2.5D / Background cells (cavity cells / hollow parts of colonies)

Since it is being considered as an optional feature to be worked on in the near future and a summary was requested, I figured I should write a design concept to make a more balanced decision.

Mechanics

To quickly explain what it even is:

In the editor, using an option on the right-click menu on cells (or I suppose it could be a separate tool like with the metaball rotation tool?) you can “move” cells into a layer below the other cells. During gameplay, these cells will not have collision with small objects, such as small chunks, or cells (that are not themselves shifted down), those would just “move over” the shifted cell.

This would have restrictions on it so that you can’t just have all your cells shifted, or “cavities” in unreasonable places:

  • Must be surrounded by other cells (shifted or not) on at least five four sides.
  • Other rules are possible, but from what I tested out do not seem very necessary for the small number of cells we use.
Some examples of what would and would not be allowed



Effects

The main mechanical benefit would be the change in collision itself:

  • If you have two toxin-firing cells in a column, the rear projectile would normally collide with the front cell. But if the front is shifted down, that no longer happens.
  • If the extracellular digestion feature is finished, this allows you to scoop up chunks and cells while having them “fully surrounded” by digestion cells for rapid effect.

Of course, many cells don’t use this kind of collision mechanic, so we can also apply any of a selection of other effects:

  • Since for any cavity, every floor would logically come with a roof, it counts as two cells. For the most part this could be simulated by doubling the specialization bonus it receives (which is currently always passed as 1 + X, I believe?).
    • I think in this case it should reasonably also cost twice as much ATP and reproduction resources.
  • Counts double for cell adjacency.
  • More directed bonus on just process speed.

Complications

While I think it’s pretty clear in the editor, real gameplay has the complication that the 4/6 rule isn’t consistent as cells are growing/die. And obviously we don’t want colonies that are effecitvely almost immortal because at some point all of their cells are hidden from normal collision. Keeping it fully consistent all the time would be too difficult on the growth order, but I think I have some solutions.

  • The best option, if possible, is to have any cells not currently following the 4/6 requirement move back into the “normal” layer. That keeps management for the player simple, but still prevents the problems mentioned above.
  • If the above is not possible, than the growth order should enforce that there is always a non-shifted cell to start with. And every shifted cell that is not connected to a non-shifted cell (directly or via a chain), should immediately die.

Final musing

In theory, we could have a system that restricts the activities of cells when they are completely surrounded by other cells, thus requiring either thin structures or cavities. But I think this would be too complex to implement at this point.

2 Likes

Why would that be? If the front cell of your colony is a background one, any damage or pilus poking etc. would just go through it and hit your other cells. So any damage reducing would be basically non-existent.

Unfortunately this would be the hardest thing to implement, and I think also to explain to the players why sometimes their background cells are normal and sometimes they actually work.

Making the check for the adjacency requirement use growth order seems simple enough. However the second point just sounds like a nerf to the “keep attached” mechanic of colonies, like mostly disabling it for background cells.

I posted some initial musings on what I could implement easily related to this here: Multicellular Parts - #22 by hhyyrylainen


Also overall I think we need an actually good name for this feature. Players are not going to understand what “background cells” are, and even I’ve been very confused with the feature. So I think a new name is kind of needed for this feature. Something like maybe cavities or cavity cells or hollow regions?

I am mainly concerned about getting into situations where the only surviving cells in a colony are cavities, so they can’t ever be fully killed (by some attackers at least). That, and what to do with cavity cells that get disconnected from the main body and are free-floating.

Maybe “cavity cells are killed if there are no non-cavity cells in the colony” would be better phrasing?

Yeah, that was what I was afraid of, but like you say, enforcing it via the growth order I think also works, and may be better in some ways.

Yeah, that’s valid, and I haven’t been able to put a definitive label on it myself before now. I think “cavities” is a good description of the mechanic as a whole. So I guess “cavity cells” will be good enough unless somebody comes up with a better name.

2 Likes

As long as the first cell is disallowed to be a background cell, you can’t get into that problem, I think. The microbe AI might be a bit dumb in targeting but would probably still eventually also hit the root cell and kill it.

Well the same effect would be from disallowing the first cell being a cavity, and I think that would help with multiple other problems you have described. So that seems like basically the same but much easier to implement?

1 Like

I don‘t think this makes sense since it assumes that normal cells all don‘t have any other „hidden“ cells on top or below them. In the multicellular stage, the cell count is an abstraction anyways since your organism with 20 cells symbolizes a larger (but still microscopic) organism with 100s or even 1000s of cells (as I recently learned from others in another thread).

If anything, those hollow portions of your organism would contain fewer (abstracted) cells than the full portions. It would therefore in my opinion make more sense if these portions recieved reduced numbers for all their positive as well as negative calculations. So everything from adjeceny bonus to osmoregulation costs to compounds produced (e.g. through chemosynthesis etc. They should also decrease your volume to surface ratio.

It does, sort of, or at least assumes that having a cavity in between lets the total amount of cells within a hex be twice as large. And lack of a body cavity is considered a limiting factor in how thick things can get.

Indeed it is, but I do always imagine that multiplication mostly in the 2D dimension. Take a look at a placozoan for example:

Placozoan cross-section

Now admittedly, that’s two (kind of three) cells stacked, not just one. But by comparison these can be 1mm across, so a lot more cells in the other dimensions. Things can be very flat. Jellyfish for example are also 2 cells thick in places where there’s not a bunch of jelly (or a cavity) in between.


To begin with, the initial proposal here is to visibly move cells down on the y axis to move them out of the way of collision. That suggests a thickening of the overall structure to make way for the cavity, not “hollowing out”, if that makes sense.


I don’t think the doubling is a strictly necessary part, but I also don’t think reducing the “virtual number of cells” makes sense.

2 Likes

I definitely see your points… In the end there is no clear way on how to interpret the nature of the implied additional 100s of cells the 2D multicellular species represent.

Maybe then the most elegant solution would be to remain agnostic on this by keeping this feature focused on its core functionality and foregoing the doubling?

1 Like