Hexagonal Life

For general discussion about Conway's Game of Life.
EricG
Posts: 199
Joined: August 19th, 2011, 5:41 pm
Location: Chicago-area, USA

Re: Hexagonal Life

Post by EricG » October 18th, 2011, 11:45 am

Golly felt clunkier using RuleTable, but after your reading your message, I increased the memory allocated, changed the step size, and everything improved.

A warning to anyone else reading this: one problem with rule tables like mine and Tropylium's is that they don't know whether the generation is odd or even, so it is possible for cells to be out of synch with each other. This could be seen as a feature, but it is a bug for emulating Golly's treatment of B0 rules. Using random fill is an easy way to accidentally end up with out-of-synch cells, and it led me to throw out a proposed spaceship-to-glider converter.

I think the answer is to create a python transition function which can use golly.getgen(),
and then use make-ruletree.py instead of using a rule table. (More things to learn how to do!)

User avatar
Andrew
Moderator
Posts: 919
Joined: June 2nd, 2009, 2:08 am
Location: Melbourne, Australia
Contact:

Re: Hexagonal Life

Post by Andrew » October 18th, 2011, 4:34 pm

I think the answer is to create a python transition function which can use golly.getgen(),
and then use make-ruletree.py instead of using a rule table.
That won't work I'm afraid. The transition function is not dynamically called every gen -- it is only interpreted *once* to make the .tree file, so the getgen() value will be constant.

If you want to generate a random pattern with only one state then use the random-fill.py script from the Golly scripts database: http://www.conwaylife.com/scripts/
Use Glu to explore CA rules on non-periodic tilings: DominoLife and HatLife

User avatar
mrob27
Posts: 1
Joined: October 19th, 2011, 2:25 am

Re: Hexagonal Life

Post by mrob27 » October 19th, 2011, 2:43 am

EricG wrote: Andrew, I'm in awe of the work that has gone into Golly, and I hope the following question doesn't sound cheeky coming from a newbie like me, but have you considered giving Golly built-in support for alternating rules?
I spent the past few hours looking at the patterns in this discussion using the old Golly and the new fixed version, and ended up duplicating a lot of what you and Andrew just did, (before noticing your recent messages :| ).

But I also wrote a Perl script that turns two "broken B0+H" rules into a RuleTable, and then used it to make RuleTables for the "B0156/S2H" and "B01245/S036H", which has that truly epic naturally-occurring spaceship. My RuleTables are a lot bigger than EricG's because I haven't found any documentation for the RuleTable language yet.

EricG, let me know if you're still interested in alternating rules, and I can smooth off some of the rough edges on the script.

137ben
Posts: 343
Joined: June 18th, 2010, 8:18 pm

Re: Hexagonal Life

Post by 137ben » October 26th, 2011, 10:21 pm

B2S3H seems to be the most like life (along with its close variations, e.g. b26s3, which is almost identical). Patterns tend to stabilize, and a glider exists, though it is rather rare.
Outer-totalistic hexagonal rules are of a particular interest to me. Does anyone know if there is a compendium of known hexagonal life-like gliders, similar to Eppstein's site for life-like rules? If not someone should make one.

Post Reply