My Guidelines for playing like the AI!

Current Game Microbe Stage
1

I don’t know which category I should put this in, but this is an idea I had over a year ago. Keep in mind that this was made with the old Auto-Evo in mind, pre-Miche system, and before a lot of the new improvements and organelles were added.

Thrive Subspecies Random Generator Guidelines

  1. Generate and divide by 0.25: 0-0.25 Add, 0.25-0.5 Delete, 0.5-0.75 Change, and 0.75-1.0 Don’t Mutate;
  2. If Add, then generate and divide by 0.083,1/12: 0-0.083 Cytoplasm, 0.084-0.167 Metabolosome, 0.168-0.251 Thylakoids, 0.252-0.335 Chemosynthezing Proteins, 0.336-0.419 Rusticyanin, 0.420-0.503 Nitrogenase, 0.504-0.587 Oxytoxisome, 0.588-0.671 Thermosynthase, 0.672-0.755 Flagellum, 0.756-0.839 Pillus, 0.840-0.923 Chemoreceptor, and 0.924-1.0 Slime Jet; if Nucleus is included, divide by 0.077 and 0.923-1.0 Nucleus;
    o 2a. Before adding, first count the number of sides starting from the center/top hex, then generate and divide by 1/side number;
    o 2b. If orientation of hex is allowed, generate and divide by 0.166, 1/6;
  3. If Delete, first count the number of hex starting from the center or top, and then generate and divide by 1/hex number to get the hex to delete;
  4. If Change, then repeat step 1, except: 0-0.333 Hex, 0.334-0.666 Membrane, and 0.667-1.0 Behavior;
    o 4a. If Hex, then generate and divide by 0.5,1/2: 0-0.5 Change Position, 0.501-1.0 Substitute;
     4a1. If Change Position, then repeat counting of hex as in step 2a to select hex and place it in a new position, and apply step 2b if applicable;
     4a2. If Substitute, then repeat the same procedure in step 2a to select hex, but then repeat step 2, step 2a, and step 2b for substituted hex;
    o 4b. If Membrane, then first generate and divide by 0.5,1/2: 0-0.5 Increase Membrane Fluidity, 0.501-1.0 Decrease Membrane Fluidity;
     4b1. Generate and divide by 0.05,1/20 to find out how much to increase/decrease;
     4b2. For membrane type, generate and divide by 0.166,1/6: 0-0.166 Single Membrane (Default), 0.167-0.332 Double Membrane, 0.333-0.498 Cellulose, 0.499-0.664 Chitin, 0.665-0.830 Calcium Carbonate, 0.831-1.0 Silica;
     4b3. To change membrane color, randomly generate each R, G, and B number (R,G,B) and divide by 1/255;
    o 4c. If Behavior, first generate and divide 0.5,1/2: 0-0.5 Increase Behavior, 0.501-1.0 Decrease Behavior;
     4c1. Generate and divide by 1/64 to find out how much to increase/decrease;
     4c2. Do this for all Behavior slides;
  5. Repeat steps 1-4 as many times allowed per MP cost, but stop when no hex can be added, deleted, or substituted/moved from position;
  6. Repeat steps 1-6 four more times to generate other Subspecies, and take a screenshot of the edit screen for each Subspecies generated for further analysis;
    o 6a. Should AI Speciation have occurred prior to generation of new Subspecies, then the number of Subspecies generated is 4 minus the number of AI Speciation Species;
  7. Now, compare the Fitness Values of the Subspecies to the Fitness Value of the new Mutated Species;
    o 7a. If only one Subspecies has a value higher Fitness Value than the Mutated Species, then replace the Mutated Species with the new Subspecies;
    o 7b. If two or more Subspecies have a higher Fitness Value than the Mutated Species, then replace the Mutated Species with an probability average of all Subspecies with higher Fitness Values and determine the proportion of each Subspecies (i.e. Ssp1/Sum of Ssp1 and Ssp2);
     7b1. To determine the proportional average hex included, overlap the shape based on the center hex;
    • 7b1i. Multiply by the proportion of each Fit Subspecies to get hex total;
    • 7b1ii. If more than one of the same hex is the same position, then place this hex type in the same position on the main Species;
    • 7b1iii. If there are no hexes of the same type in the same position, then place the hex that gives the highest Fitness Value in the same position on the main Species;
    • 7b1iv. Should MP be limiting, then only position/change the hexes that give the highest Fitness Value for the lowest MP from each Fit Subspecies;
     7b2. To determine the average Membrane, first add the Membrane Fluidities of the Subspecies, and then divide by the number of Subspecies;
    • 7b2i. For determining Membrane type, add the Membrane type that is common to two or more Subspecies;
    • 7b2ii. If there are no Membranes of the same type, then place the Membrane that gives the highest Fitness Value on the main Species;
    • 7b2iii. For determining the proportional average Membrane color, add the proportion of the RGB color value for each Subspecies;
     7b3. To determine the average Behavior, first add the Behavior of the Subspecies, and then divide by the number of Subspecies for each Behavior slide;
  8. Compare Fitness Values of new Species in current and nearby locations, if feasible;
    o 8a. Pick the location with the highest Fitness Value for new Species;
    o 8b. If two or more locations have the same Fitness Value, then generate and divide by the number of location;

Old Species → (1 hex) Mutant → (2 hexes) Ecotype → (3 hexes) Subspecies → (4 hexes) New Species

image
image955×398 67.2 KB

I welcome any suggestions or improvements.

2 Likes
2

Nice work here! But why would the player play like autoevo?

1 Like
3

For fun. Plus, it would be interesting to actually see the intermediates generated by Auto-Evo. However, since players cannot see these generated intermediates themselves, this is the next best thing.

Also, congratulations on reaching Ascended!

1 Like
4

What sort of cells do you mean by “intermediates”?

1 Like
5

My understanding of how Auto-Evo used to work, and I don’t know if it still works this way, was that Auto-Evo would generate 5 intermediate “subspecies” via mutations. If a “subspecies” had a better score than the current species, then it would split off into a new “species”.

1 Like
6

Would those subspecies only last during calculations or would they be interactable with in the gameplay prior to their splitting into a new species?

1 Like
7

I think the “subspecies” only last during the Auto-Evo calculations. Someone else who is better versed in Auto-Evo can probably confirm or deny this.

1 Like
8

Are the “ecotypes” also like this?

1 Like
9

Yes, that’s the old behaviour. The new auto-evo can generate more than 100 variants to try to find a better form of the species (by finding a variant that can capture more miches). Luckily that only happens to the biggest species (which is also a major contributing factor to auto-evo taking longer in the late game) and other species will try a bit over 20 variants in an auto-evo run.

Yes, and it is one of the slowest parts of auto-evo so a perfect code upgrade would make it so that species only used for trying mutations would be eliminated entirely or more realistically just made to be recycled more to avoid unnecessary object allocations and other processing work. Relevant issue:

3 Likes
10

Within the guidelines I created? Yes, although, I only did the Subspecies step when testing the guidelines over a year ago.

In real-life? Ecotypes are deemed important enough to be studied in both Zoology and Botany. I am not an Evolutionary Biologist, so I do not know the current state of research for Ecotypes.

1 Like
11

I wonder if in future stages subspecies category will actually be present in the gameplay…

1 Like
12

One can only hope so. Also, we might need more beefy computers to handle the more intensive Auto-Evo should Subspecies be included in the calculations.

1 Like
13

Then it’s unlikely they’ll be added.

1 Like
14

I tried doing this on a new save and it made the game unreasonably hard. It was on the normal difficulty…

3 Likes
15

That’s the point of it. Autoevo doesn’t put one species above the rest.

1 Like
16

I made this a challenge for myself. So it works as a challenge.

2 Likes
17

And challenges don’t have to be fun…

1 Like