Searching for Conduits and Reflectors in Sparse Rules

For scripts to aid with computation or simulation in cellular automata.
Post Reply
Posts: 26
Joined: March 9th, 2021, 4:53 pm

Searching for Conduits and Reflectors in Sparse Rules

Post by Francisco » April 10th, 2022, 12:08 pm

I am trying to search for guns in the generations rule 23/3/3. I have yet to see any kind of gun in the rule so I am trying very hard not to narrow my search space I.E. I am looking for the most common/most easily searchable kind of gun in the rule. Efforts to look for the kind of volatile oscillators common in most guns though algorithms like apgsearch or a modified version of torus have not yielded much luck, making me suspect that the rarity of volatile, rapidly growing and changing patterns in the rule would make it easier to search for guns by first searching for conduits, converters, reflectors, splitters, etc. and constructing a track out of them.

What is the best way to search for these kinds of patterns in this kind of rule? I considered creating a modified version of catalyst(search algorithm) of catforce, but I wonder if searching only for stable catalysts is narrowing the search space too much, given that so many of stable catalysts in normal CGOL rely on some of the still lives being completely destroyed and replaced, and the fact that in 23/3/3 explosive patterns are rare and ones that leave behind debris are rarer. I am considering the possibility that taking advantage of sparking oscillators like the ones below is the best way to find catalysts, however I worry that may expand the search space too much.

Code: Select all

x = 71, y = 25, rule = 23/3/3
Here is also an eating behavior that may also prove useful for searching for catalysts.

Code: Select all

x = 31, y = 25, rule = 23/3/3
Any help would be appreciated, thank you.

User avatar
Posts: 204
Joined: June 15th, 2020, 6:20 pm

Re: Searching for Conduits and Reflectors in Sparse Rules

Post by silversmith » May 20th, 2022, 9:03 pm

One strategy which could lead to some more advanced circuitry are periodic conduits for the b-heptomino.

To search for these conduits, I can think of two possible approaches. One is to find some more, higher period sparkers, for which the only program I can think of is rlifesrc. The other is to try enumerating how sparks can interact with the b-hept and other objects. It could be similar to hotdogPi's script, except for generations rules. I am currently working on a program which might work for this purpose.

Here is an example of a partial conduit.

Code: Select all

x = 41, y = 41, rule = 23/3/3
Edit: A working conduit

Code: Select all

x = 40, y = 33, rule = 23/3/3
A simulator with the tools I couldn’t find elsewhere:

Post Reply