On gliders breaking down still life sets

[Rhombic]

Code: Select all

x = 18, y = 10, rule = B3/S23
17bo$15b2o$b2o3bo9b2o$o2bobobo$o2bobo2bo$b2o3bobo$7bo$4bo$3bobo$4bo!

I leave that example for reference, where a glider collides with that set of 3 still lifes producing a 1708-generation development. From most other angles, the collision results in the fairly boring appearance of blocks and blinkers, together with the odd loaf, ending at a much lower number of generations.

The ways gliders collide with small-sized still lifes have been thoroughly studied, but we will usually encounter still life patterns close enough together that the "reaction" of one will the glider will eventually include the other nearby ones. This makes it very difficult to analyse, since the relative positions of blocks, tubs, loaves, etc. can be totally arbitrary and have significantly different results, as well as the angle and position of attack of the glider.

Hence, I would like to know whether some general rules/ideas could be drawn about how still lifes will generally interact with each other. Even if rudimentary and basic, it will probably save time from not attempting attacks from unlikely positions. (Note: one of my ideas to think about this has been to follow glider syntheses, but it is way too complex to approach either way)

[mniemiec]

Hence, I would like to know whether some general rules/ideas could be drawn about how still lifes will generally interact with each other. Even if rudimentary and basic, it will probably save time from not attempting attacks from unlikely positions. (Note: one of my ideas to think about this has been to follow glider syntheses, but it is way too complex to approach either way)

Life behavior is extremely chaotic. Even a change of a single bit in a large pattern can have extreme repercussions to the outcome. One observation that does tend to be true, more often than not, is that patterns (and sub-patterns) with odd axes of symmetry are more likely to expand than ones with even axes - e.g. Pi heptominos, B heptominos, traffic lights, honeyfarms, shuttles, pentadecathlons, etc.

[Gamedziner]

Code: Select all

x = 18, y = 10, rule = B3/S23
17bo$15b2o$b2o3bo9b2o$o2bobobo$o2bobo2bo$b2o3bobo$7bo$4bo$3bobo$4bo!

I leave that example for reference, where a glider collides with that set of 3 still lifes producing a 1708-generation development. From most other angles, the collision results in the fairly boring appearance of blocks and blinkers, together with the odd loaf, ending at a much lower number of generations.

Part of what makes this take so long is that generation 41 contains the Pi heptomino.

Life behavior is extremely chaotic. Even a change of a single bit in a large pattern can have extreme repercussions to the outcome. One observation that does tend to be true, more often than not, is that patterns (and sub-patterns) with odd axes of symmetry are more likely to expand than ones with even axes - e.g. Pi heptominos, B heptominos, traffic lights, honeyfarms, shuttles, pentadecathlons, etc.

This is likely because you have to have one or three cells near each other with odd symmetry, and 3 cells cause another cell to be born.