Multicellular Stage: an Actual Roadmap

Alright everyone, Microbe stage is finishing up, which mean that, after some frantic debugging, we are going to be hands-on-keyboard for multicellular. With over ten years to plan, we are obviously ready and everybody knows what they’re doing, right? :grin:

In all seriousness, I’m on the record for not having much confidence the microbe stage would even finish, but I am actually very optimistic for the multicellular stage. Not only do I think our current generation of volunteers might be our most dedicated and versatile in quite some time, but a lot of the work is done for us: many, probably most, mechanics of multicellular were already created for microbe stage. Frankly, I think it’s also helping us that people haven’t been focusing on multicellular all these years to keep scope creeping semi-finished ideas to take away focus.

I only saw the last 4 years of Microbe stage work, but I’ve seen a lot of talent and energy go to waste. Devs talk about some great gameplay idea, decisively decide to take more time to decide things, write a very long and interesting forum thread that newer game designers don’t read because it’s super long, then leave, and the whole concept gets forgotten or cut. We can’t make progress on a feature for a year because there’s no 3D modelers to make models, then a 3D modeler shows up and we can’t decide what to ask from them, and the modeler understandably leaves for somewhere else. One programmer makes a really cool prototype and leaves before ever incorporating it into the game, while another programmer makes a very good PR which withers on the vine after 6 months, while nobody else works on that feature because there’s already a PR for it.

I can’t solve all of this (and we probably never need to; Thrive is a chaotic project by nature and likely always will be), but while the multicellular concepts are as uncluttered as they ever will be I want to make three lists:

  • Stuff that we know we’re going to do, at least mostly understand how to design, and need a programmer to implement.
  • Stuff we want to do, but need to come to a decision on or prototype on so we can put something in the first list before we finish the stage.
  • Models/art stuff we know will be in multicellular stage, even if we don’t know how exactly we are going to use it, or if we don’t have the other resources to implement it.

I know that there have been other attempts to make similar lists; we can put the results there or draw information from there if you want. I expect we will eventually populate our release roadmap with this kind of stuff, which was the closest thing to a GDD we were using by the last couple years of Microbe Stage development (pop quiz: how many GDDs does Thrive have? How many can you find?)

I would rather see a low-resolution list of things for the whole stage, than a very detailed breakdown of every PR we need to make. I expect any plans made here to change, that’s okay. What I really want is a list of what we are and are not trying to fit into Multicellular stage (not macroscopic), as of right now, and what needs to happen next for each of them. What needs to happen next might be “decide if and how we’re really going to do this”, but that itself is a task.

@Gamedungeon I’m going to put you on the spot here, since you are the lead game designer. The next few months are likely to be a good season in my life to whip up a prototype for you, although the skeleton of Multicellular may not be in a strong enough state to learn much from them just yet.

4 Likes

I’ll start, based on making an AI summarize this thread for me and some common sense things:

TO IMPLEMENT:

  • AI for multicellular
  • Autoevo for multicellular
    • making autoevo not skip multicellular species
    • mutation strategies for autoevo (I have some solid ideas for this to start with so I don’t consider this a “to decide”, but feel free to disagree with me)
    • making existing selection pressures work for multicellular organisms

TO DECIDE

  • Size-based penalties
  • What are the other reproduction types, and how do they work?
  • editor costs and mechanics of placing vs changing cells
  • life cycles

TO DRAW

  • intra-cellular matrix (right? We’re still doing that?)
  • Shells? Other external parts that microbes won’t have?
2 Likes

Congrats to GameDungeon on the promotion!

IMO, there are two mechanics that will be definitive to this stage…

  1. Cell Specialization - What incentivizes players to make cells which are actually specialized instead of the more generalist cells we see currently?
  2. Cell Adjacency - What puzzles will the player have to solve when it comes to the arrangement of their colony?

If the two questions above are answered handedly, we can approximate general phenomena that life exhibits as they become Multicellular, and have engaging gameplay.

When it comes to Cell Specialization, I feel pretty confident that @Rathalos has found a good solution. It’s described in this post in the “The Suggestion” section, but the basic idea is: cells in a colony receive a bonus, and that bonus strength and type is determined by how much of their cell is composed of a specific part. I’d say that is candidate A in terms of an idea that is ready for the “let’s actually see what it does” prototype.

When it comes to Cell Adjacency, that’s a bit less concrete. But a suggestion that has come up is essentially having interactions between cells based on their composition. So a cluster of cells with a lot of vacuoles would boost storage for those cells, a cluster of cells with a lot of pulling cilia would result in a black hole, etc.


One other thing that would be good to discuss is to have a shared vision of the fundamentals of the Multicellular Stage. I think it would be a disservice to just see it as an extension of the Microbe Stage - what different challenges, objectives, goals, and “moods” are we presenting the player in this stage?

I detailed this more in this post in the sections titled “The Premise” & “Multicellular Stage Expectations”. But for me, the gist of what I think we should see between the two stages is…

Microbe Stage

Here, you’re exploring your home. What works here? What doesn’t work? And how will you keep up with all the swings?

  • Surviving on a volatile and unstable planet, with generally more dramatic events.
  • Overcoming constraints of the metabolisms you choose.
  • Carving out a place for yourself against other microbes.
  • Overall, trying to keep up as resources vary and environmental events occur.

Multicellular Stage

Here, you’ve settled in a bit, but the competition is picking up. How will you grow? How will you defend yourself? Can you keep up with all the competition around you?

  • Maximizing stats, and creating a powerful organism.
  • Much more “dangerous” AI species with more powerful abilities.
  • Surviving a much more chaotic game scene, with projectiles, cell goop, and destruction around you.
  • Overall, a desire for “more” - more speed, more efficiency, more projectiles, stronger abilities, etc.

IMO, The Microbe Stage is essentially the player proving they have what it takes not to starve on their new planet. The Multicellular Stage assumes that you’ve proved that well, so it then challenges you with throwing more, and more, and more at you, with dangerous abilities on colonies which have somewhat goofy physics. Can you keep up?

If we nail the general “vibe” we’re going for, we can develop with a pretty cohesive thought. Thrive obviously benefits tremendously from a diversity of thought, but I do think not having the same chasms we’ve had at some points with the Microbe Stage would help our design be more intent.

1 Like

Is this a typo?

Oh, looks like Deus is already here.

This needs to be dynamically generated by the code. So a graphics programmer is needed or this needs to be shelved until such a time one is available.


I think any multicellular roadmap should start with the items we need the most to no longer need to call multicellular a prototype. And based on latest player feedback I remember, the most important thing is to solve how multicellular feels way too slow to progress (both in how long it takes to grow your colony and how much you need to place more cells, though this second part could be alleviated by having actual stage-specific stuff to interact with so the player doesn’t feel like they just need to get past multicellular ASAP).

1 Like

Alright, after a great start, sounds like we have a little more:

TO IMPLEMENT:

  • AI for multicellular

  • Autoevo for multicellular

    • making autoevo not skip multicellular species

    • mutation strategies for autoevo (I have some solid ideas for this to start with so I don’t consider this a “to decide”, but feel free to disagree with me)

    • making existing selection pressures work for multicellular organisms

  • Cell bonuses for specialization

  • Improve reproduction progress rate

TO DECIDE

  • Multicellular unique abilities

  • Size-based penalties

  • What are the other reproduction types, and how do they work?

  • editor costs and mechanics of placing vs changing cells

  • life cycles

TO DRAW

  • intra-cellular matrix (graphics programmer needed)

  • Shells? Other external parts that microbes won’t have?

1 Like

I think for now it is best to keep discussion on the actual design concepts of the multicellular stage over in the thread Deus linked.

So as for the purpose of this thread: I guess it would be nice to set up a living document or a page of the wiki just to list the things there is a broad consensus on. So indeed just like the release roadmap we already have, but perhaps without the “time-line” for implementation. I think that will work better and be more visible in the long run than keeping track of a thread like this.

I did suggest some mechanics for this in the other thread, but I don’t think there is a consensus on the necessity of this. So I wouldn’t put this on a list to encourage people to produce models for yet.

As for other things to add:

TO IMPLEMENT:

  • Change the multicellular stage startpoint, probably to a blob of unspecialised cells. (I put this here because Deus seemed to agree on this, and I have seen no counter-opinion)

TO DECIDE:

  • Do we want non-cell elements to be available in the multicellular stage/editor?
  • Progression mechanics from microbe stage, through multicellular into macroscopic.
  • Symmetry mechanics (yes/no, here or only in macroscopic?)

I haven’t settled on the exact replacement yet (still researching for inspiration), but I do think we should replace “reach X cells” as the progression requirement with something else. This should probably be much closer to Microbe Stage, where you evolve specific parts (nucleus, binding agent) in order to be able to advance.

As far as colony growth speed goes, I think it should simply just be sped up. We can multiply the passive gain of reproduction compounds by (close to) the number of cells the final organism is supposed to have. I think we can manage to make multicellular generation gameplay interesting enough to last longer than a microbe generation, but not by toooo much.

2 Likes