the flotation mod...with it being out of date/abandoned, it's a perfect excuse to bring buoyancy to the actual game! 

[Maris]

Maybe there could be some buildings for WIND. Some existing ones:

And by default there is no wind, so no horizontal movement until something happens like an item or liquid dropping onto the surface.

[ThePyro]

Next step – flying solid resources! We already have gas layering mechanics in the game (light gases rise, heavy ones sink). What if we could attach a chunk of Granite to a bunch of Hydrogen-filled Balloons? They'd float up! Wes would be thrilled to see his balloons used like this 🎈🎈🎈

[Maris]

On 3/3/2026 at 12:14 PM, ThePyro said:

Next step – flying

Yes, and there already exist Flydo and flying creatures, so wind could affect them too.

Come to think of it, this could be a broader concept - WIND.
It affects both things floating in liquid and things flying through gas.

Similar to other buildings, a wind-generating structure could have a defined area of effect where the wind occurs. And similar to other areas of effect (like an Auto-Sweeper), this zone wouldn't pass through obstacles. Anyway, there's a lot to think about here. Could turn out really cool!

Just imagine, wind affects flying creatures. They shall not pass!
How many new designs will this create?

[gigamoi]

4 hours ago, Maris said:

Come to think of it, this could be a broader concept - WIND.

Would flatulent dupes affect this system?

[Bluefoxfire]

Made me think of net tiles. Basically catches any items but lets everything else through, including critters and ladders (and by extension dupes).

[Nexto21]

As much as I love the idea of buoyancy, I think the game engine might explode. Considering how mixed gases and critter movement already impact performance, calculating the movement of all that debris would be a nightmare. Let’s be honest: density is incredibly buggy. Currently, if you drop a dense liquid into a tank of light liquid, it just keeps stacking in a single tile. I daren’t guess what would happen if floating debris hit that tile—it might shoot straight through the core of the asteroid!
I may be overthinking it, but buyency influence gases as well and concidering the ridicules gass presures some of us have  in some palces  they may be some levitating debries in some bases

But I love the idea of building a debris lift and messing with overpressurized liquids

[Maris]

On 3/8/2026 at 10:38 PM, Nexto21 said:

As much as I love the idea of buoyancy, I think the game engine might explode. Considering how mixed gases and critter movement already impact performance, calculating the movement of all that debris would be a nightmare. Let’s be honest: density is incredibly buggy. Currently, if you drop a dense liquid into a tank of light liquid, it just keeps stacking in a single tile. I daren’t guess what would happen if floating debris hit that tile—it might shoot straight through the core of the asteroid!
I may be overthinking it, but buyency influence gases as well and concidering the ridicules gass presures some of us have  in some palces  they may be some levitating debries in some bases

But I love the idea of building a debris lift and messing with overpressurized liquids

Sorry for the off-topic, but I think it's relevant here.

Critters cause lag due to unoptimized A* pathfinding, and this could be improved. A virtual layer for path tracking could be created, where the game marks whether a room has been calculated or not. The algorithm for each critter would be:

1. Check if the room for the critter's current cell has been calculated.
2. If not, run A*, calculate the room volume, and write the result (the volume value) to every cell of that room on the virtual (invisible) layer.
3. If yes, just take the volume value and use it for critter happiness.
4. If a tile was built or a block was destroyed, clear the virtual layer at the end of the tick. Otherwise it's a reusable layer that should persist for a long time without recalculation.

Even if room volumes are recalculated every tick, each cell would only be used in A* once. So 1000 Shine Bugs in a 2x2 room wouldn't cause any lag at all. Sure, there are nuances for Pacus, but those can be solved by adding another virtual layer for "liquid rooms".

As for liquids, this also seems fixable. Say there are three cells stacked vertically (C1 top, C2 middle, C3 bottom). If high pressure has built up in the bottom cell, the game could check not just the cell above, but two cells above. And if C1 = C2 by liquid type, merge them (C2 flows into C1), and C3 spreads across 2 cells accordingly.

Yes, this slightly increases the load, but only barely, because it's just basic if-then-else. That is:
IF a cell contains liquid AND pressure is too high AND the cell above contains a different liquid type, THEN ...
This condition would almost never trigger.

[ThePyro]

On 3/9/2026 at 12:38 AM, Nexto21 said:

As much as I love the idea of buoyancy, I think the game engine might explode. 

I don't really get why this would cause major issues. With liquids and gases, it's pretty involved—you have to account for the max pressure of the resource in each tile, compare masses and types in neighboring tiles, decide where things stack and where they don't... Floating or flying resources would be simpler since they only move strictly vertically, and the sideways flow (wind or liquid stream) is confined to a fixed area rather than spreading across the whole map. That would require way less computational than handling gas mixing or, say, calculating critter flight paths.

[Maris]

I hope the devs will consider introducing such an update. And if it's a DLC, the player will even have a choice. In any case, they can add a checkbox in the options to disable this physics feature, for those concerned about the CPU load.

[ThePyro]

It would be great if one of the community managers would post on the forum at least occasionally. Three months and they've been completely silent. Is it really that hard to write a few sentences like, "Great idea, we're already working on it," or "It's in the plans, stay tuned for updates," or even "This will never happen because it doesn't fit into our vision for the game." Any feedback from them would be important and appreciated.

Hey, Klei guys, do you read the forum? 🫣