Thread for basic questions

[mniemiec]

How can I measure how explosive a rule is?

This is an extremely difficult question whose difficulty is on par with "what does pattern X do in Y generations". For example, a rule's explosivity might depend on whether a B heptomino is a short-lived pattern, a long-lived methuselah, a clean spaceship, a dirty puffer, a breeder, or a replicator - and the only way to determine which of those applies is to actually run the B heptomino to completion. Other common patterns (e.g. R pentomino, T tetromino, pi heptomino, etc.) would also have a dramatic effect on the explosivity of a rule, and the evolutions of these are often incredibly fragile, and can behave differently based on very slight perturbations of rules that otherwise don't have a significant effect on a rule's behavior.

B8 or S8 occur very rarely, and usually have very little effect - except that they affect how traffic lights and honeyfarms evolve, and since those two patterns occur very frequently in Life, B8 and S8 have a disproprtionally larger influence on rule behaviors in rules similar to Life, much more than they ought to based on random neighborhood distributions.

[Schiaparelliorbust]

How do puffer and fuse orbits form? I mean how and why do they stabilize? Even something as dirty as Puffer 2 can stabilize relatively quickly.

[mniemiec]

How do puffer and fuse orbits form? I mean how and why do they stabilize? Even something as dirty as Puffer 2 can stabilize relatively quickly.

They stabilize in exactly the same way that methuselahs do. If you consider every generation as a state in an arbitrarily large (probably finite) state machine, each state leads to another state; if any state leads back to a previously-visited state, forming a loop, you have periodicity.

Determining whether an arbitrary pattern achieves periodicity by time p (or whether it ever does) in general is a difficult problem, equivalent to actually evaluating the pattern p generations and seeing what it does, which is potentially O(p!^2).

[yujh]

From rlifesrc discussion in GitHub: how can I only upload one object to catagolue to a specific place?

[hotdogPi]

I've also come across a few rules where it's explosive if the entire pattern is symmetric along a horizontal or diagonal line (which one it is depends on the rule), but not in any other case. If I place a single cell anywhere not on the axis of symmetry, it settles quickly.

[Schiaparelliorbust]

I've also come across a few rules where it's explosive if the entire pattern is symmetric along a horizontal or diagonal line (which one it is depends on the rule), but not in any other case. If I place a single cell anywhere not on the axis of symmetry, it settles quickly.

Yes, I've always found that to be a bit strange. Why does that happen? Sometimes I see gliders moving in between the axes causing a sort of feedback loop, but in rules that lack common gliders, I don't see how it happens.

[yujh]

Why did I get notified from the quoting of someone in my foe list??

[Schiaparelliorbust]

How do you input known cells in rlifesrc? Is there any way I can convert an RLE to json (the format it wants)?

[yujh]

How do you input known cells in rlifesrc?

They said:

Known cells

Cells whose states are known before the search.

Input should be in Json format, e.g. [{"coord":[0,0,0],"state":0},{"coord":[1,1,0],"state":1}], where:

coord is the coordinates of the cell. [x, y, t] means the cell at (x, y) on the generation t.
state is the state of the cell. For non-Generations rules, 0 means dead, 1 means alive.

Also, I’ve just talked to alephalpha to see if they can make that support rle.

[PC101]

Are there turners in the Jason Summers pattern collection that involve collisions of gliders and XWSSs? I was able to find some turners in that database that turns gliders, but they don't involve the gliders actually colliding with the XWSSs.
If they aren't in the Jason Summers collection are they in any other database? If so where can I find the database?

[ihatecorderships]

Hello! I have a basic question about the precompiled version of apgsearch 5.x for windows. When I use it, it always says things like"Linear-growth pattern detected: ←[1;32myl144_1_16_afb5f3db909e60548f086e22ee3353ac←[0m" I want to see these patterns in Golly, but I don't know what all of those symbols mean, and I can't just copy and paste them. Can someone help me?

[bubblegum]

Hello! I have a basic question about the precompiled version of Catagolue 5.x for windows. When I use it, it always says things like"Linear-growth pattern detected: ←[1;32myl144_1_16_afb5f3db909e60548f086e22ee3353ac←[0m" I want to see these patterns in Golly, but I don't know what all of those symbols mean, and I can't just copy and paste them. Can someone help me?

They're technical (I don't know why there are these random characters; the pattern you're describing is yl144_1_16_afb5f3db909e60548f086e22ee3353ac). `yl` identifies the pattern as a linear growth. The 144 is the period of the population growth (the actual BLSE is p288 but it can't know that), the 1 is the first "moment" of the pattern as calculated by some function; the 16 is the population increase per population growth period; the hexadecmial mess at the end is the second and third "moments", separated by a # and hashed.

[wwei47]

the 1 is the first "moment" of the pattern as calculated by some function

I'm pretty sure that it's just the period of the population of the ash.

[bubblegum]

I'm pretty sure that it's just the period of the population of the ash.

It can't tell the difference between the ash and the non-ash.

[BokaBB]

In my Pi puffer life thread I cannot make the big LifeViewer collections openable. Also,parts of them get removed and (some) a weird line(s) appear(s). However,that last likely appeared because I had a Draw brush back then but is pretty weirdly behaving and putting bigger lines.
Solution?
Have a good day!

BokaBB

[dvgrn]

In my Pi puffer life thread I cannot make the big LifeViewer collections openable. Also,parts of them get removed and (some) a weird line(s) appear(s). However,that last likely appeared because I had a Draw brush back then but is pretty weirdly behaving and putting bigger lines.
Solution?

I went ahead and fixed this. It looks like your copy/paste operation (done via a phone, I understand) probably did an HTML-style copy of the RLE.

When you copy text off of an HTML page, any newlines get replaced by spaces -- which puts the entire RLE text on one line, and most CA viewers don't handle that too well (they need the newline between the header and the main RLE, anyway.)

[EvinZL]

Is it possible to build a "fast HBK" treating the half-bakery reaction as a climber? (probably yes)
If so, why hasn't anybody build one yet? (or thought of it yet?)

EDIT:

Hello! I have a basic question about the precompiled version of apgsearch 5.x for windows. When I use it, it always says things like"Linear-growth pattern detected: ←[1;32myl144_1_16_afb5f3db909e60548f086e22ee3353ac←[0m" I want to see these patterns in Golly, but I don't know what all of those symbols mean, and I can't just copy and paste them. Can someone help me?

Similar to what bubblegum said. You make a link like bubblegum did to the cata page. Then (at least how I do it) click oon one of the soups and run it to get the pattern.

The other strange symbols (presumably you're on windows if you're wondering about them) are ANSI escape codes - ←[1;32m makes green text and ←[0m resets formatting.

[dvgrn]

Is it possible to build a "fast HBK" treating the half-bakery reaction as a climber? (probably yes)
If so, why hasn't anybody build one yet? (or thought of it yet?)

It's definitely been thought of. I remember vaguely working through the implications back when the half-baked knightships were still in design mode. I don't think anyone ever figured out exactly what a front end for one of these half-baked caterpillars would look like, though... all I remember is that it looked complicated, and the HBK was easier to complete. Your mileage may vary, though!

[MathAndCode]

Is there some repeating evolutionary sequence and some k so that for all n>k, it is impossible to create an n-cell predecessor of that evolutionary sequence so that no proper subset of the live cells of that predecessor is a predecessor of the same evolutionary sequence?

[toroidalet]

Is there some repeating evolutionary sequence and some k so that for all n>k, it is impossible to create an n-cell predecessor of that evolutionary sequence so that no proper subset of the live cells of that predecessor is a predecessor of the same evolutionary sequence?

Take a glider synthesis that uses the pi created by this fuse:

Code: Select all

x = 154, y = 152, rule = B3/S23
152b2o$151bobo$150bo$149bo$148bo$147bo$146bo$145bo$144bo$143bo$142bo$
141bo$140bo$139bo$138bo$137bo$136bo$135bo$134bo$133bo$132bo$131bo$130b
o$129bo$128bo$127bo$126bo$125bo$124bo$123bo$122bo$121bo$120bo$119bo$
118bo$117bo$116bo$115bo$114bo$113bo$112bo$111bo$110bo$109bo$108bo$107b
o$106bo$105bo$104bo$103bo$102bo$101bo$100bo$99bo$98bo$97bo$96bo$95bo$
94bo$93bo$92bo$91bo$90bo$89bo$88bo$87bo$86bo$85bo$84bo$83bo$82bo$81bo$
80bo$79bo$78bo$77bo$76bo$75bo$74bo$73bo$72bo$71bo$70bo$69bo$68bo$67bo$
66bo$65bo$64bo$63bo$62bo$61bo$60bo$59bo$58bo$57bo$56bo$55bo$54bo$53bo$
52bo$51bo$50bo$49bo$48bo$47bo$46bo$45bo$44bo$43bo$42bo$41bo$40bo$39bo$
38bo$37bo$36bo$35bo$34bo$33bo$32bo$31bo$30bo$29bo$28bo$27bo$26bo$25bo$
24bo$23bo$22bo$21bo$20bo$19bo$18bo$17bo$16bo$15bo$14bo$13bo$12bo$11bo$
10bo$9bo$8bo$7bo$6bo$5bo$4bo$3bo$2bo$2o!

No cell from any of the gliders can be removed, and any cell of the fuse will also cause the synthesis to collapse. It is theoretically possible for an oscillator or still life or spaceship to be unsynthesizable but there are no known proven examples. (we can construct a pattern that synthesizes the target iff it can be synthesized with gliders by simulating every possible synthesis to see if it works)

I don't think anyone ever figured out exactly what a front end for one of these half-baked caterpillars would look like, though... all I remember is that it looked complicated, and the HBK was easier to complete.

An idea for the frontend would be to shoot LWSS's and reflect them like so (or a better reaction) to create a slow salvo that makes the half bakeries:

Code: Select all

x = 193, y = 140, rule = B3/S23
152bo$153b2o$152b2o3$153bo$151bobo$152b2o3$155bobo7bo$156b2o5bobo$156b
o7b2o3$163bobo$164b2o$164bo2$168bo$169bo5bobo$167b3o6b2o$176bo2$176bo$
177bo$175b3o3$179bo8bo$180b2o7bo$179b2o6b3o3$187bo$188b2o$187b2o3$192b
o$190bobo$191b2o15$166bo2bo$170bo$166bo3bo$167b4o18$127b2o$125b2ob2o$
125b4o$126b2o17$84b4o$83bo3bo$87bo$83bo2bo16$43b2o$42b4o$42b2ob2o$44b
2o17$o2bo$4bo$o3bo$b4o!

The body of the perpendicular HBK would look something like this:

Code: Select all

x = 399, y = 540, rule = B3/S23
19bobo$20b2o$20bo25$2bobo$3b2o$3bo7$19bobo$20b2o$20bo18$66bobo$67b2o$
67bo23$48bo$49b2o$48b2o12$71bo$69bobo$70b2o18$118bo$116bobo$117b2o23$
99bo$100bo$98b3o12$120bo$121b2o$120b2o22$170bobo$171b2o$171bo23$152bo$
153b2o$152b2o6$o$b2o201b2o$2o201bo2bo$204bobo$205bob2o$206bo2bo$175bo
31bobo$173bobo32bo$174b2o3$170b2o$169bo2bo$170bobo$23bo147bob2o$21bobo
148bo2bo$22b2o149bobo$174bo2$186b2o$185bo2bo$186bobo$187bob2o$188bo2bo
$189bobo33b2o$190bo33bo2bo$225bobo$226bob2o$227bo2bo$228bobo$229bo4$
191b2o$25bobo162bo2bo$26b2o163bobo$26bo165bob2o$193bo2bo$194bobo$195bo
2$207b2o$206bo2bo$42bobo162bobo$43b2o163bob2o$43bo165bo2bo$210bobo33b
2o$211bo33bo2bo$246bobo$247bob2o$248bo2bo$249bobo$250bo3$38bo$39b2o
171b2o$38b2o171bo2bo$212bobo$213bob2o$214bo2bo$215bobo$216bo2$55bo172b
2o$56b2o169bo2bo$55b2o171bobo$229bob2o$230bo2bo$231bobo33b2o$232bo33bo
2bo$267bobo$268bob2o$269bo2bo$270bobo$271bo4$233b2o$232bo2bo$233bobo$
234bob2o$235bo2bo$236bobo$237bo2$61bo187b2o$62b2o184bo2bo$61b2o186bobo
$250bob2o$251bo2bo$252bobo33b2o$253bo33bo2bo$288bobo$289bob2o$290bo2bo
$291bobo$292bo3$84bo$82bobo169b2o$83b2o168bo2bo$254bobo$255bob2o$79b2o
175bo2bo$78bo2bo175bobo$79bobo176bo$80bob2o$81bo2bo185b2o$82bobo184bo
2bo$83bo186bobo$271bob2o$95b2o175bo2bo$94bo2bo175bobo33b2o$95bobo176bo
33bo2bo$96bob2o209bobo$97bo2bo209bob2o$98bobo210bo2bo$99bo212bobo$313b
o4$275b2o$274bo2bo$275bobo$276bob2o$100b2o175bo2bo$99bo2bo175bobo$100b
obo176bo$101bob2o$102bo2bo185b2o$103bobo184bo2bo$104bo186bobo$292bob2o
$116b2o175bo2bo$115bo2bo175bobo33b2o$116bobo176bo33bo2bo$117bob2o209bo
bo$118bo2bo209bob2o$119bobo210bo2bo$120bo212bobo$334bo4$296b2o$295bo2b
o$296bobo$297bob2o$121b2o175bo2bo$120bo2bo175bobo$121bobo176bo$122bob
2o$123bo2bo185b2o$124bobo184bo2bo$125bo186bobo$313bob2o$137b2o175bo2bo
$136bo2bo175bobo33b2o$137bobo176bo33bo2bo$138bob2o209bobo$139bo2bo209b
ob2o$140bobo210bo2bo$141bo212bobo$355bo4$317b2o$316bo2bo$317bobo$318bo
b2o$142b2o175bo2bo$141bo2bo175bobo$142bobo176bo$143bob2o$144bo2bo185b
2o$145bobo184bo2bo$146bo186bobo$334bob2o$158b2o175bo2bo$157bo2bo175bob
o33b2o$158bobo176bo33bo2bo$159bob2o209bobo$160bo2bo209bob2o$161bobo
210bo2bo$162bo212bobo$376bo4$338b2o$337bo2bo$338bobo$339bob2o$163b2o
175bo2bo$162bo2bo175bobo$163bobo176bo$164bob2o$165bo2bo185b2o$166bobo
184bo2bo$167bo186bobo$355bob2o$179b2o175bo2bo$178bo2bo175bobo33b2o$
179bobo176bo33bo2bo$180bob2o209bobo$181bo2bo209bob2o$182bobo210bo2bo$
183bo212bobo$397bo4$359b2o$358bo2bo$359bobo$360bob2o$184b2o175bo2bo$
183bo2bo175bobo$184bobo176bo$185bob2o$186bo2bo185b2o$187bobo184bo2bo$
188bo186bobo$376bob2o$200b2o175bo2bo$199bo2bo175bobo$200bobo176bo$201b
ob2o$202bo2bo$203bobo$204bo9$205b2o$204bo2bo$205bobo$206bob2o$207bo2bo
$208bobo$209bo2$221b2o$220bo2bo$221bobo$222bob2o$223bo2bo$224bobo$225b
o9$226b2o$225bo2bo$226bobo$227bob2o$228bo2bo$229bobo$230bo2$242b2o$
241bo2bo$242bobo$243bob2o$244bo2bo$245bobo$246bo9$247b2o$246bo2bo$247b
obo$248bob2o$249bo2bo$250bobo$251bo2$263b2o$262bo2bo$263bobo$264bob2o$
265bo2bo$266bobo$267bo9$268b2o$267bo2bo$268bobo$269bob2o$270bo2bo$271b
obo$272bo2$284b2o$283bo2bo$284bobo$285bob2o$286bo2bo$287bobo$288bo!

On the 2-HB side the glider stream will move approach the future half-bakeries, but it's not great for slow salvos because it only has 2 half-diagonals it can work on without hitting the previous target. (once it's formed we might be able to bootstrap something and fire gliders)

[cgoler2]

What symmetry was the pufferfish found in?

[dvgrn]

What symmetry was the pufferfish found in?

D8_1, and D4_+1 (the post directly below the linked one). You can find the discovery post and soup by going to the first link in the pufferfish LifeWiki article.

[MathAndCode]

Take a glider synthesis that uses the pi created by this fuse:

That should take care most cases, but there are some problematic cases that remain. The ones that I can think of are the block, beehive, traffic light, pond, pi ash, and the constellation produced from this cluster:

Code: Select all

x = 13, y = 12, rule = B3/S23
11b2o$10bo$9bo$8bo$7bo$6bo$5bo$4bo$3bo$2bo$bo$o!

At least some of these can be taken care of by using XWSS predecessors instead of gliders, using different fuses, and/or using different wick, but the block seems especially tricky.



Edit: I just found a way to create arbitrarily large block predecessors where every cell is necessary using a glider predecessor from mirage and an unsupported diagonal line.

Code: Select all

x = 51, y = 51, rule = B3/S23
o$bo$2bo$3bo$4bo$5bo$6bo$7bo$8bo$9bo$10bo$11bo$12bo$13bo$14bo$15bo$16b
o$17bo$18bo$19bo$20bo$21bo$22bo$23bo$24bo$25bo$26bo$27bo$28bo$29bo$30b
o$15b2o14bo$18bo13bo$16bo16bo$17bo16bo$15bo19bo$36bo$37bo$38bo$39bo$40b
o$41bo$42bo$43bo$44bo$45bo$46bo$47bo$48bo$49bo$50bo!

Another edit: This should take care of beehives:

Code: Select all

x = 36, y = 36, rule = B3/S23
35bo$34bo$33bo$32bo$31bo$30bo$29bo$28bo$27bo$26bo$13bo11bo$13bo10bo$10b
2o11bo$9bo3bo8bo$11bo9bo$20bo$19bo$18bo$17bo$16bo$15bo$14bo$13bo$12bo
$11bo$10bo$9bo$8bo$7bo$6bo$5bo$4bo$3bo$2bo$bo$o!

Yet another edit: I have taken care of blinkers, traffic lights, loaves, and pi ash.

Code: Select all

x = 29, y = 29, rule = B3/S23
28bo$27bo$26bo$25bo$24bo$23bo$22bo$8bo12bo$9b2o9bo$8b2o9bo$18bo$17bo$
16bo$15bo$14bo$13bo$12bo$11bo$10bo$9bo$8bo$7bo$6bo$5bo$4bo$3bo$2bo$bo
$o!

Code: Select all

x = 27, y = 27, rule = B3/S23
o$bo$2bo$3bo$4bo$5bo$6bo$7bo$8bo$9bo$10bo$11bo$12bo$13bo$14bo$6b2o7bo
$9bo6bo$7bo9bo$8bo9bo$6bo12bo$20bo$21bo$22bo$23bo$24bo$25bo$26bo!

Code: Select all

x = 42, y = 42, rule = B3/S23
o$bo$2bo$3bo$4bo$5bo$6bo$7bo$8bo$9bo$10bo$11bo$12bo$13bo$14bo$15bo$16b
o$17bo$18bo$19bo$20bo$21bo$22bo$23bo$24bo$25bo$26bo$14b2o11bo$17bo10b
o$15bo13bo$16bo13bo$14bo16bo$32bo$33bo$34bo$35bo$36bo$37bo$38bo$39bo$
40bo$41bo!

Code: Select all

x = 56, y = 62, rule = B3/S23
o$bo$2bo$3bo$4bo$5bo$6bo$7bo$8bo$9bo$10bo$11bo$12bo$13bo$14bo$15bo$16b
o$17bo$18bo$19bo$20bo$21bo$22bo$23bo$24bo$25bo$26bo$27bo$28bo$29bo$30b
o$15b2o14bo$18bo13bo$16bo16bo$17bo16bo$15bo19bo$36bo$37bo$38bo$39bo$40b
o$41bo$42bo$43bo$44bo$45bo$46bo$47bo$48bo$49bo$50bo9$54bo$53b2o$53bob
o!

I don't think that there's anything else in that category that needs to be worried about.



Fourth edit: I realized that I had forgotten about gliders (although they arguably fall into a different category).

Code: Select all

x = 61, y = 61, rule = B3/S23
60bo$59bo$58bo$57bo$56bo$55bo$54bo$53bo$52bo$51bo$50bo$49bo$48bo$47bo
$46bo$45bo$44bo$43bo$42bo$41bo$40bo$39bo$38bo$37bo$36bo$13b4o18bo$12b
o4bo16bo$18bo14bo$19bo12bo$31bo$30bo$29bo$28bo$27bo$26bo$25bo$24bo$23b
o$22bo$21bo$20bo$19bo$18bo$17bo$16bo$15bo$14bo$13bo$12bo$11bo$10bo$9b
o$8bo$7bo$6bo$5bo$4bo$3bo$2bo$bo$o!

Fifth edit: It turns out that the glider predecessor that I used for the beehive solution is a superset of a beehive predecessor.

Code: Select all

x = 5, y = 5, rule = DoubleB3S23
4.A$4.C$.2C$A3.C$2.C!

Luckily, I was able to find another glider predecessor that doesn't have that problem.

Code: Select all

x = 36, y = 36, rule = B3/S23
35bo$34bo$33bo$32bo$31bo$30bo$29bo$28bo$27bo$26bo$25bo$13bo10bo$10b2o
bo9bo$9bo12bo$11bo9bo$20bo$19bo$18bo$17bo$16bo$15bo$14bo$13bo$12bo$11b
o$10bo$9bo$8bo$7bo$6bo$5bo$4bo$3bo$2bo$bo$o!

[muzik]

Are there any other known non-period-1 billiard table oscillators where the rotor is bounded by four diagonally connected line segments, opposite ends parallel and equal length, and the rotor makes direct contact with this bounding region without ever killing it?

Code: Select all

x = 48, y = 24, rule = B3/S23
14b2ob2o$14bo3bo10b2o11b2o$15b3o11b2o11b2o2$4b2o9b3o9b4o9b4o$2obo2bob
2ob2obo3bob2ob2obo4bob2ob2obo2bobob2o$bobo2bobo3bobo3bobo3bobo2bobobo
3bobo3b2obo$o2bo2bo2bobo2bo3bo2bobo2bo4bo2bobo2bo4bo2bo$2o2b2o2b2ob2o
2b3o2b2ob2o2b4o2b2ob2o2b4o2b2o2$15b3o11b2o9b2o$14bo3bo10b2o10bo$14b2ob
2o20bo$29b4o6b2o$28bobo2bo$28bo2bobo5b4o$25b2ob2o3bob2obo4bob2o$25b2ob
o4bob2obobo2bob2o$29b4o5bobo2bo$38bo2bobo$31b2o6b4o$31b2o$41b2o$41b2o!

[wwei47]

That should take care most cases, but there are some problematic cases that remain. The ones that I can think of are the block, beehive, traffic light, pond, pi ash, and the constellation produced from this cluster:

I don't get why those in particular are problematic. Here's one for LWSS, with your method:

Code: Select all

x = 45, y = 45, rule = B3/S23
o$bo$2bo$3bo$4bo$5bo$6bo$7bo$8bo$9bo$10bo$11bo$12bo$13bo$14bo$15bo$16b
o$17bo$18bo$19bo$20bo$21bo$22bo$23bo$24bo$25bo$26bo$27bo$28bo$29bo$17b
2o11bo$20bo10bo$18bo13bo$19bo13bo$17bo16bo$35bo$36bo$37bo$38bo$39bo$40b
o$41bo$42bo$43bo$44bo!

Why does mirage have so many unnecessary cells?