Better evolution

We like few-celled organisms, not bugs!
Post Reply
Dries Demeire
Posts: 3
Joined: Sat Nov 12, 2016 9:56 pm

Better evolution

Post by Dries Demeire » Sun Jun 16, 2019 1:24 am

Dear Petter I hope you read this.
I was trying to evolve an organism and was thinking why it is so hard in the game to evolve cool organisms from scratch without human help. As i was thinking i thought of how evolution works in nature.

And suddenly i had a eureka moment:
My idea is that instead of having discrete cell modes you had a mix off cel modes. For example:
a 100% fagocyte (takes up food at normal rate)

Can transition to a 99% fago 1% flagello (takes up food at 99% speed)(swims 1%speed for same amount of energy normal flagellocyte uses)

to a 100% flagellocyte.

This way a cell can get just that tiny bit better by shifting 1% away from it's previous state and evolution will occur.

Cells are always a combination off all cel modes and the sum of all percentages of cell modes is 100% this would mean that all cell types would need some kind of scalability for example:
50% gluocyte (kind of sticky but not as much)
secrocyte (more nutrient per expelled substance)
Lipocyte (50 less dense nutrient storage)

Same for all other functions for exemple:
In stereocyte : 100% off excretion distributed between all modes (cell/wall/light/speed)


You could say that what i'm saying is that everywhere where you have to pick something, should actually be a slider. Apart from cell modes (M1,M2 etc.)
And to keep thinks simple you can make this turn on, only when you toggle pro mode so newbies still have the simple version.

Or I was thinking no cell modes anymore (fago, flagello etc, )
But 1 cell with all sliders of all types.
And every slider has a nutrient cost when it's higher.
For cell properties like sensors and signal transmission i would make it very low cost so "intelligent behaviour" can develop and maybe logic gates even?

This would make for a better evolution but possibly very hard to make (or you have your own vision which i respect of ofcourse) i have no idea. But i am super enthusiastic thinking about it!
If you do it i'll donate again 😋
I am nothing but cells
User avatar
countingtls
Posts: 25
Joined: Mon Jun 10, 2019 10:36 am

Re: Better evolution

Post by countingtls » Mon Jun 17, 2019 3:02 pm

Dries Demeire wrote:
Sun Jun 16, 2019 1:24 am
Dear Petter I hope you read this.
I was trying to evolve an organism and was thinking why it is so hard in the game to evolve cool organisms from scratch without human help. As i was thinking i thought of how evolution works in nature.

And suddenly i had a eureka moment:
My idea is that instead of having discrete cell modes you had a mix off cel modes. For example:
a 100% fagocyte (takes up food at normal rate)

Can transition to a 99% fago 1% flagello (takes up food at 99% speed)(swims 1%speed for same amount of energy normal flagellocyte uses)

to a 100% flagellocyte.

This way a cell can get just that tiny bit better by shifting 1% away from it's previous state and evolution will occur.

Cells are always a combination off all cel modes and the sum of all percentages of cell modes is 100% this would mean that all cell types would need some kind of scalability for example:
50% gluocyte (kind of sticky but not as much)
secrocyte (more nutrient per expelled substance)
Lipocyte (50 less dense nutrient storage)

Same for all other functions for exemple:
In stereocyte : 100% off excretion distributed between all modes (cell/wall/light/speed)


You could say that what i'm saying is that everywhere where you have to pick something, should actually be a slider. Apart from cell modes (M1,M2 etc.)
And to keep thinks simple you can make this turn on, only when you toggle pro mode so newbies still have the simple version.

Or I was thinking no cell modes anymore (fago, flagello etc, )
But 1 cell with all sliders of all types.
And every slider has a nutrient cost when it's higher.
For cell properties like sensors and signal transmission i would make it very low cost so "intelligent behaviour" can develop and maybe logic gates even?

This would make for a better evolution but possibly very hard to make (or you have your own vision which i respect of ofcourse) i have no idea. But i am super enthusiastic thinking about it!
If you do it i'll donate again 😋
After running like millions of h of and hundreds of different pure evolution simulations. I find that cells using Photocyte as energy source has no trouble getting more complex and interconnected over time, they will use what every possible to stay as long as the basic self replicating Devorocyte existed to push them toward the edge and defense, or gravity to force them to float, or rotating light to store energy, they tend to get pretty complex as long as they are just on the verge of getting wiped out.

The cells using Phagocyte as energy source tend to be slightly different, they often evolved into utilize "single cell" switch, with uneven split ratio, I think that's the easiest way of moving around by "exploding" one of the new cell, and at the same time they can "reproduce" when there is cluster of food dots. If adding gravity they tend to sink and float with multiple types of Buoyocyte. They can evolve to using Flagellocyte to run in circle or with two cells in the front to be more mobile, the trouble is the distance between food needs to be sufficiently far part, but at the same time enough of them exist so they can reproduce and let evolution to take its course, so they tend to evolve much slower with much lower total number, while Photocyte energy types can easily explode to the max.

And the last type of using Devorocyte as energy source, or parasite using Glueocyte are the hardest. They rely heavily on the above two to shrive. They are also the easiest to "stay alive", they can be very successful without any complex splitting chain, just divide and divide. I've only seem mobile Devorocyte using the exploding new cell trick to move around (and reproduce at the same time), and acting as mobile hunter for "chasing" Phagocyte, or eating the edge Photocyte, but they are hard to stay, either needing way too high density of food for their reproduction, and just stay that way, or when Photocyte evolved to include Keratinocyte in their chain and just cannot get enough energy and spread as fast as the simpler divide and connect chain of Devorocyte. Glueocyte energy is the same, they just need a very rapid breeding host to exist and divide happily, however just stay that way, pushing to stay simpler when Keratinocyte start to appear.

The more complex cell type, I often found them as "excess" features out of the long stabilized species, they have extra energy to burn and have no trouble adding extra cells in their "life cycle", sometimes they are even helpful in identifying where the mutation started. But I think these kind of "neutral" mutation most often just way too random to be included in the existing function. It's like random "contamination" all over again, but this time much slower and depending on the mutation rate, and the number of "species" around, which is much lower in total number, normally not over 5, even on a very large dish. So they might eventually start to have real function over a very very long period of time, purely by chance it would needs to be very lucky to get the right combo and parameters.

(Virocyte usually tends to be too sucessful and screw things up, clean the dish by accident too easily thus restart from scratch, just don't include them)

I think indeed the trick is always about how to link reproduction with energy function in the chain. The simpler it is, the easier for evolution to begin its work. The rest is more about scale and quantity and time, and then the interaction between "species" as a higher level. The trouble is if they don't need to be complex enough to survive, they will be happily stay as they are. Doing enough mutation experiments I tend to find those can survive a little radiation over a very long period of time is often those with redundant chains (copy of the similar function) several times, so even if they behave like the same, they are actually multiple different Modes of the same function. The much delicate complex chain we players designs are simply lacking this protection, and easily degraded. Genes tend to have redundancy for protection, and a way to protect the reproduction process so evolution can continue. Right now this mechanism is most often act at the cell mode level, where its "gene" sequence is actually fairly limited with only 40 lines of codes so to speak to work with. Imagine only able to write a program with just 40 lines of code they either need to be very simple or very intricate and easily break.

The idea of self sufficient cell with all the reproduction and different functions wrapped into one with sliders is interesting, but I think it would only push the issue further down into the "slide scale", which will be much more difficult to check. Although I think it would mean each cell is essentially its own species, and might have the opposite effect of making them lived way too easily where just one cell is enough, and never needing to create any interconnected function from others and never needs to be specialized.
Bl0z
Posts: 2
Joined: Tue Jul 02, 2019 8:21 pm

Re: Better evolution

Post by Bl0z » Thu Jul 04, 2019 2:30 am

I really thought exactly the same as you, before this post of course. I recently published a post about mutations, if you want to check it out
Post Reply