Thread for basic questions

[Sokwe]

Can you tell me more about the differences in the 1-letter finds? I am not sure whether to use qfind or afind.

It depends on what you want to find, although if you're considering afind, you should probably try to get ikpx2 or gfind working instead. Together they can do almost anything afind can, but quite a bit faster (and multi-threaded in the case of ikpx2). Also, as I recall, afind is a bit buggy and may miss things.

qfind is much faster than afind, but it can only search front-to-back for orthogonal ships and is essentially limited to a logical width of 10 (true width depends on symmetry). It is the most effective tool when there are many partial results with some rows in common or when you're looking for a large number of ships.

afind can search for ships with "floating rows" (that is, new rows can be placed with some shift rather than all in a single column) and can search for ships that travel diagonally or even obliquely, but ikpx2 also does these things and does them better. The two "advantages" to afind over ikpx2 are that it can (I think) search for ships with gcd(period,offset)>1 (but so can gfind in all cases where this would be useful), and it can search for short, wide ships (but you should probably use WLS, JLS, or LLSSS for such searches).

Basically, afind was a great program, and I used it to find several new ships in Life-like CA back in the day, but I don't think it's really useful anymore. Everything it does can be done better by some other program.

Edit: afind may still be the best at finding a large number of short, wide ships in cases where there are many ships with columns in common. I haven't used it in years, so I can't say for sure.

[Haycat2009]

Just for clarification: ixpx2 can only complete partials, not find new ones, right?

What about zfind and gfind?

[confocaloid]

Just for clarification: ixpx2 can only complete partials, not find new ones, right?

What about zfind and gfind?

All of these can be used to find complete new spaceships. If you're interested, try looking them up / reading the forums.

[Haycat2009]

How do I find the activation stage of a billiard table/spaceship?

[confocaloid]

How do I find the activation stage of a billiard table/spaceship?

The tutorial Tutorials/Glider syntheses which you just edited includes a section "Other methods" with several links to relevant forum posts.
TL,DR: as far as I understand, it is still more an art than a science.

Regarding your edits in Tutorials/Glider syntheses, I think "As a hint, it should take 14 gliders" is not very honest as a hint. A newcomer is not likely to find the same synthesis, or a synthesis with the same cost. I think that sentence should be discarded.

[Haycat2009]

How do I find the activation stage of a billiard table/spaceship?

The tutorial Tutorials/Glider syntheses which you just edited includes a section "Other methods" with several links to relevant forum posts.
TL,DR: as far as I understand, it is still more an art than a science.

Regarding your edits in Tutorials/Glider syntheses, I think "As a hint, it should take 14 gliders" is not very honest as a hint. A newcomer is not likely to find the same synthesis, or a synthesis with the same cost. I think that sentence should be discarded.

3 for snake, 1 for boat-bit, 4 for conversion, 3 for other conversion, 3 for adding mango with long 9.

[confocaloid]

3 for snake, 1 for boat-bit, 4 for conversion, 3 for other conversion, 3 for adding mango with long 9.

And what if they used a suboptimal component? They still got a working synthesis (i.e. solved the problem), but they didn't get 14 gliders.
The "hint" is not really a hint.

[qqd]

It is possible for an INT rule to have an elementary unidimensional spaceship? If it is true, then can you engineer one example of such a rule?

[confocaloid]

It is possible for an INT rule to have an elementary unidimensional spaceship? If it is true, then can you engineer one example of such a rule?

The way I understand terms, the following spaceship is elementary and unidimensional-

Unidimensional spaceship, 2c/36o, simple rulestring

Code: Select all

x = 11, y = 1, rule = B2-a3-ijnr4/S12i
ob2ob6o!

I don't know whether the rule was/is engineered or not, though.

(added later)
For one-dimensional spaceships, see also viewtopic.php?p=137657#p137657 and viewtopic.php?p=93077#p93077

[snowman]

The battleship is a large example of an elementary 1xN spaceship:

Code: Select all

x = 21, y = 1, rule = B2ce4-cint67e8/S14567
2o3bob2obob2obob2obo!

Many other examples are possible. Here's a 1xN c/4

Code: Select all

x = 4, y = 1, rule = B2c3/S1e34i
ob2o!

[Haycat2009]

c/137 things with low population.

Well, a pi-R squared spaceship would be more likely to be 3c/10. Pi has a 3c/10 movement, neither has c/137.

Also, nice joke.

[dvgrn]

c/137 things with low population.

Well, a pi-R squared spaceship would be more likely to be 3c/10. Pi has a 3c/10 movement, neither has c/137.

Taking this as a basic question, since this thread is where you put the comment: "more likely" is very hard to quantify in this case. It seems quite unlikely that the two R-pentominoes would manage to interact with two pi-heptominoes in a way that would support the 3c/10 movement of both pi objects, while also throwing the R pentominoes such a large distance forward so quickly to just the right new location.

By contrast, there are a lot of configurations of two pi predecessors and two R predecessors that fit inside 55x34. The idea is that those four active objects quickly interact with each other and produce a near-motionless blob of chaos, which just happens to resolve into the same configuration offset by one cell after 137 ticks (or maybe by N cells after N*137 ticks, I suppose.)

That's extraordinarily unlikely also, of course. There might not be much of any way except by intuition to estimate the relatively unlikeliness of the two possibilities.

[Haycat2009]

c/137 things with low population.

Well, a pi-R squared spaceship would be more likely to be 3c/10. Pi has a 3c/10 movement, neither has c/137.

Taking this as a basic question, since this thread is where you put the comment: "more likely" is very hard to quantify in this case. It seems quite unlikely that the two R-pentominoes would manage to interact with two pi-heptominoes in a way that would support the 3c/10 movement of both pi objects, while also throwing the R pentominoes such a large distance forward so quickly to just the right new location.

By contrast, there are a lot of configurations of two pi predecessors and two R predecessors that fit inside 55x34. The idea is that those four active objects quickly interact with each other and produce a near-motionless blob of chaos, which just happens to resolve into the same configuration offset by one cell after 137 ticks (or maybe by N cells after N*137 ticks, I suppose.)

That's extraordinarily unlikely also, of course. There might not be much of any way except by intuition to estimate the relatively unlikeliness of the two possibilities.

Well, R is the most common sequence, so I do not see why it could not reform with the correct spacing after being hit with pi exhaust, especially under symmetry. The pi moves forward by itself, so we do not have to work about it.

In your example, the pi cannot move forward by itself, as it would have to move at a speed not typical to it. Assuming symmetry, there are 2 objects that need to be reformed. For context, imagine the snark. Then think about an alt universe where neither of the transparent blocks were known.

3c/10 is more likely. The partial we have is excellent by partial standards - I do not see why we cannot have a 3c/10 spaceship.

[confocaloid]

3c/10 is more likely. The partial we have is excellent by partial standards - I do not see why we cannot have a 3c/10 spaceship.

I'd say 3c/10o is less likely. This is because it is an obvious target for exploration (so many a Lifenthusiast has looked into it) and nobody managed to solve it as of yet.
It is quite possible, that nobody made it because it cannot be done. Every possible combination just happens to fail, in some way or another.

By contrast, here is an existing working R-squared spaceship in a Life-like cellular automaton differing from Conway's Life in a single Hensel condition (B8, which also happens to be a neighbour-counting condition):

[...]
Perhaps the most interesting of these is the one ship that has the two Rs in exactly the same phase (the fourth ship from the top). This means that the following simple pattern is an oblique ship predecessor in b38/s23:

Code: Select all

x = 12, y = 37, rule = B38/S23
2bo$3o$bo18$11bo$9b3o$10bo12$7b2o$6bobo$7bo!

[...]

It is not p137, but then Pedestrian Life is not Life either.

[Haycat2009]

I'd say 3c/10o is less likely. This is because it is an obvious target for exploration (so many a Lifenthusiast has looked into it) and nobody managed to solve it as of yet.
...
It is not p137, but then Pedestrian Life is not Life either.

Yes, but p30 (A composite) is lower than p137 (which is also a prime). Lower period spaceships are more likely to be found.

[Haycat2009]

Are there Eater 4 replacements for asymmetric reactions?

[TYCF]

Is this a generation of a known active region?

Code: Select all

x = 5, y = 6, rule = B3/S23
3bo$2bobo$2o$o$o$bo!

[confocaloid]

Is this a generation of a known active region?

Code: Select all

x = 5, y = 6, rule = B3/S23
3bo$2bobo$2o$o$o$bo!

I remember I saw the temporary tub before.
The pattern is gliderless (there are no escaping gliders), so it is possible to identify the final pattern via https://catagolue.hatsya.com/object
That leads to xp2_70gggy633zezy9111 "U-turner ash". It remains to check whether the pattern you posted is a successor of the U-turner, or a predecessor of U-turner, or if it just converges to the same sequence.

[TYCF]

Is this a generation of a known active region?

Code: Select all

x = 5, y = 6, rule = B3/S23
3bo$2bobo$2o$o$o$bo!

I remember I saw the temporary tub before.
The pattern is gliderless (there are no escaping gliders), so it is possible to identify the final pattern via https://catagolue.hatsya.com/object
That leads to xp2_70gggy633zezy9111 "U-turner ash". It remains to check whether the pattern you posted is a successor of the U-turner, or a predecessor of U-turner, or if it just converges to the same sequence.

I think it is a predecessor.

Generation 29.

Code: Select all

x = 9, y = 5, rule = B3/S23
6bo$2o5b2o$ob2o$b3o$2b2o!

[confocaloid]

Only if you ignore the spark. Otherwise I think it just converges to the same sequence as the U-turner object.

[eRroR_6o6]

Was wondering, is there a stable reflector that reflects a glider one way, switches to some other state of itself, and then reflects a glider on the same lane as the original glider a different way?

[confocaloid]

Was wondering, is there a stable reflector that reflects a glider one way, switches to some other state of itself, and then reflects a glider on the same lane as the original glider a different way?

This device accepts input gliders on a single lane, and sends output gliders in two opposite directions on the same lane:

Code: Select all

x = 110, y = 47, rule = B3/S23
bo$2bo$3o2$46b2o$46b2o12b2obob2o$53bob2obo2bob2o2bo$53b2obob2obo4bobo
13bo$61b5obo12b3o$57b2obo4bo13bo$57b2obobo3bo12b2o$61bobo3bo$62bo3b2o
3$89b2o$89b2o$54b2o11bo$54b2o9b3o$39b2o24bo$21bo16bo2bo23bo$22bo14bob
2o$20b3o14bo56b2o$36b2o56bo$51b2o37bo9b2o$34b4o13bo38bo8bo2bo2b2o$33bo
3bo14b3o24b2o9b3o6bobo4bo$33b2o19bo11b2o11b2o11bo7bo5bob2o$67bo35b2obo
bo$64b3o36bo2bo2bo$64bo35bo4bo2b2o$100b5o2$102b2obo$23b2o77bob2o$22bob
o5b2o$22bo7b2o$21b2o2$35bo$31b2obobo$30bobobobo$27bo2bobobobob2o$27b4o
b2o2bo2bo$31bo4b2o$29bobo$29b2o!

[...]
Here is a capped p2023 glider gun:

Code: Select all

x = 130, y = 98, rule = StateInvestigator
43.EA$42.EA.A6.E$42.A2.A$43.2A7.A3.R$53.A$51.3A7$53.2A$53.2A9$40.E5.E
$39.B7.E$48.E$44.E3.D$37.E2.2DE2.E3.E$37.E2.E5.E2.D$37.BE9.D$48.C2$
49.3A$49.A2.A$49.A$49.A46.2A$50.A.A43.A$94.A.A$94.2A$47.E$42.EB2EBC
29.2A18.2A$42.D11.EDE21.A18.A$41.C15.D20.A.A14.A.A$40.D.D15.E20.2A14.
2A$42.2EB12.D$44.CDE7.ED2E2.D$40.B.DE.ED7.B2.E2.E$41.E2.EDCD5.DE2.D.D
$41.E2.E2.E8.E.2E$41.D2.D2.E5.2EDE2.D$41.ED.E2.D5.DCE4.E$42.E4.CD7.D
3.E$44.E3.EA4.E.E2.D$41.E6.DED2.DE.E2.E$49.E3.EDED.DE$41.D3.D5.E3.D2E
2.C$C2.R37.CD.DC11.C$43.E6.D4.DE.D.D$11.A28.E2.2E5.C3EDC3.C$12.2A26.D
2.D.E4.EDE2.2E2.E$11.2A27.E.D2ED3.ED3.E.D2.D25.2A7.2A$41.E4.2E.E4.D2E
D.E26.A7.A$45.D4.ED2E4.E24.3A9.3A$43.D2E5.E2.D28.A13.A$46.D3.D3.2ED$
45.CED.DE$48.2E$48.DC$46.CD$48.E$52.A$49.A2BAB$47.2B3.A$47.A8.2A$50.B
A3.A.A$57.A5$126.2A$126.A$127.3A$129.A14$46.R$34.B!

[...]

EDIT3: opposite directions in the exact same lane

Code: Select all

x = 27, y = 20, rule = LifeHistory
6.2C$5.B2C2B$6.4B2.3B$5.11B$4.12B$3.13B$2.15B$2.9B2D4B$11B2D4B$7BE9B
2C$7BEBE5B.C2BC2.2C$7B3E4B2.CBCB3.C$.8BE4B3.C4B.C.2C$2.11B6.B2C.C.C$
3.7B9.BCB.C2.C$5.4B8.C4.C2.2C$17.5C2$19.2C.C$19.C.2C!

[...]

[Haycat2009]

Asking again: Are there eater 4 variants meant for asymmetric reactions?

[b-engine]

Asking again: Are there eater 4 variants meant for asymmetric reactions?

Very likely no. Eater 4 is mainly used for symmetric reactions, and other catalysts are available for asymmetric reaction, according to what I see in Tutorials/Catalyses. See Tutorials/Catalyses#Eater_4.

[Haycat2009]

Asking again: Are there eater 4 variants meant for asymmetric reactions?

Very likely no. Eater 4 is mainly used for symmetric reactions, and other catalysts are available for asymmetric reaction, according to what I see in Tutorials/Catalyses. See Tutorials/Catalyses#Eater_4.

So is there anything else that can regenerate 4 cells in a corner-shaped formation?