Kite Template Challenge

Post your own challenge to your fellow researchers!
Post Reply
User avatar
wapcaplet
Posts: 363
Joined: Thu Dec 08, 2016 9:52 pm

Kite Template Challenge

Post by wapcaplet » Wed Dec 28, 2016 8:42 pm

We have so many difficult challenges! Here is one that's meant to be easy(ish) and fun.

Substrate: wapcaplet-challenge-kite-template.substrate
kite-template.png
kite-template.png (5.25 KiB) Viewed 2759 times
On this substrate are some very basic "kite" or "diamond" smart swimmers. They can maintain a population of 60 to 80 pretty well, but rarely exceed 100 cells. For this challenge, load the kite's genome from the M1 cell in the microscope--this will be your template.

Rules: You must use the template M1-M7 cells, and may not use any modes M8-M20, nor are you allowed to change the existing split modes. But you can modify any sliders, adhesin settings and other checkboxes for any mode M1-M7. In other words, you may only make "point mutations" to the template organism. You may do this with evolution or by engineering.

Goal 1 (easy): Derive a new kite swimmer that can out-compete the white swimmer from 1 initial cell, leaving no original white cells.
Goal 2 (difficult): Get your swimmer to reach a stable population of at least 200 cells.
Goal 3 (?): Check "Keep adhesin" for M1's children, turning the swimmer into a serpent. Keeping these on, derive an organism that can still out-compete the individual white swimmers.

For goal 2, I have reached 200 cells but cannot maintain it. The higher-difficulty goals will be dependent on how well Alast and Bwisialo's optimized kites can do here--I am sure it's impressive :-) I have not come anywhere close to goal 3, but it seems possible--opinions?

Tips:
  • After loading M1 to the genome editor, change the color of a few cells before saving your genome.
  • When trying out alterations to your cells, give them a new color as well, so you can better observe how they compare.
  • Relative nutrient priorities play a big role in reproduction rate
  • Flagellocyte swim speed and angle can improve accuracy
  • (others?)
User avatar
bwisialo
Posts: 2953
Joined: Sat Aug 15, 2015 10:22 pm

Re: Kite Template Challenge

Post by bwisialo » Thu Dec 29, 2016 4:22 am

Nice idea. Many in-app challenges are focused on optimization, like Macrophages IV. And there's the basic template of the standard swimmer, though thete are plenty of other approaches. But there aren't really straightforward ways for people to learn how to optimize. People have to learn it insofar as challenges require it, but they mostly have to learn on there own. I'm not sure that it's realistic to provide tutorials or instructional challenges to "teach" some optimization basics, but something like that would be good, especially for doing challenges after Friction. Anyway...

1 is easy with basic changes, as intended

2 seems not so difficult but maybe not so easy either. I made some of typical changes and got 425, sustaining 275.

3 is tough partly because individual swimmers are so much more efficient, but also because the substrate is small, and nutrient settings lead to near depltion of all food chunks at times, all of which deliver huge disadvantages to a serpent species trying to get off the ground and compete.
amor fati
User avatar
wapcaplet
Posts: 363
Joined: Thu Dec 08, 2016 9:52 pm

Re: Kite Template Challenge

Post by wapcaplet » Fri Dec 30, 2016 12:00 am

Using the starting substrate, a few thousand simulated hours of evolution at modest mutation rate (0.010 - 0.020) produced a winner that can maintain ~240 cells--so goal 2 is pretty easy after all :-) Maybe 300-400 not so much?
User avatar
TragmatiS
Posts: 154
Joined: Sun Nov 13, 2016 5:07 pm
Location: Russia

Re: Kite Template Challenge

Post by TragmatiS » Thu Jan 19, 2017 4:33 pm

Goals 1 and 2 solved via slight adjustments. Easier than expected. Maybe 300?
I`m Russian so there may be typos.
Post Reply