Contiguous cells & differentiation

Future Game Multicellular Stage
1

I’m not sure if this has been implemented yet, but I’d guess it hasn’t.

One thing I remember from my multicellular runs on an older version was that my cells had sizeable gaps. Even when I created wall-cells to form a sort of skin, they weren’t particularly effective, as attacks could get through, and they could be easily separated. I suggested methods of solving this - such as a way to adjust the strength of cell connections, I believe - but I’m not sure what happened with them, as it’s only very recently that I could access the more updated Windows versions.

Running on this outdated knowledge, I’ve just conceived of a solution to a problem, the latter of which may no longer exist, and the former of which may have been implemented. This solution is to simply make the walls of cells contiguous. They connect together with no gaps, so that there’s no way to attack any cells that are behind other cells on all hexes. This would be awkward with cells of highly variant sizes, but that’s already awkward and probably needs a solution in and of itself.

The second thought is something that, similarly, is running on an outdated knowledge of the multicellular stage because I don’t want to spend an hour reaching it just to test this. This is the fact that it’s strategic to, before beginning to develop complex cell types, first develop a cell type which is empty. This saves MP. Say your stem cell serves 5 different functions (digestion, movement, metabolism, photosynthesis, and ammonia production, for the case of a hybrid predator/plant). If you create 5 cell types for each of these, it’ll be pretty difficult, as you’ll have to pay the MP costs for removing all the unneeded structures individually for each cell type, before adding in the desired structure to fill the cell. However, if you first create a dummy cell type that you don’t intend to use, you can delete the offensive structures in it, then duplicate that into the actually intended cell types, saving MP over time. I’m pretty sure this isn’t an intended behavior.

I don’t know how to fix it, though, if fixing is needed. One could make it cheaper to edit cell types and more expensive to place them, to shift MP costs such that this isn’t as useful of a trick. One could take the brutish method and force the player to actually use cell types they create. I don’t know. It seems like something to at least think about, whether or not it’s something that ultimately needs fixing.

1 Like
2

I think this idea needs more development, as the cell shapes just may not fit together so what should be done? Are the visible gaps just filled in with invisible physics collisions or what?

I think this is really the opposite of what should be done. Because in real life areas of certain cell type can grow and shrink very quickly in evolution, but entirely new cell types evolving is a much slower process (AFAIK, I’m not a theorist).

2 Likes
3

This could be a rather realistic option - if an organism stores data of a cell type but doesn’t ever use it, why bother not to remove it’s information to free up some resources needed elsewhere?

1 Like
4

Just did a test run of Multicellular and I did notice that sometimes the cells just outright float. The hexes don’t seem to be particularly effective with cells of different sizes. I don’t know enough biology to know if cells in multicellular organisms are often of different sizes, but I’d assume they would be, so I’m not sure how to deal with it.

The brute-force solution is to just connect all the cells together regardless of size, so there’s a lot of empty space or conversely they’re packed tighter than they could normally be, so that there’s not any gaps. I envision a large multicellular organism as having its internal cells appear roughly hexagonal, with only the outer cells having rounded bounds.

The cell type evolution thing still seems like a rather thorny problem, given that, as long as cells can differentiate from non-stem (and I’d assume they should be able to do that) cell types, you can always take that shortcut of making an empty cell type to build off of, which would definitely boost your evolution speed once it’s established. The brute-force method is really the only solution I can think of right now.

1 Like
5

There would be massive overlaps, like probably close to 50% of the graphics would overlap and look super glitchy and mushed together. And I’d have to read like endless bug reports about it so that isn’t a solution.

6

Regarding cell differentiation, stem-cells themselves are known to be undifferentiated and versatile cells that then can be turned into more specialized cells. So players who adopt your strategy are actually representing a real evolutionary trend, which absolutely is intentional on our behalf. Having a trimmed, stock cell that you can turn into a more specialized cell is a pretty cool layer of evolutionary storytelling in my opinion.

Of course, we don’t know exactly when such behavior emerged, but it definitely is there, and is an important part of multicellular life.

3 Likes
7

Maybe create some kind of algorithm for intercellular substance?

8

The kind of effect the mucocyst creates could be used as a start, maybe?

9

I had a few ideas to deal with these issues

I actually mentioned this in the ideas for multicellular thread, my idea was to be able to modify a specific cell already placed within your organism and change it into a different cell type. Then simply remove the ability to duplicate cells in the cell tab, now you can only place them down or edit them. Then simply remove any cell type that is not a part of your organism, possibly after warning the player first

Ok so bear with me because this going to become extremely long and technical.
The organisms don’t look very great at the moment
We want them to either look like this
1000px-Multieditorconcept

Or this

Either way they need to connect consistently and properly without any overlapping or free-floating cells. This is the three step solution I came up with, it’s not perfect but it should help with atleast thinking about how this is going to be done going forward.

Fluidity would effect how the cells would mesh, fully fluid cells would mesh the best with other cells without really caring how the fluid cell is shaped, while fully rigid cells would need to be specifically shapped In order to meld nicely.

Larger surface area of connected cells would also play a role,
Two fully rigid cells that are only connected to by a small strip of membrane would be much easier to break apart then two fluid cells that meld perfectly.

Step 1: points
Each cell is given six connect points correlating to the six sides of the hex it is in. The points are placed on the hexes closest to each side. If there are multiple hexes then the point is placed in between them.
Step 2: move
Each cell is then placed in the order they will appear, this of course will result in a lot of messy overlap, so after every cell is placed each cell is moved based on how fluid and new they are. The newest and most rigid cells get moved the most while the most fluid cells simply stop caring about overlap at all (as long as less than 50% of cell is overlaping).
Step 3: Refine
After this step is done there is probably still edges and gaps that need to be fixed. so this is when edges of the individual cells are redefined to make the whole look better. If every cell is fully rigid this step is skipped. This step is also a 3-step process

  1. First it finds the center of each cell and cuts it into six triangles

  2. then it compares each triangle to each neighboring triangle

  3. The neighboring triangles are then adjusted to meld together by comparing the amount of hexes from the center to the neighboring edges of both triangles and adjusting accordingly.

This will, in theory, hopefully make an algorithm that can mesh cells together in a way that looks nice.

10

Oh, yeah, that’s a great idea, and it makes a lot of sense.

The issue with a dummy cell type also probably wouldn’t apply to macroscopic tissues. I’d imagine that, at that point, it would be a lot cheaper to refit stuff that’s already there than to build a tissue from scratch.’

That image of the blue multicellular organism is basically what I was thinking of as a better shape for multicellulars, and the green microscope image is what they actually are right now.

In respects to how this is an observed behavior IRL - making the stem cell simple and easy to adapt is still kind of difficult, since, you know, that’s your stem cell. It needs to be able to support not only itself, but several other cells while the organism gets itself established. It can’t be small unless all the other cells are small, and with a nucleus, you just aren’t going to be able to have small cells.

11

It’s actually kinda funny since the second one is an actual multicellular organism.


Looking up images for microscopic multicellular organisms, and with a bit of digging, there’s a fascinating variety of microscopic multicellular organisms with a large variety of forms.
I hope one day the game will be able to emulate many of these forms.

4 Likes
12

Yeah, looking at those, you could def. see a good deal of variance. A lot of those individual ones look like they could be emulated with a few sliders and have a gameplay effect. For instance, do you want a loose colony, or a tight one? Presumably that slider would have similar effects to the fluidity/rigidity slider we already have.

Those are some weird cells.

1 Like
13

I can guess that the slider will be associated with intracellular matrix, a feature yet to come into the game.