Thread for basic questions

[Book]

A gliderless methuselah is a relatively long-living pattern that does not emit any escaping gliders.

emu methuselah

[mniemiec]

A gliderless methuselah is a relatively long-living pattern that does not emit any escaping gliders.

emu methuselah

It's right in the name: methuselahs = slash the emu!

[Book]

What is the point of inflation as defined in the wiki article of that name? I ask because the article begs for such an explanation.

[confocaloid]

What is the point of inflation as defined in the wiki article of that name? I ask because the article begs for such an explanation.

Is it unclear what "inflation" means, as defined there? I think the first sentence defines it.

It could be useful to add an explanation how to interpret Catagolue soups for pseudosymmetries like iD2_+1 or iiii8x32. For asymmetric soups (iC1, iiC1, iiiC1, ...) I can take the soup from Catagolue and scale it by a factor of 2^k if there are k i's. For other pseudosymmetries I do not know how to get the original soup from the soup returned by Catagolue.

[Book]

What is the point of inflation as defined in the wiki article of that name? I ask because the article begs for such an explanation.

Is it unclear what "inflation" means, as defined there? I think the first sentence defines it.

It could be useful to add an explanation how to interpret Catagolue soups for pseudosymmetries like iD2_+1 or iiii8x32. For asymmetric soups (iC1, iiC1, iiiC1, ...) I can take the soup from Catagolue and scale it by a factor of 2^k if there are k i's. For other pseudosymmetries I do not know how to get the original soup from the soup returned by Catagolue.

The issue is not the meaning of inflation. The issue is why do it at all? Certainly not just because you can. So what is the value of inflating a pattern?

[confocaloid]

The issue is not the meaning of inflation. The issue is why do it at all? Certainly not just because you can. So what is the value of inflating a pattern?

Well, the same question can be asked about many things that the community apparently finds worthwhile to do. I guess that depends on who you're asking.

For what it's worth (probably not much), I think it's exactly just because you can. For one thing, inflation is a well-defined operator (which in itself is sufficiently rare if you think about it), and moreover it can be repeated multiple times. Then one can (just to give an example) run parallel censuses in C1 and iC1 (and iiC1, ..) and compare frequencies of occurring objects.

If you are asking about some other "value" - perhaps about some "value for the community" - then probably I cannot answer that question, simply because I do not think I have any right whatsoever to say "we" and speak from the community. I'm a mere hobbyist who happens to be interested in cellular automata.

[HotWheels9232]

Why is it OCA DOTY and not OCA POTY?

[Ian07]

Why is it OCA DOTY and not OCA POTY?

Because entire rules are often submitted as a single entry.

[confocaloid]

If you enlarge the tub by a factor of 3 and evolve the resulting 36-cell pattern for six ticks, you will get the tub back.

Code: Select all

x = 9, y = 9, rule = B3/S23
3b3o$3b3o$3b3o$3o3b3o$3o3b3o$3o3b3o$3b3o$3b3o$3b3o!
#C [[ GRID STOP 6 ]]

Is there a (nonempty) pattern which has the 2x-scaled copy of itself as a predecessor?
Edit: here is a 34-tick solution:

Code: Select all

x = 6, y = 4, rule = B3/S23
4o$4o$2o2b2o$2o2b2o!
#C [[ GRID STOP 34 ]]

[volcanrb]

Is there a known lower bound on the possible asymptotic growth rate for the population of a pattern in CGoL wrt number of steps? For example could there exist a pattern whose growth rate is slower than log^(i)(t) for all i, where log^(i) denotes log composed with itself i times?

[pzq_alex]

Is there a known lower bound on the possible asymptotic growth rate for the population of a pattern in CGoL wrt number of steps? For example could there exist a pattern whose growth rate is slower than log^(i)(t) for all i, where log^(i) denotes log composed with itself i times?

If a pattern grows without bound, then at generation T its history (cells that ever have been on) has a diameter (minimum d such that it fits in a d by d box) of Omega(sqrt(log(t))).

[dl-rs]

Uh what does the numbers in a tree rule mean?

[dvgrn]

Uh what does the numbers in a tree rule mean?

See Golly's documentation on @TREE format.

[GUYTU6J]

Is there a known lower bound on the possible asymptotic growth rate for the population of a pattern in CGoL wrt number of steps? For example could there exist a pattern whose growth rate is slower than log^(i)(t) for all i, where log^(i) denotes log composed with itself i times?

See recursive filter and exponential filter.

---

Why does the term "first natural block" refer to the first block dropped by Herschel instead of one laid by B-heptomino? The latter appears at T=43 of the R-pentomino (or T=15 for B-heptomino) when Herschel is yet to form.

[dvgrn]

Why does the term "first natural block" refer to the first block dropped by Herschel instead of one laid by B-heptomino? The latter appears at T=43 of the R-pentomino (or T=15 for B-heptomino) when Herschel is yet to form.

Well... it's for the same reason that "first natural glider" refers to the first glider emitted by a Herschel, not the first glider ever seen. Herschels were the reference object for the universal signal circuitry that was under development in the mid-1990s, when both of these terms were invented. Or at least when the "first natural glider" term was invented -- I'm not sure anybody ever actually said "first natural block" until the next millennium sometime.

I think that the wiki's current statement that "named for being the first naturally occurring glider that was observed in the early days of Life research (by Conway's group when tracking the evolution of the R-pentomino)" is not really true -- or maybe it's sort-of-true but misleading. "First natural glider" was never intended to mean The First Ever Natural Glider though it does happen to be that.

I can't even find any uses of "first natural glider" before some email messages in 2003 where I mentioned them several times. That doesn't mean I invented the term, though: Stephen Silver posted (via email to LifeCA) a list of Herschel-to-glider converters in 1998, and labeled two of them...

"1st natural glider"
and
"2nd natural glider".

And those two strings with "1st" and "2nd" seem never to have gotten used again! But the abbreviation "FNG" didn't come along until much later; I seem to remember not liking it much when it first showed up.

Anyway, the "first natural glider" was so named specifically to distinguish it from the "second natural glider", and specifically in the context of Herschels. Same with "first natural block", though really that term has never gotten much of any actual use -- I've only ever seen the abbreviation "FNB" a couple of times, for example, here on the forums.

[GUYTU6J]

...
20P2 has 21 cells.
21P2 has 16 cells.
22P2 has 20 cells.
23P2 has 17 cells.
24P2 has 18 cells.
25P2 has 18 cells.
26P2 has 20 cells.

None of these objects seem to have those names in Achim Flammenkamp's census, or in Mark Niemiec's database, or on pentadecathlon.com, or in the Life Lexicon, or in any old email discussions that I can find.

I'd say this is an old non-standard naming convention left over from when the LifeWiki was first being created over a decade ago, and it's somehow persisted all this time without anyone noticing that they weren't real names. Let's switch over to naming those objects by apgcode, ASAP, and remove the redirects from "{n}P2".

These have been tabulated in beacon variants. I am going to change the original articles to redirects to that page because I think they are not notable.

[hotdogPi]

I support the change. I don't want to lose any information, though, such as why the common ones are more common. (I added a paragraph, but it could be expanded with more information on each one where it applies.)

[mniemiec]

...
20P2 has 21 cells.
21P2 has 16 cells.
22P2 has 20 cells.
23P2 has 17 cells.
24P2 has 18 cells.
25P2 has 18 cells.
26P2 has 20 cells.

None of these objects seem to have those names in Achim Flammenkamp's census, or in Mark Niemiec's database, or on pentadecathlon.com, or in the Life Lexicon, or in any old email discussions that I can find.

I'd say this is an old non-standard naming convention left over from when the LifeWiki was first being created over a decade ago, and it's somehow persisted all this time without anyone noticing that they weren't real names. Let's switch over to naming those objects by apgcode, ASAP, and remove the redirects from "{n}P2".

I had always thought that these originally came from Achim's census - just arbitrary name assigned sequentially to various frequently-occurring P2s. I agree that the names are obsolescent, and should be deprecated.

[GUYTU6J]

I support the change. I don't want to lose any information, though, such as why the common ones are more common. (I added a paragraph, but it could be expanded with more information on each one where it applies.)

I have added them to the Notes section. Now please check the commonness data...

Two related questions:

1) For position among natural period-2 oscillators on Catagolue, I checked the b3s23/C1/xp2 census. Can someone check the position among natural oscillators or natural objects?
2) For position on Achim's census, how did they check those below 50 in 2009? This page only lists the top 50 or so.

[Ian07]

2) For position on Achim's census, how did they check those below 50 in 2009? This page only lists the top 50 or so.

Here is the full list.

[ihatecorderships]

What is the difference between a Garden of Eden and a orphan?
Different question: Why does Kazyan not have a green name? I have read in a few places that they have the powers of a moderator.

[Ian07]

What is the difference between a Garden of Eden and a orphan?

If I understand correctly, an orphan is the set of cells required to be either on or off for a pattern to have the property of being a Garden of Eden, and it can therefore be extended in any way past that and still be a GoE.

[GUYTU6J]

Take the infamously large methuselah 40514M, and shift one or two of the five subpatterns by any offset in the range (±10, ±10) with respect to its/their original position. Do all of the resulting patterns that do not exhibit infinite growth settle within a lifespan below 40514 ticks?

[hotdogPi]

Take the infamously large methuselah 40514M, and shift one or two of the five subpatterns by any offset in the range (±10, ±10) with respect to its/their original position. Do all of the resulting patterns that do not exhibit infinite growth settle within a lifespan below 40514 ticks?

There are 1.06×10^14 soups in C1 so far.
There are about 130 methuselahs that last at least 40514 ticks in C1.
This is a ratio of about 8×10^11.
Moving any object within ±10/±10 is a 21×21 box, ×2 because blinkers can be in either phase. (21×21×2)^2 * (5 choose 2) ≈ 7.8×10^6.

It's likely that it's the longest if a maximum of two (or even three) can be shifted, but if you shift four or five, there's likely to be one that lasts longer.

[dl-rs]

I'm uh unable to download the pdf from Book, read it online, buy it or open Lulu.com. Does anyone have a PDF?