Is auto-evo biased against heterotrophs? I see an awful lot of chemotrophs in my runs, and I’m wondering if auto-evo likes them because they can supply themselves with food.
Not sure about this. Don’t know how to phrase it. Might be an issue?
Is auto-evo biased against heterotrophs? I see an awful lot of chemotrophs in my runs, and I’m wondering if auto-evo likes them because they can supply themselves with food.
Not sure about this. Don’t know how to phrase it. Might be an issue?
Fairly sure this happens because there are many means for autotrophic species to ward off heterotrophs, leading to these later ones having a harder time surviving.
Those heterotroph wards don’t seem to work when my Gigantic Engulfer #46 genocides a patch for food.
Player cells are a whole different breed from the auto-evo cells. Furthermore they are guided by an actual person.
I think auto-evo also doesn’t really have an accurate model of what serves as a good defense.
When I eat a toxic cell, it will sit there until it is digested or I am about to die, which is not very good for either of our fitness. Luckily, that doesn’t matter, because I am a living God and cannot be defeated by any mortal cell.
When my cilia are gravitating the local spikey boi, I will eat him. I don’t care that he has a spike. I am big and strong, and he only has one, singular spike. He could have six spikes, for full protection, but auto-evo cells seem to never evolve more than one.
Auto-evo has no effect on the gameplay part of Thrive. So you are mixing up the microbe AI component and auto-evo. Though I’ll grant you that the auto-evo simulation needs to encourage changing behaviour values to ones that the microbe AI would then look for and make the gameplay cells act better.
Speaking of, I’ve heard that auto-evo supposedly takes a few behavior statistics into account. Which ones?
You likely won’t get an answer to your question, but I’ll mention how someone could come up with the answer: by checking the usage of the behaviour variables within the auto-evo SelectionPressure
derived classes.
I could have an idea on how to balance the behaviour editor but it is probably not a good solution.
If the player doesn’t follow the rules of their species’ behaviour they set (like trying to move while being a sessile-ish creature), perhaps they could end up doing worse (like having their stamina bar run out faster when moving around, or having a worse time detecting prey) than if they followed what they set. Maybe following the behaviour rules would even yield some extra benefits for the player?