Talk:Conway's Game of Life

Rewrite

This article could actually use a rewrite since right now it is more or less just the Wikipedia article. I'll think about re-working some of the sections to make the examples and whatnot more relevant given the larger collection of Life pages available here. Nathaniel 03:43, 4 February 2009 (UTC)

Yeah, I noticed. The beginning is probably still fine since wiki's introduction is okay, but you should probably go a bit more in-depth further on. (meanwhile, I'll contribute some more in the evening, and I've got a Game Theory midterm to study for till then) Elithrion 14:08, 4 February 2009 (UTC)

Variations on Life

A note on the Variations on Life section: I have found that rules for life-like cellular automata are more commonly written in the form "Bxx/Syy", 'xx' and 'yy' being a list of numbers whose values range form one to eight. Those numbers represented by 'xx' would be the number of live cells around a dead cell necessary to cause a birth, while those numbers represented by 'yy' would be the number of live cells surrounding a live cell that would cause that cell to survive; so the rule for Life would be "B3/S23". Currently the section is written such that the rule is formatted "yy/xx", with survival first and birth second, and with no label to easily identify the meaning of each set of numbers. I have found that the format which I described earlier is far easier to read and avoids the ambiguity of the current format. I therefore suggest that it be changed.

Update: I see this is mentioned at the "rules" section on the cellular automaton page, but the format throughout the wiki seems to be the "S/B" form. I propose this be changed to the other form ("Bxx/Syy") to make it slightly clearer.

01:31, 20 April 2009 (UTC)

Sokwe 22:23, 19 April 2009 (UTC)

That switch is fine by me, as long as the wording of the rules section is modified to point out that we're using B/S as the standard rather than S/B. Also, I still want S/B described there, since it's also widely-used by other sources. Also, at some point we should probably move the variations on Life section to the cellular automaton article (with a different title) since it doesn't make quite as much sense in LifeWiki as it did in Wikipedia. Maybe a brief subsection describing Life-like cellular automata could be included in this article (with a "Main article: Life-like cellular automaton") thing included here though. Nathaniel 11:44, 21 April 2009 (UTC)

Chaos

Do any infinitely growing patterns exist that grow chaotically (DotLife or DryLife-esque)? 8:32AM 5/28/2017 NY time

In the sense that I think you mean -- a methuselah that creates an increasing patch of random ash, but keeps growing indefinitely -- it doesn't seem too likely. If the pattern is truly chaotic, then as T approaches infinity the probability approaches 1 that the chaos will eventually settle down, obeying a power law that was demonstrated fairly clearly by Nathaniel's online soup-search project several years back. Of course many patterns will keep growing indefinitely, but it will be by some mechanism along the lines of a switch engine, with a definite pattern to it.

If the odds of an N-tick methuselah decrease by a factor of two for every thousand ticks, you can make a rough guess that the longest-running chaotic methuselah starting from a 20x20 soup might take between 100,000 and 1,000,000 ticks to settle down -- but no one is going to find out for sure any time soon.

However, see Nick Gotts' experiments for examples of apparently unending novelty generated by various kinds of puffers. This is more like "controlled chaos" than the randomly-growing-blob behavior you're asking about.

-- This would be a much better question for a forum thread like the basic questions discussion. As you can see, this talk page hasn't had any activity since 2009, and should probably be left for discussing issues related to the actual content of the page. Answers aren't nearly as likely to show up in random places like this, especially if you don't sign your name...! I'm only looking here because I'm going through and patching up some of your edits from yesterday. Dvgrn (talk) 09:27, 29 May 2017 (UTC)

Random soup

Re: this edit:
> putting back soup; the animation of the octomino is nice but I don't see why we can't have the soup as well, it gives a better picture of the nature of Life
One problem is that these random soups break a convention "when the infobox pattern is clicked, show the pattern in a new window". On many pattern pages (for example Glider), the infobox shows a pattern, and clicking on the pattern launches a LifeViewer window. Random soups shown for rules do not follow this convention -- when the soup is clicked (intentionally or accidentally), it expands on the page. A reader who accidentally clicks on the soup can be puzzled what happened and how to go back. I tried to solve this before, but could not find how to fix the template.
Another problem is that the initial soup takes only a small fraction of the available display area, and cells are very small. Maybe a randomly filled torus with larger cells would look better?
In general, I think it would be better to have a non-random pattern (in the infobox or outside the infobox), chosen by a human being to illustrate how the rule works, with the viewer adjusted accordingly. See also Template talk:Rule. Confocal (talk) 01:43, 2 August 2023 (UTC)

See also this forum post by me in the "lifeviewer bug" thread, and the accompanying test page. Confocal (talk) 09:54, 8 August 2023 (UTC)

Is CGoL Turing-complete?

标题

--LittleLWSS(talk) 2024年8月5日 15:21（一）农历七月初二