For general discussion about Conway's Game of Life.
dvgrn
Moderator
Posts: 6710
Joined: May 17th, 2009, 11:00 pm
Contact:

### Re: Thread for basic questions

Moosey wrote:
toroidalet wrote:Actually, the lifespan of bounded patterns is the busy beaver function of one of its sides (provided both sides are long enough; the other side can be constant), since there are finite Turing machines.
What would the answer be if Turing machines are not allowed? Just, hypothetically
Well, I just ran some crackpot numbers based on the current 16x16 methuselah results from Catagolue. If each 1K bin on average contains 5/9ths of the number of methuselahs in the previous bin, then if we ran all 2^256 possible 16x16 patterns and distributed them among 1K bins, the last bin that would be likely to have a methuselah in it would be the 329K bin.

Up to now we've found methuselahs that last as long as 47,575 ticks, but according to my very inaccurate math there should be one that lasts almost 330,000 ticks.

Slight changes in assumptions will make huge differences in the estimate of longest-lived 16x16, but it does seem pretty likely that the longest lifespan will be six digits long -- unless some lucky weird mechanism shows up that isn't like the random ash we've seen so far. Maybe there's something based on interacting natural glider guns, for example.

A for awesome
Posts: 2036
Joined: September 13th, 2014, 5:36 pm
Location: 0x-1
Contact:

### Re: Thread for basic questions

dvgrn wrote:-- unless some lucky weird mechanism shows up that isn't like the random ash we've seen so far. Maybe there's something based on interacting natural glider guns, for example.
Personally, my money would be on an ark that shoots backward gliders at debris, which, after a large number of ticks (possibly as many as millions, given the size of the search space), sends a glider or gliders forward into the engine, destroying it. (There are similar small symmetrical examples in Golly's pattern collection IIRC.)
x₁=ηx
V ⃰_η=c²√(Λη)
K=(Λu²)/2
Pₐ=1−1/(∫^∞_t₀(p(t)ˡ⁽ᵗ⁾)dt)

$$x_1=\eta x$$
$$V^*_\eta=c^2\sqrt{\Lambda\eta}$$
$$K=\frac{\Lambda u^2}2$$
$$P_a=1-\frac1{\int^\infty_{t_0}p(t)^{l(t)}dt}$$

http://conwaylife.com/wiki/A_for_all

Aidan F. Pierce

dvgrn
Moderator
Posts: 6710
Joined: May 17th, 2009, 11:00 pm
Contact:

### Re: Thread for basic questions

A for awesome wrote:Personally, my money would be on an ark that shoots backward gliders at debris, which, after a large number of ticks (possibly as many as millions, given the size of the search space), sends a glider or gliders forward into the engine, destroying it.
Stopping both switch engines in an ark seems like a fairly tough trick shot, maybe more likely with multiple synchronized gliders (?). I'm not motivated enough to try to calculate the odds that that will happen starting from some 16x16 soup, but I suppose 2^256 is big enough that it's pretty likely to be out there somewhere.

If one switch engine gets away, then we're in a category of long-lived pattern that seems not to count, or at least apgsearch isn't designed to collect it as a methuselah.

Patterns that "go boring" with infinite growth don't seem to be getting added to the official Methuse List, even though now there's finally a non-symmetrical 16x16 soup that makes a long-lived infinite growth pattern.

So my basic question #1 is, do people think it would be okay to add a separate section to the Methuse List for neverending-population-growth long-lived patterns with small bounding boxes, say up to 32x32? Seems like there's some interesting stuff out there, that might be worth tracking as long as they don't get mixed up with regular methuselahs which are only neverending-bounding-box-growth.

Basic question #2: are there any Gottsian 32x32 starting patterns for which it's not even known if they ever go boring or not?

Moosey
Posts: 3268
Joined: January 27th, 2019, 5:54 pm
Location: A house, or perhaps the OCA board. Or [click to not expand]
Contact:

### Re: Thread for basic questions

dvgrn wrote:
A for awesome wrote:Personally, my money would be on an ark that shoots backward gliders at debris, which, after a large number of ticks (possibly as many as millions, given the size of the search space), sends a glider or gliders forward into the engine, destroying it.
Stopping both switch engines in an ark seems like a fairly tough trick shot, maybe more likely with multiple synchronized gliders (?). I'm not motivated enough to try to calculate the odds that that will happen starting from some 16x16 soup, but I suppose 2^256 is big enough that it's pretty likely to be out there somewhere.
Couldn’t you just break one SE so that the other is destabilized?
I am a prolific creator of many rather pathetic googological functions

My CA rules can be found here

Also, the tree game
Bill Watterson once wrote: "How do soldiers killing each other solve the world's problems?"

dvgrn
Moderator
Posts: 6710
Joined: May 17th, 2009, 11:00 pm
Contact:

### Re: Thread for basic questions

Moosey wrote:Couldn’t you just break one SE so that the other is destabilized?
Theoretically, maybe. In practice that adds a whole lot of unlikeliness to the equation. It all depends on what switch engine pair we're talking about. The pair has to be firing backward gliders. Usually those arrangements protect themselves fairly well from gliders coming from behind. For example, see if you can find a forward glider that disrupts even one switch engine of the two in the 133,100-tick not-methuselah.

Run it in LifeHistory to check for targets. There are a few ticks where you could disrupt the backward-firing Herschel, where the reaction envelope just barely sticks out past the backward glider stream. On the opposite side there are no blue cells sticking out beyond the repetitive ash at all. So that's not a lot of options.

Looks like the most hopeful idea is to send a glider to kick back one of the sideways gliders, and cross your fingers that it will be possible to make some huge ugly chaotic explosion that can overtake both switch engines. But the kicked-back glider's collision point will be pretty far back in the ash trail; by the time the hypothetical chaos gets going, it will have an uphill battle spreading far enough to catch the switch engines. And there are only just so many possible kickbacks before the returning glider is just hitting settled ash -- no more new options.

What are the odds that the mess created by the backward glider stream will just happen to create a perfectly aligned forward glider that stops both switch engines? I can't estimate without surveying all switch-engine pairs that produce backward gliders, and trying all possible forward gliders against all of them to see if any are stoppable.

But it's going to be painfully low odds, because the kind of novelty produced by a single backward glider stream has a fairly low half-life: at any time it has good odds of building itself an eater, or drilling through the ash and going boring that way.

googoIpIex
Posts: 260
Joined: February 28th, 2019, 4:49 pm
Location: Sqrt(-1)

### Re: Thread for basic questions

is there any way to reach the end result (with the unstable block) without creating the piece of junk?

Code: Select all

x = 13, y = 14, rule = B3/S23
2o3b2o$o2bo2bo$b5o2$b3o$bo2bo$3bobo$4bobo$5b2o3$10b3o$10bo$11bo!

woomy on a vroomy

fluffykitty
Posts: 653
Joined: June 14th, 2014, 5:03 pm
Contact:

### Re: Thread for basic questions

Does this work?

Code: Select all

x = 16, y = 17, rule = B3/S23
2o3b2o6bo$o2bo2bo6bobo$b5o7b2o2$b3o$bo2bo$3b2o8$b3o$3bo$2bo!

I like making rules

googoIpIex
Posts: 260
Joined: February 28th, 2019, 4:49 pm
Location: Sqrt(-1)

### Re: Thread for basic questions

Almost, but not quite.

I'm looking for one which results in this pattern:

Code: Select all

x = 7, y = 6, rule = B3/S23
2o3b2o$o2bo2bo$b5o2$b2o$b2o!

woomy on a vroomy

dvgrn
Moderator
Posts: 6710
Joined: May 17th, 2009, 11:00 pm
Contact:

### Re: Thread for basic questions

googoIpIex wrote:Almost, but not quite.

I'm looking for one which results in this pattern:

Code: Select all

x = 7, y = 6, rule = B3/S23
2o3b2o$o2bo2bo$b5o2$b2o$b2o!

Can you use a recipe like the one for the tub-supported variant of the siamese-bookends object?

Code: Select all

x = 21, y = 11, rule = B3/S23
7b2o9b2o$7b2o8bo2bo$17bo2bo$18b2o2$8b3o$16b3o$16bo$b2o14bo$obo$2bo! For example, there must be a way to build the following by adding the block at the top as a last step: Code: Select all x = 7, y = 7, rule = B3/S23 b2o$b2o2$2bo$bobob2o$o2bob2o$b2o!
At T=2 this produces a great-grandfather of the siamese bookends, but if the single cell at the right can be continuously suppressed by ... whatever else you're adding (?) ... then it will settle into the right shape with no problem, and you have the maximum amount of space to do something with that part of the induction coil.

Of course if you're also making changes on the siamese-bookends side, this probably won't work. Seemed like it might be worth mentioning just in case.

A for awesome
Posts: 2036
Joined: September 13th, 2014, 5:36 pm
Location: 0x-1
Contact:

### Re: Thread for basic questions

googoIpIex wrote:Almost, but not quite.

I'm looking for one which results in this pattern:

Code: Select all

x = 7, y = 6, rule = B3/S23
2o3b2o$o2bo2bo$b5o2$b2o$b2o!

Does this work?

Code: Select all

x = 9, y = 11, rule = B3/S23
2b2o3b2o$2bo2bo2bo$3b5o2$3b3o$2bo2bo$3b2o$6b2o$b2o3bobo$obo3bo$2bo! x₁=ηx V ⃰_η=c²√(Λη) K=(Λu²)/2 Pₐ=1−1/(∫^∞_t₀(p(t)ˡ⁽ᵗ⁾)dt) $$x_1=\eta x$$ $$V^*_\eta=c^2\sqrt{\Lambda\eta}$$ $$K=\frac{\Lambda u^2}2$$ $$P_a=1-\frac1{\int^\infty_{t_0}p(t)^{l(t)}dt}$$ http://conwaylife.com/wiki/A_for_all Aidan F. Pierce googoIpIex Posts: 260 Joined: February 28th, 2019, 4:49 pm Location: Sqrt(-1) ### Re: Thread for basic questions Yes, it does. Thanks! woomy on a vroomy Gustone Posts: 614 Joined: March 6th, 2019, 2:26 am ### Re: Thread for basic questions How do I MANUALLY upload a soup to catagogue? My favourite oscillator of all time Code: Select all x = 15, y = 13, rule = B3/S23 7bo2$3b2o5b2o$b2o4bo4b2o$5b2ob2o$bobo7bobo$bo2bobobobo2bo$5obobob5o$o
4bo3bo4bo$b3obobobob3o$3bob2obo2bo$8bobo$8b2o!


testitemqlstudop
Posts: 1362
Joined: July 21st, 2016, 11:45 am
Location: in catagolue
Contact:

### Re: Thread for basic questions

With the stdin symmetry (compile with --symmetry stdin)

Moosey
Posts: 3268
Joined: January 27th, 2019, 5:54 pm
Location: A house, or perhaps the OCA board. Or [click to not expand]
Contact:

### Re: Thread for basic questions

testitemqlstudop wrote:With the stdin symmetry (compile with --symmetry stdin)
And then what? How would I upload a spider or something as a soup?

(I didn’t ask the original question, but I’m curious now.)
I am a prolific creator of many rather pathetic googological functions

My CA rules can be found here

Also, the tree game
Bill Watterson once wrote: "How do soldiers killing each other solve the world's problems?"

Ian07
Posts: 551
Joined: September 22nd, 2018, 8:48 am

### Re: Thread for basic questions

Does Period-156 glider gun actually have a 16G synthesis? The synthesis cost was added by Entity Valkyrie a few days after he first joined but with no RLE provided. I tried to synthesize it myself, but couldn't stop the gliders from crossing paths (which to be fair is more likely because of my incompetence with syntheses).

Goldtiger997
Posts: 599
Joined: June 21st, 2016, 8:00 am

### Re: Thread for basic questions

Ian07 wrote:Does Period-156 glider gun actually have a 16G synthesis? The synthesis cost was added by Entity Valkyrie a few days after he first joined but with no RLE provided. I tried to synthesize it myself, but couldn't stop the gliders from crossing paths (which to be fair is more likely because of my incompetence with syntheses).
The crossing paths problem is actually fairly hard to solve, and I don't think it can be solved (in 16G or less) with only the collisions in two-glider-collisions.rle. I suspect the synthesis that Entity Valkyrie found had crossing paths in it, since close to when they first joined they posted quite a few invalid syntheses (such as this one).

Anyway, I found two distinct ways of synthesizing the gun in 16 gliders:

Code: Select all

x = 74, y = 74, rule = B3/S23
9bo$10bo$8b3o6$2bo$obo$b2o$49bo$41bo5b2o$21bo18bo7b2o$22b2o16b3o$21b2o
6$13bobo$14b2o$14bo$60bobo$60b2o$61bo5$59bo$59bobo$59b2o7$13b2o$12bobo$14bo5$12bo$12b2o$11bobo$59bo$58b2o$58bobo6$51b2o$31b3o16b2o$24b2o7bo 18bo$25b2o5bo$24bo$71b2o$71bobo$71bo6$63b3o$63bo$64bo! Code: Select all x = 148, y = 46, rule = B3/S23 3bo$bobo$2b2o62b2o58b2o5bo$66b2o58b2o3b2o$132b2o3$3bobo$3b2o$4bo56b3o
57b3o2$2bo$obo$b2o3$56bo88bobo$45bo11bo55bo31b2o$46bo8b3o55bo32bo$44b 3o4b2o60bo$52b2o$51bo$106b2o38b2o$106b2o38b2o$82bo$80b2o$81b2o4b3o50bo
$76b3o8bo19bo32bo$76bo11bo18b2o31bo$77bo28bobo3$11b2o$11bobo$11bo2$9bo 60b3o57b3o$9b2o$8bobo3$120b2o$66b2o53b2o3b2o$10b2o54b2o52bo5b2o$10bobo$10bo!

dvgrn
Moderator
Posts: 6710
Joined: May 17th, 2009, 11:00 pm
Contact:

### Re: Thread for basic questions

Goldtiger997 wrote:Anyway, I found two distinct ways of synthesizing the gun in 16 gliders...
Did you try the 3G way of making that block+pi?

Code: Select all

x = 59, y = 87, rule = LifeHistory
57.E$56.E$56.3E9$.E$2.2E5.E$.2E7.E34.E$8.3E32.2E$30.A.A11.2E$30.2A$31.A7$21.2D$21.2D4$22.2D17.A$21.D.D17.A.A$21.D19.2A$21.D.D$22.2D2$.A$
2.2A$.2A6$33.2D.2D$.2D4.3D23.D3.D3.2D$.2D3.D3.D23.3D4.2D$6.2D.2D6$41.
2A$40.2A$42.A2$20.2D$20.D.D$.2A19.D$A.A17.D.D$2.A17.2D4$21.2D$21.2D3$
2.E$2.2E$.E.E2$12.A$12.2A18.3E$11.A.A18.E$33.E4$3.E$3.2E$2.E.E3$48.2E
$48.E.E$48.E!
A quicker 4G way of making the pi+block would be needed in the other two locations. I checked the three 2G pi collisions,

Code: Select all

x = 218, y = 66, rule = LifeHistory
25.A$25.A.A$25.2A$83.A$84.A$82.3A2$157.A$155.A.A$156.2A17$2D$D.D$2.D$
D.D$2D$112.2D$112.D.D$114.D74.2D$112.D.D74.D.D$112.2D77.D$189.D.D$
189.2D20$29.2A$28.2A$30.A$78.3A$80.A$79.A$215.3A$215.A$216.A! and the two methods you used in your 16-glider synthesis (yellow gliders, first pattern), and there were always conflicts with incoming gliders one way or another. So maybe this 3G option (green gliders, first pattern) isn't as hopeful as it looks. Macbi Posts: 734 Joined: March 29th, 2009, 4:58 am ### Re: Thread for basic questions Does this help? Code: Select all x = 29, y = 37, rule = LifeHistory .C$2BC$3CB$.4B$2.4B$3.4B$4.4B$5.4B$6.4B4.2B$7.4B2.4B$8.4B.4B$9.10B$10.9B2.BC$9.11B2CB$8.13B2C$8.9BC5B$9.9BC4B$10.6B3C5B$10.15B$9.16B$9. 16B$9.16B$10.15B$10.14B$15.11B$16.11B$16.12B$17.11B$16.13B$17.8BDBDB$17.8BDBD$17.2B2D4B3D$18.B2D6B$18.8B$19.6B$20.4B$21.B!  testitemqlstudop Posts: 1362 Joined: July 21st, 2016, 11:45 am Location: in catagolue Contact: ### Re: Thread for basic questions Moosey wrote: testitemqlstudop wrote:With the stdin symmetry (compile with --symmetry stdin) And then what? How would I upload a spider or something as a soup? (I didn’t ask the original question, but I’m curious now.) When running apgluxe, paste in the RLE for spider. Macbi Posts: 734 Joined: March 29th, 2009, 4:58 am ### Re: Thread for basic questions Yeah it does help: Code: Select all x = 64, y = 80, rule = B3/S23 61bobo$61b2o$62bo14$34bo11bobo$34bobo9b2o$34b2o11bo3$49b3o$49bo$50bo7$
46bo$44b2o$45b2o4$18bo$19bo$17b3o3$44b3o$44bo$45bo4$17b2o$18b2o$17bo7$
13bo$14bo$12b3o3$16bo11b2o$16b2o9bobo$15bobo11bo14$bo$b2o$obo!

Is there a way to submit to the catalogue in such a way that the two cleanup gliders are separate?

wildmyron
Posts: 1397
Joined: August 9th, 2013, 12:45 am

### Re: Thread for basic questions

Macbi wrote:Is there a way to submit to the catalogue in such a way that the two cleanup gliders are separate?
I don't think the synthesis submission on Catagolue supports linear growth, but I may be mistaken
The latest version of the 5S Project contains over 226,000 spaceships. There is also a GitHub mirror of the collection. Tabulated pages up to period 160 (out of date) are available on the LifeWiki.

Moosey
Posts: 3268
Joined: January 27th, 2019, 5:54 pm
Location: A house, or perhaps the OCA board. Or [click to not expand]
Contact:

### Re: Thread for basic questions

After this discovery by nick gotts, I’ve started to wonder:
Why don’t we have bunnies of higher order? Why isn’t there a bunnies 12? Why does it only go to bunnies 11?

What is the longest lived 12-cell pattern known?

Obviously it lives at least as long as bunnies 11, since you can add a dot to it so that it evolves exactly the same— so that throws out a lot of possibilities.
I am a prolific creator of many rather pathetic googological functions

My CA rules can be found here

Also, the tree game
Bill Watterson once wrote: "How do soldiers killing each other solve the world's problems?"

Posts: 1964
Joined: November 8th, 2014, 8:48 pm
Location: Getting a snacker from R-Bee's

### Re: Thread for basic questions

Does 3D.lua support custom rules, and if so, under what limitations and how would one go about making them?
LifeWiki: Like Wikipedia but with more spaceships. [citation needed]

Andrew
Moderator
Posts: 771
Joined: June 2nd, 2009, 2:08 am
Location: Melbourne, Australia
Contact:

### Re: Thread for basic questions

BlinkerSpawn wrote:Does 3D.lua support custom rules ... ?
No, but you could always write a variant of 3D.lua that supported whatever rules you wanted. Best to start with the 3D.lua script included in Golly 3.2 rather than the highly optimized version in 3.3.

Ethanagor
Posts: 80
Joined: March 15th, 2017, 7:34 pm
Location: the Milky Way galaxy

### Re: Thread for basic questions

This is probably a really stupid question, but I'm pretty new to 1D automata. can someone explain to me why W90 doesn't work?
"It's not easy having a good time. Even smiling makes my face ache." - Frank N. Furter