Thread for basic questions

[confocaloid]

Link to a different view on the issue: viewtopic.php?p=172264#p172264

The quotes I posted by themselves count as expressions of choice of multiple people, who actually wanted to discuss some CA-related stuff, and who picked the word 'condition' as a natural/intuitive choice in the context.

In the following posts, I posted a number of quotes from CA-related discussions / texts:
viewtopic.php?p=171595#p171595
viewtopic.php?p=171596#p171596
viewtopic.php?p=171671#p171671
viewtopic.php?p=172091#p172091

question:
would the definitions that you gave make the whole "condition" vs. "transision" mess on the OCA:Tlife page just an equivalent of arguing over whether a Granny Smith is an apple or a fruit?

I think no.
Here is a link to the result of all changes so far in the article: https://conwaylife.com/w/index.php?diff ... did=140860
The lead section of OCA:tlife compares the rule to Life ('differing by only two ???')
The section "Rules" says: 'The two differing ???, illustrated:'
The section "B36/S2-i34q" says: 'This rule is commonly known as "tHighLife", due to it sharing a B6 ??? with {{rl|HighLife}}.'
All three cases above are similar (two different rules are viewed as sets of '???', and something is said about how the two sets are related or distinct). The question is what word/phrase to use in place of '???'

For me at least, it is easier to understand a rule as a set of condition-transition pairs.
So it makes sense to say that tHighLife and HighLife share a B6 condition.
Another way to say the same would be that the two rules have a B6 transition rule in common.
Another way to say the same would be that in both rules, a cell is born when it has six alive neighbours.
(The latter possibility might be preferable, since the (condition/transition rule) is written down in plain text, without using either of the words under debate.)

I believe that writing "tHighLife and HighLife share a B6 transition" is incorrect and confusing.
Transitions happen when some pattern is actually evolved (evolved according to some evolution rule, e.g. some rule defined by a rulestring or a ruletable). At every tick, each cell in the pattern transitions from its previous cellstate to some cellstate (different or the same).
A rule definition (e.g. rulestring or ruletable) does not consist of transitions. Instead, a rule definition prescribes that certain transitions happen in certain circumstances.

Does this help?

[C28]

Link to a different view on the issue: viewtopic.php?p=172264#p172264

The quotes I posted by themselves count as expressions of choice of multiple people, who actually wanted to discuss some CA-related stuff, and who picked the word 'condition' as a natural/intuitive choice in the context.

question:
would the definitions that you gave make the whole "condition" vs. "transision" mess on the OCA:Tlife page just an equivalent of arguing over whether a Granny Smith is an apple or a fruit?

I think no.
Here is a link to the result of all changes so far in the article: https://conwaylife.com/w/index.php?diff ... did=140860
The lead section of OCA:tlife compares the rule to Life ('differing by only two ???')
The section "Rules" says: 'The two differing ???, illustrated:'
The section "B36/S2-i34q" says: 'This rule is commonly known as "tHighLife", due to it sharing a B6 ??? with {{rl|HighLife}}.'
All three cases above are similar (two different rules are viewed as sets of '???', and something is said about how the two sets are related or distinct). The question is what word/phrase to use in place of '???'

For me at least, it is easier to understand a rule as a set of condition-transition pairs.
So it makes sense to say that tHighLife and HighLife share a B6 condition.
Another way to say the same would be that the two rules have a B6 transition rule in common.
Another way to say the same would be that in both rules, a cell is born when it has six alive neighbours.
(The latter possibility might be preferable, since the (condition/transition rule) is written down in plain text, without using either of the words under debate.)

I believe that writing "tHighLife and HighLife share a B6 transition" is incorrect and confusing.
Transitions happen when some pattern is actually evolved (evolved according to some evolution rule, e.g. some rule defined by a rulestring or a ruletable). At every tick, each cell in the pattern transitions from its previous cellstate to some cellstate (different or the same).
A rule definition (e.g. rulestring or ruletable) does not consist of transitions. Instead, a rule definition prescribes that certain transitions happen in certain circumstances.

Does this help?

i give up. i am more confused about the situation and i regret ever asking about it. please forget that this ever happened as i feel extremely stupid over my inability to understand.

[confocaloid]

That was not my intention in any case.
In short, a condition is what has to be true ("B6" = "a cell is born when it has 6 alive neighbours") and a transition is what happens ("a cell is born").
Both are used in informal discussions in a similar way, but the wiki is written for newcomers/readers, rather than for experienced rulegolfers who already know the stuff.

[TYCF]

What is the smallest glider splitter that produces two gliders in the same lane?

[dvgrn]

What is the smallest glider splitter that produces two gliders in the same lane?

Assuming you're talking about starting from a glider -- these things are called "G-to-G0" converters. Unfortunately the tandem-glider outputs were too awkward to keep in newer versions of the Elementary Conduits Collection, but older versions have quite a few H-to-G0s collected.

There still aren't any really nice compact direct G-to-G0s, I don't think. Everything is a composite G-to-H-to-another-H-to-G0, like the ones in the doubled-signal toolkit:

Code: Select all

x = 80, y = 70, rule = LifeHistory
$17.A$17.3A$10.A9.A$8.3A8.2A$2.A4.A11.B$.A.A3.2A7.4B3.3B$.A.A2.3B4.15B
$2A.2A2B5.17B$A2.B.AB2.20B$2.2A.A.22B$3.A.2A.20B$3.A2.24B$2.2A3.19B2.
2A$6.A2.17B2.A48.ABA$5.A.A2.11B.2B5.3A44.B2AB$6.A5.8B3.B7.A43.3BA$9.11B
3.2A49.4B$8.12B4.A48.4B$8.12B4.A.A45.4B$8.11B6.2A44.4B$8.B3D4B.4B49.4B
$8.2BD4B4.2A48.4B$8.2B3D2B4.A48.4B$8.6B6.3A44.4B$7.7B8.A43.4B$6.8B51.
4B$5.8B51.4B$4.9B50.4B$3.4B.6B48.4B$2.4B.7B47.4B$.4B2.6B47.4B$4B3.6B46.
4B$2.B4.6B45.4B$6.8B43.4B$7.8B41.4B$6.9B40.4B$6.9B39.4B$6.10B37.4B$6.
5B2A3B36.4B$6.5B2A4B3.2A29.4B$6.11B3.A29.4B$6.4BD7BA.A28.4B$8.B3D4B2.
2A4.2A22.4B$2.2A4.D2B2D2B9.A22.4B$3.A4.6B7.BA.A21.4B$3.A.AB.6B3.3B.B2A
10.A10.4B$4.2AB.14B12.3A7.4B$6.16B15.A5.4B$7.14B15.2A4.4B$6.16B14.9B6.
2A$6.18B14.6B7.A$7.20B2.2B2.B3.6B5.2A.A$8.15BD19B4.A2.A$9.7B.4BDBD20B
3.B2A$9.6B2.4B3D9B2A14B$8.6B3.4BD11B2A13B$8.4B6.29B$8.2B2AB12.2B.B.17B
$6.2AB.2A12.3B4.15B$5.A.AB11.2A.2BA5.15B$5.A14.A.2BA.A5.13B$4.2A15.A2.
BA.A2.13B$22.A3.A.A.2A4.8B$20.A.4A2.AB2A6.6B$19.A.A3.A.A2.B8.5B$19.A.
A2.A2.2A.2A9.B.B$20.A3.2A2.A.A9.3B$28.A.A9.B2AB$29.A11.2A!

I'm not guaranteeing that that one is the smallest -- there's probably one with a slightly better bounding box now -- but it's in the same ballpark as everything else that's known at the moment, I think.

Given some of the searches that people are working on these days, I'm thinking it's quite likely that the next big POTY-type celebrated discovery might end up being a small elementary G-to-2G.

(Probably the G-to-2G won't be a G-to-G0, and that wouldn't be as useful as two output gliders in different directions, anyway ... but if an elementary G-to-G0 shows up, especially if it's Spartan-ish, that will still be very useful in making circuitry like the doubled-signal toolkit a lot more compact.)

[HerscheltheHerschel]

Is there a three-block shifter loop?

[Quantum-mechanics]

Are these puffers know?

Code: Select all

x = 21, y = 110, rule = B3/S23
15b4o$14b6o$5bo7bob3ob2o$5b2o6b2o3b2o$5b2o$6b3o6b2o$7b2o9bo$19bo$7bo7b
o3bo$5b3o3bo4b4o$5b2o5b2o$4bo6b3o$5b2o5b2o$5b3o3bo4b4o$7bo7bo3bo$19bo
$7b2o9bo$6b3o6b2o$5b2o$5b2o6b2o3b2o$5bo7bob3ob2o$14b6o$15b4o10$14b4o$
13b6o$4bo7bob3ob2o$4b2o6b2o3b2o$4b2o$5b3o6b2o$6b2o9bo$18bo$6bo7bo3bo$
4b3o3bo4b4o$4b2o5b2o$3bo6b3o$4b2o5b2o$4b3o3bo4b4o$6bo7bo3bo$18bo$14bo
2bo28$13b4o$12b6o$3bo7bob3ob2o$3b2o6b2o3b2o$3b2o$4b3o6b2o$5b2o9bo$17b
o$5bo7bo3bo$3b3o3bo4b4o$3b2o5b2o$2bo6b3o$3b2o5b2o$3b3o3bo4b4o$5bo7bo3b
o$17bo$5b2o9bo$4b3o6b2o$3b2o$3b2o11b2o$2b2o6bob4ob2o$2bo7bo2b5o$10bo3b
3o$4bo$3bob2o$3bobo8b4o$4bo8bo3bo$4b2o11bo$b3o5b2o2bo2bo$3o6b3o$b3o5b
2o2bo2bo$4b2o11bo$3bo9bo3bo$14b4o!

[dvgrn]

Is there a three-block shifter loop?

Yes, there are the adjustable oscillators from way back, and then this modern construction by David Bell seems like it must count as a loop. The link is right there in the three-block shifter article.

[HerscheltheHerschel]

Is there a three-block shifter loop?

Yes, there are the adjustable oscillators from way back, and then this modern construction by David Bell seems like it must count as a loop. The link is right there in the three-block shifter article.

I was referring to a loop involving only three-block shifters.

[dvgrn]

Are these puffers [known]?

(Moved question to Basic Questions, since these definitely aren't spaceships.)

These puffers are probably not in the "knowable" category. Someone may possibly have seen them before, but wouldn't have kept track of them -- there are just too many medium-to-large puffers that make big messes like this, so people only pay attention to the smallest possible examples of each period and/or output.

For example, there's a smaller period-48 blinker puffer than the one in your pattern, in [url=https://github.com/Matthias-Merzenich/j ... 2-0048.rle]jslife-moving[/size]. It's always worth looking there for sample puffers that people have previously thought were worth collecting.

Very likely the other two "untamed" puffers could be tamed, with some effort, to make various "tamed" high-period puffers, but they're likely to be fairly large and awkward so they probably wouldn't end up in jslife-moving -- unless I'm failing to see some highly unusual feature of them that would make them noteworthy, which is always possible.

[dvgrn]

I was referring to a loop involving only three-block shifters.

Um. Given that they're shifters and not reflectors, how do you plan to make them into a loop?

[confocaloid]

Is there a three-block shifter loop?

I was referring to a loop involving only three-block shifters.

I don't know if this counts as a legitimate example of a loop, but you can always move into a bounded universe. Klein bottle:

Code: Select all

x = 8, y = 12, rule = B3/S23:K64*,50
o$b2o$2o$6b2o$6b2o6$2o4b2o$2o4b2o!

Torus:

Code: Select all

x = 114, y = 100, rule = B3/S23:T114,100
42$51bobo$52b2o$52bo2$59b2o$59b2o6$53b2o4b2o$53b2o4b2o29$5bo$5bobo$5b2o!

If the "bottom" glider in second example is advanced by 1 tick then it is reflected by a spark.

Code: Select all

x = 114, y = 100, rule = B3/S23:T114,100
42$51bobo$52b2o$52bo2$59b2o$59b2o6$53b2o4b2o$53b2o4b2o29$6bo$4b2o$5b2o!

[EvinZL]

Why is the silverfish so much smaller than the centipede?

[dvgrn]

Why is the silverfish so much smaller than the centipede?

Long story short, in 2014 chris_c had found a way to support a block-trail-creating "silverfish" front end of a 31c/240 spaceship, using only two streams of upward-traveling spaceships instead of the six in the centipede. The number of upward spaceship streams really accounts for almost all of the size difference between centipede and silverfish; it's expensive to use a slow salvo to build edge-shot upward HWSSes at the end of an existing stream, without some of the intermediate targets getting in the way of the stream.

Short story longer: none of the old rakes from the centipede worked in the silverfish design, because the block trail spacing was a little different. In 2014, the idea of starting all over and rebuilding from the ground up had been too exhausting to think about, so we just stopped with the centipede.

in 2018 I was originally going to just write up how the centipede worked, for the relevant chapter of the Life textbook -- but when I went back and read through the thread, there was this beautiful new silverfish front-end design just waiting to be used.

By that time it was clear that the big-pattern stuff in the second half of the the book tended to get explained a lot more clearly, if I had just recently done the relevant work and hadn't forgotten all the details yet -- so rebuilding all the rakes and doing a fresh job of assembling them seemed to be the way to go.

[HerscheltheHerschel]

Has quadratic growth ever appeared naturally or seminaturally in B3/S23?

[dvgrn]

Has quadratic growth ever appeared naturally or seminaturally in B3/S23?

Not yet! Read the LifeWiki quadratic growth article. It's a safe bet that any new quadratic-growth soups that show up on Catagolue will immediately get mentioned there. Right now the status is this (emphasis mine):

... quadratic growth could conceivably occur naturally, since many different pairs of two interacting switch engines have occurred from random soups.

[get_Snacked]

where is the range-1 von Neumann INT transition chart?

[confocaloid]

where is the range-1 von Neumann INT transition chart?

Are you looking for a mapping like this (from range-1 von Neumann conditions to sets of Hensel conditions), or the question is about something else?
Edit: added third column for the case with two coexisting universes rotated 45 degrees, where only diagonal neighbours are considered.

Code: Select all

R1 von Neumann condition          Hensel conditions   Rotated (corners only)
----------------------------------------------------------------------------
0 alive neighbours                01c2cn3c4c          01e2ei3e4e
1 alive neighbour                 1e2ka3inyq4ny5e     1c2ka3kajr4jr5c
2 opposite alive neighbours       2i3r4itz5r6i        2n3q4qwz5q6n
2 non-opposite alive neighbours   2e3kaj4kaqw5kaj6e   2c3iny4kait5iny6c
3 alive neighbours                3e4jr5inyq6ka7e     3c4ny5kajr6ka7c
4 alive neighbours                4e5c6cn7c8          4c5e6ei7e8

[Disaster16439]

How do I run zfind? It keeps giving me errors. The error is either access to file denied, or no file or directory. An example of how you run zfind would help, too.
My 138th post!
138=2x3x23=3xperiod of the twin bees shuttle=6xthe period of the 2-wing C2 OPlife oscillator found by may13

[confocaloid]

Forum thread: zfind discussion

Download the ntzfind files:

Code: Select all

# git clone https://github.com/rokicki/ntzfind.git
# cd ntzfind
# ls
ntzfind.cpp  README.md  tab.cpp

Compile (I used the command from README.md) and run with 'c' to print usage instructions:

Code: Select all

# g++ -std=c++11 -O3 -march=native -o ntzfind ntzfind.cpp
# ./ntzfind c

Code: Select all

Usage: "zfind options"
  e.g. "zfind B3/S23 p3 k1 w6 v" searches Life (rule B3/S23) for
  c/3 orthogonal spaceships with even bilateral symmetry and a
  search width of 6 (full width 12).

Available options:
  bNN/sNN searches for spaceships in the specified rule (default: b3/s23)

  pNN  searches for spaceships with period NN
  kNN  searches for spaceships that travel NN cells every period
  wNN  searches for spaceships with search width NN
       (full width depends on symmetry type)

  lNN  terminates the search if it reaches a depth of NN (default: 2000)
  mNN  disallows spaceships longer than a depth of NN
       (the spaceship length is approximately depth/period)
  fNN  disallows spaceships that do not have the full period by a depth of NN
  tNN  disallows full-period rows of width greater than NN
  sNN  terminates the search if NN spaceships are found (default: 1)

  xNN  searches for spaceships that travel NN cells horizontally every period
       (only values of 0 and 1 are currently supported)
       when using x1, one phase of the spaceship will have a width of 1 less
       than the width specified by the 'w' parameter
  NNN  when using x1, NN is the phase with the smaller width (default: 0)

  dNN  dumps the search state every 2^NN calculations (minimum: 20)
  j    dumps the state at start of search

  a    searches for asymmetric spaceships
  u    searches for odd bilaterally symmetric spaceships
  v    searches for even bilaterally symmetric spaceships
  g    searches for symmetric spaceships with gutters (empty center column)

  o    uses naive search order (search will take longer when no ships exist)
  r    uses randomized search order
  n    uses popcount search order

and performs the command. Available commands: 
  RNNN restricts memory usage to NNN megabytes
  CNNN uses about NNN megabytes for lookahead cache

Run the example from the usage instructions, only with rule changed to B3/S234c:

Code: Select all

# ./ntzfind B3/S234c p3 k1 w6 v

Code: Select all

Rule: B3/S234c
Period: 3
Offset: 1
Width:  6
Symmetry: even
Depth limit: 2000
Stop search if a ship is found.
Starting search

.....oo.....
.oo.o..o.oo.
.oo......oo.
.o.o....o.o.
...oo..oo...
..oo.oo.oo..
....o..o....
..o......o..
..o......o..
............
..oooooooo..
.oo......oo.
............
Length: 39
Spaceship found. (1)

Current depth: 2001
Calculations: 3034
CPU time: 0.005920 seconds
Search terminated: spaceship found.

Find the c/11o from 11life:

Code: Select all

# ./ntzfind B3/S2-n36c p11 k1 w3 u

Code: Select all

Rule: B3/S2-n36c
Period: 11
Offset: 1
Width:  3
Symmetry: odd
Depth limit: 2000
Stop search if a ship is found.
Starting search

..o..
.o.o.
.....
oo.oo
.....
.....
Length: 64
Spaceship found. (1)

Current depth: 2001
Calculations: 54352
CPU time: 0.009734 seconds
Search terminated: spaceship found.

[get_Snacked]

Are you looking for a mapping like this (from range-1 von Neumann conditions to sets of Hensel conditions), or the question is about something else?
Edit: added third column for the case with two coexisting universes rotated 45 degrees, where only diagonal neighbours are considered.

I'm just looking for whatever letters are used for conditions in R1 VN rulestrings, analogously to R1 Moore.

[Disaster16439]

Forum thread: zfind discussion

Download the ntzfind files:

Code: Select all

# git clone https://github.com/rokicki/ntzfind.git
# cd ntzfind
# ls
ntzfind.cpp  README.md  tab.cpp

Compile (I used the command from README.md) and run with 'c' to print usage instructions:

Code: Select all

# g++ -std=c++11 -O3 -march=native -o ntzfind ntzfind.cpp
# ./ntzfind c

Code: Select all

Usage: "zfind options"
  e.g. "zfind B3/S23 p3 k1 w6 v" searches Life (rule B3/S23) for
  c/3 orthogonal spaceships with even bilateral symmetry and a
  search width of 6 (full width 12).

Available options:
  bNN/sNN searches for spaceships in the specified rule (default: b3/s23)

  pNN  searches for spaceships with period NN
  kNN  searches for spaceships that travel NN cells every period
  wNN  searches for spaceships with search width NN
       (full width depends on symmetry type)

  lNN  terminates the search if it reaches a depth of NN (default: 2000)
  mNN  disallows spaceships longer than a depth of NN
       (the spaceship length is approximately depth/period)
  fNN  disallows spaceships that do not have the full period by a depth of NN
  tNN  disallows full-period rows of width greater than NN
  sNN  terminates the search if NN spaceships are found (default: 1)

  xNN  searches for spaceships that travel NN cells horizontally every period
       (only values of 0 and 1 are currently supported)
       when using x1, one phase of the spaceship will have a width of 1 less
       than the width specified by the 'w' parameter
  NNN  when using x1, NN is the phase with the smaller width (default: 0)

  dNN  dumps the search state every 2^NN calculations (minimum: 20)
  j    dumps the state at start of search

  a    searches for asymmetric spaceships
  u    searches for odd bilaterally symmetric spaceships
  v    searches for even bilaterally symmetric spaceships
  g    searches for symmetric spaceships with gutters (empty center column)

  o    uses naive search order (search will take longer when no ships exist)
  r    uses randomized search order
  n    uses popcount search order

and performs the command. Available commands: 
  RNNN restricts memory usage to NNN megabytes
  CNNN uses about NNN megabytes for lookahead cache

Run the example from the usage instructions, only with rule changed to B3/S234c:

Code: Select all

# ./ntzfind B3/S234c p3 k1 w6 v

Code: Select all

Rule: B3/S234c
Period: 3
Offset: 1
Width:  6
Symmetry: even
Depth limit: 2000
Stop search if a ship is found.
Starting search

.....oo.....
.oo.o..o.oo.
.oo......oo.
.o.o....o.o.
...oo..oo...
..oo.oo.oo..
....o..o....
..o......o..
..o......o..
............
..oooooooo..
.oo......oo.
............
Length: 39
Spaceship found. (1)

Current depth: 2001
Calculations: 3034
CPU time: 0.005920 seconds
Search terminated: spaceship found.

Find the c/11o from 11life:

Code: Select all

# ./ntzfind B3/S2-n36c p11 k1 w3 u

Code: Select all

Rule: B3/S2-n36c
Period: 11
Offset: 1
Width:  3
Symmetry: odd
Depth limit: 2000
Stop search if a ship is found.
Starting search

..o..
.o.o.
.....
oo.oo
.....
.....
Length: 64
Spaceship found. (1)

Current depth: 2001
Calculations: 54352
CPU time: 0.009734 seconds
Search terminated: spaceship found.

The thing is, I don't understand what the . is.

My 140th post! 140=2^2x5x7

[confocaloid]

Forum thread: zfind discussion
Download the ntzfind files: [...]
Compile (I used the command from README.md) and run with 'c' to print usage instructions: [...]

The thing is, I don't understand what the . is.

Maybe Tutorials/Compiling will help?

Are you looking for a mapping like this (from range-1 von Neumann conditions to sets of Hensel conditions), or the question is about something else?
Edit: added third column for the case with two coexisting universes rotated 45 degrees, where only diagonal neighbours are considered.

I'm just looking for whatever letters are used for conditions in R1 VN rulestrings, analogously to R1 Moore.

For 2-state isotropic CA using the range-1 von Neumann neighbourhood, the configuration needs to be specified only when there are two alive neighbours (either two opposite neighbours, or not). I don't remember any established notation to distinguish between these two that would be analogous to Hensel notation.

(There were discussions/proposals for range-2 von Neumann neighbourhood:)

We just had a conversation on Discord that might make this a lot easier to use. [...]

[...]Alright this thing is complicated.

[C28]

what is the protocal for if a lifewiki user and/or conwaylife.com forum user dies?

[Disaster16439]

Forum thread: zfind discussion
Download the ntzfind files: [...]
Compile (I used the command from README.md) and run with 'c' to print usage instructions: [...]

The thing is, I don't understand what the . is.

Maybe Tutorials/Compiling will help?

No, it still doesn't explain what the . in ./zfind is
My 142nd post!
142=2x71