Developing Guide To: Microbial Stage

Good afternoon/night/morning/ambiguous-space-time-period thivers.

We have been lately discussing about the lack of Focus on our development pipeline, and IMHO, to gain focus we need something clear and concise about we're heading towards. Since everything I saw lately was a lot of blablabla and some discussion about underwater civilizations, I'm going to open this thread as a recopilation of serious and feasible ideas that could get into thrive's first stage

My idea is to gather all the features proposed from the very beginning and list them all in this first post, so everyone developing anything of this stage can have a quick look and see what's the next important step to take. As the features are being listed, we'll discuss it's feasibility, and when something passes the "proposed" step, we'll give them a priority from 1 to 3.

All of us are free to propose new rules if we have a better way of organising this list .

As a sign of good will, an since I consider myself a very recent newcomer on the project, I will not list any feature right now, as I consider there's other people who can point to features that have been lurking since the very conception of this project.

So here it is, let's start the first:

Proposed features:

- Simplified Auto-evo system:

• One single Biome

• Specialized Food Web

• Simpler statistics for mutation decisions

Confirmed features:

Discarded features:

"Features" don't have much to do with programming, do they?

roadkillguy wrote:

"Features" don't have much to do with programming, do they?

I'm asking about "overall" features. Personally I see programming as a method for solving problems and. turning concepts into "real" things.

My arachnid sense tells me that someone called roadkill-something is going to discuss lots of features the constructive way, amirite?

Ok, i'm gonna start with stuff from the current concept. This one is kind of hard to explain so i'm going to quote directly from the page here.



Proposed Feature: If a cell is unsuccessful, the computer will access what it needs to be less unsuccessful. Really there's only 2 factors I could gauge success with and that was the cell's population, and if that population can produce more members than members are being consumed. That can be put into a simple equation. If that cell population is selected that turn, and is found that more members have been lost then gained.


I think this is a pretty good idea, and it's simple. I'd say if I get three other member approvals it's a confirmed feature. Now, discuss and object.

zippybomb wrote:

Ok, i'm gonna start with stuff from the current concept. This one is kind of hard to explain so i'm going to quote directly from the page here.



Proposed Feature: If a cell is unsuccessful, the computer will access what it needs to be less unsuccessful. Really there's only 2 factors I could gauge success with and that was the cell's population, and if that population can produce more members than members are being consumed. That can be put into a simple equation. If that cell population is selected that turn, and is found that more members have been lost then gained.


I think this is a pretty good idea, and it's simple. I'd say if I get three other member approvals it's a confirmed feature. Now, discuss and object.

I think this idea will be good if used alongside a random evolver. If after X random evolution attemps without clear success, try to artificially balance it. This will ensure more npcs if random real evolution was unfavorable to others.

Any programmer giving advice on how shall we implement this?

You're doing fairly well here. evolutionary fitness is based on ability to produce fertile offspring. This, for us, is impossible ti measure, since our evolution happens on a generation by generation basis. however, if you read our procedural evolution system, you'll gain a better understanding of how evolution works in the game. The great thing about cell stage is that it's much less complex than multicellular stage, so we can run the procedures almost exactly in the same way, but they'll be easier to keep track of. I see no reason to reinvent the wheel and make a different evolution system for cell stage.

Ok, this idea's discarded then? Add it to trash pile then uverion. One problem solved.

~sciocont wrote:

You're doing fairly well here. evolutionary fitness is based on ability to produce fertile offspring. This, for us, is impossible ti measure, since our evolution happens on a generation by generation basis. however, if you read our procedural evolution system, you'll gain a better understanding of how evolution works in the game. The great thing about cell stage is that it's much less complex than multicellular stage, so we can run the procedures almost exactly in the same way, but they'll be easier to keep track of. I see no reason to reinvent the wheel and make a different evolution system for cell stage.

I've just looked at the auto-evo paper, in my oppinion it seems fairly simple yet effective, I'll be happy to see a formal programmer oppinion about it, but even with my little C++ skills I'll see this fairly achievable

Even though, I see the auto-evo concept a bit generalized. We don't really need to re-invent the well as you said, but IMO specializing this concept for a trully cellular one will reduce the resources needed (human and computational) Defining specialized concept:

- One single Biome. We're talking about biomes containing a number of species, I'll change this to one single biome. We're moving on a effectivelly small pond of about the size of a spoonfull of water, wasting resources for tenths of biomes is overkill, sure this was implicit, but making it explicit will help us all

- Food web. Probably, due to my lazyness, I'm skipping the paper for food webs, but giving it a little tought, I think that "food webs" can change his meaning by adapting to every game/stage. Cell FW will be fairly simple compared to creature FW, and if we ever get to civ stage, the FW can change to something similar to a Political Web, representing the planetary/galactic dominance of a civ. But this is another story, and mus be told on another occassion

- Simpler statistics for mutation decisions. Cells are simpler than multicellular organisms, so when they mutate there are thinks that the system can skip in order not to consume excessive resources. I think that giving stats for Size, Speed and Armor would suffice for the organism, and using a separate "Additional parts stats" for deffensive, offensive organs and types ov movement will suffice. I would ask the biologists here to advice us: ¿Is there any chance for a cell to become toxic when consumed by other organism? (not counting virical strategies, as I will count them as offensive parts.

I'll replace the first idea with these ones with a link to auto-evo wiki

uverion, very funny sig.
but yes this all sounds good to me, anyone want to type their opinions?

I think microbe stage should consist of no more than five biomes, but we should focus on one first. Food webs shouldn't be hard coded, but we're going to have to be clever to ensure that they work and don't collapse very easily. It will also be a bit hard to figure out how to program in the logic for the webs in the first place, but we can discuss that in the appropriate thread. I definitely agree that mutations should be simpler, but they needn't be simplified- for unicellular organisms, they'll already be very simple. the most important things we need to know are diet, membrane construction, what kinds of organelles it has, where they are, surface area, volume, etc.

ThankYou for summing that up for us, sciocont.
Who here has prgramming abilities?

I did make a concept for cellular evolution some time back. It isn't the greatest but it could help flesh things out I guess: https://thrivegame.canadaboard.net/t665p30-cellular-evolution-concept-questions
(look for my post with spoilers, that sums it up.)

uverion wrote:

zippybomb wrote:

Ok, i'm gonna start with stuff from the current concept. This one is kind of hard to explain so i'm going to quote directly from the page here.



Proposed Feature: If a cell is unsuccessful, the computer will access what it needs to be less unsuccessful. Really there's only 2 factors I could gauge success with and that was the cell's population, and if that population can produce more members than members are being consumed. That can be put into a simple equation. If that cell population is selected that turn, and is found that more members have been lost then gained.


I think this is a pretty good idea, and it's simple. I'd say if I get three other member approvals it's a confirmed feature. Now, discuss and object.

I think this idea will be good if used alongside a random evolver. If after X random evolution attemps without clear success, try to artificially balance it. This will ensure more npcs if random real evolution was unfavorable to others.

Any programmer giving advice on how shall we implement this?

I think that this version of the evo has already been discarded as there has already been a discussion on the evo system. So, I think it's discarded.


- One single Biome. We're talking about biomes containing a number of species, I'll change this to one single biome. We're moving on a effectivelly small pond of about the size of a spoonfull of water, wasting resources for tenths of biomes is overkill, sure this was implicit, but making it explicit will help us all [/quote]


I think this is a great idea. EEven though you will spawn in the ocean, it's going to be on a volcanic vent, you're going to be so tiny it's not going to be possible to move to another biome.

Dr_Chillgood wrote:

I did make a concept for cellular evolution some time back. It isn't the greatest but it could help flesh things out I guess: https://thrivegame.canadaboard.net/t665p30-cellular-evolution-concept-questions
(look for my post with spoilers, that sums it up.)

Yes, I used a quote from it.

zippybomb wrote:

Dr_Chillgood wrote:

I did make a concept for cellular evolution some time back. It isn't the greatest but it could help flesh things out I guess: https://thrivegame.canadaboard.net/t665p30-cellular-evolution-concept-questions
(look for my post with spoilers, that sums it up.)

Yes, I used a quote from it.

Ah I see. It's helpful for me to read the entire post to not create a redundancy in the thread, isn't it? :3

Dr_Chillgood wrote:

zippybomb wrote:

Dr_Chillgood wrote:

I did make a concept for cellular evolution some time back. It isn't the greatest but it could help flesh things out I guess: https://thrivegame.canadaboard.net/t665p30-cellular-evolution-concept-questions
(look for my post with spoilers, that sums it up.)

Yes, I used a quote from it.

Ah I see. It's helpful for me to read the entire post to not create a redundancy in the thread, isn't it? :3

Yes, we're trying to keep it concise. Anyone else think the single biome is a good idea for cell stage?

maybe, I wanto to see what Sciocont says

~sciocont wrote:

Food webs shouldn't be hard coded, but we're going to have to be clever to ensure that they work and don't collapse very easily. It will also be a bit hard to figure out how to program in the logic for the webs in the first place, but we can discuss that in the appropriate thread.

Ok, we should open that thread and call the coding armies to gather in there. We can start conceptualizing it and even pseudo-coding it

~sciocont wrote:

I definitely agree that mutations should be simpler, but they needn't be simplified- for unicellular organisms, they'll already be very simple. the most important things we need to know are diet, membrane construction, what kinds of organelles it has, where they are, surface area, volume, etc.

So we should start concreting what statistics go into the cell chapter right? Hope there will be more discussing topics tomorrow morning, it's getting pretty late here in Spain (3:40 a.m.) Good night thrivers!

uverion wrote:

~sciocont wrote:

Food webs shouldn't be hard coded, but we're going to have to be clever to ensure that they work and don't collapse very easily. It will also be a bit hard to figure out how to program in the logic for the webs in the first place, but we can discuss that in the appropriate thread.

Ok, we should open that thread and call the coding armies to gather in there. We can start conceptualizing it and even pseudo-coding it

~sciocont wrote:

I definitely agree that mutations should be simpler, but they needn't be simplified- for unicellular organisms, they'll already be very simple. the most important things we need to know are diet, membrane construction, what kinds of organelles it has, where they are, surface area, volume, etc.

So we should start concreting what statistics go into the cell chapter right? Hope there will be more discussing topics tomorrow morning, it's getting pretty late here in Spain (3:40 a.m.) Good night thrivers!

-it already exists, search it and start posting.
-go to one of our cell stage threads.