ConwayLife.com

Forums for Conway's Game of Life
https://conwaylife.com/forums/

Rule request thread

https://conwaylife.com/forums/viewtopic.php?f=11&t=2030

Page 11 of 44

Re: Rule request thread

Posted: June 29th, 2017, 7:03 pm
by Tezcatlipoca
Still would like to see a rule to try to mimic biochemistry. I wonder how much complexity could arise with a rule like this.

These molecules should "snap together" like those organic chemistry models used in education sets. But if they are then able to organize random configurations of the composing molecules into a shape identical to itself, it could be a lot like what we think biological life's precursor must have been. Take the following where the symbol pairs 'q' and 'b', 'p' and 'd', 'n' and 'u', '[' and ']', and '{' and '}' are each the same shape but in different orientations:

Code: Select all

. . . . . . . . . . . . . . . 
. . . . . . . . . . . . . . .  
. . . . . . . . . . . . . . .  
. . . . . . . . . . . . . . .  
. . . . . . . . . . . . . . .  
. . . . . [ . } . . . . . . .  
. . . . . b u d . . . . . . .  
. . . . . . . . . . . . . . .  
. . . . . . . . . . . . . . .  
. . . . . . . . . . . . . . .  
. . . . . . . . . . . . . . .  
. . . . . . . . . . . . . . .  
. . . . . . . . . . . . . . .
Where:
  • shape 'b' and shape 'u' always have a tendency to arrange with 'b' on the left hand side of 'u' as it is oriented, when in within the proximity of two spaces in any direction, so that if in generation 1:

Code: Select all

. . . . . . . . . . . . . . .  
. . . . . . . . . . . . . . .  
. . . . . . u . . . . . . . .  
. . . . . . . . b . . . . . .  
. . . . . . . . . . . . . . .  
. . . . . . . . . . . . . . .
then in generation 2:

Code: Select all

. . . . . . . . . . . . . . .  
. . . . . . . . . . . . . . .  
. . . . . b u . . . . . . .  
. . . . . . . . . . . . . . .  
. . . . . . . . . . . . . . .  
. . . . . . . . . . . . . . .
And similarly:
  • 'u' and 'd' always have a tendency to arrange with 'd' on the right hand side of 'u' as it is oriented when in proximity
  • 'd' and '}' always have a tendency to arrange with 'o' above '}' as it is oriented when in proximity
  • 'b' and '[' always have a tendency to arrange with 'c' above '[' as it is oriented when in proximity
  • Once any of these pairs get into their preferred orientations, they move together as a unit.
Also:
  • '[' and '}' always have a tendency to arrange such that '{' brackets '[' as it is oriented when in proximity, but this arrangement is not strongly preferred and so does not lock into place or persist if the arrangement is being pulled into a different orientation or location by another affinity
[/list]
It would tend to form up like this:

Code: Select all

. . . . . . . . . . . . . . . 
. . . . . . . . . . . . . . .  
. . . . . . . . . . . . . . .  
. . . . . . . . . . . . . . .  
. . . p n q p n q . . . . . . 
. . . { [ ] } [ ] . . . . . .  
. . . . b u d b u d . . . . . 
. . . . . . . . . . . . . . .  
. . . . . . . . . . . . . . .  
. . . . . . . . . . . . . . .  
. . . . . . . . . . . . . . .  
. . . . . . . . . . . . . . .
A single instance of this isn't very interesting, but I hope it could get a lot more complex when each shape is more complex, can rotate in four different orientations rather than two, and in the interplay between partially formed configurations being drawn in different directions by different rules. The idea is that simple rules determining how different states are oriented relative to each other might form complexity.

I know in the above example those are shapes and and not states which is what they would have to be for a rule, but "shapes" could be formed by states that have affinities for other states in orientation when in proximity. Then you just set a big fill, a random distribution of all these states and see how they assemble.

I am really interested to see if this could work.

Re: Rule request thread

Posted: June 29th, 2017, 8:09 pm
by blah
Tezcatlipoca wrote:Still would like to see a rule to try to mimic biochemistry. I wonder how much complexity could arise with a rule like this.
I don't believe discrete cellular automata are a good way to simulate this. Why not, for example, have a set of particles moving around in continuous space (like NodesWin, also see here), rather than conforming to a grid?

Have you already seen that thing I linked? It seems relevant to your interests. Anyway, I know that if I really put in the effort I could create a particle physics system with various directional attractions if I cared enough. Maybe one day I will (probably not), but what's the point? This is a big question to me; why simulate evolution? What would we learn about evolution that we don't already know?

edit: 128th post! yeah

Re: Rule request thread

Posted: June 29th, 2017, 9:35 pm
by Tezcatlipoca
blah wrote:Have you already seen that thing I linked?
No, I haven't seen this one yet. Thanks for sharing. It looks like it could be interesting, and I am now looking forward to trying it out.

BUT I have played with a lot of evolution sims--that's how I got here a few years ago in looking for new ones--and there are a few reasons why even though there are much more specific and seemingly appropriate ways to make evolution like games, I am still excited about this possibility.

I was directed to about three rulesets (by dvgrn I think--thanks dvgrn!) when I first came here asking about the automata that were most bio-life like. One of those three was Shapeloop (viewtopic.php?f=11&t=1440&p=13228&hilit ... oop#p13228) and I found that to be the most interesting evolution sim I had ever played with, even though I am not sure it was designed with that in mind. The way the shapes propagate is eerily life-like. I was able to "evolve" hundreds of wildly different shapes from a few ancestors/lineages over millions of generations. It has a really good, direct simulacra of biological genetics for one thing, that I haven't seen anywhere else. Others in the community came up with such cool stuff as I was doing it, and it was great to be able to take that right in with a copy and paste to play with it.

Golly itself is a big asset too. The way you can fill space with random states and ability to speed things up and change the scale so profoundly were great. There are many other reasons, and some suspicions I have about why even though a rule in Golly is not the most direct and efficient way to do a simulation like this, it could still produce interesting results. A year or two ago I made a post like this before there was a rule request thread. User fluffykitty messaged me about it and we started to give it a try for about two afternoons. He or she build the rule entirely and I just tried to help with how in discussions of how it should all behave.

Code: Select all

@RULE biology_0_2
0 empty
1 b
2 u
3 d
4 out
5 in
6 interaction
7 signal out
8 signal in
9 bonded b
10 bonded u
11 bonded d
12 special interaction
13 shadow
14 corner protector

4,5,7,8 unused

ins try to match

particles only interact in unambiguous situations

@TABLE
n_states:15
neighborhood:Moore
symmetries:rotate4reflect
#all
var a1={0,1,2,3,4,5,6,7,8,9,10,11,12,13,14}
var a2={0,1,2,3,4,5,6,7,8,9,10,11,12,13,14}
var a3={0,1,2,3,4,5,6,7,8,9,10,11,12,13,14}
var a4={0,1,2,3,4,5,6,7,8,9,10,11,12,13,14}
var a5={0,1,2,3,4,5,6,7,8,9,10,11,12,13,14}
var a6={0,1,2,3,4,5,6,7,8,9,10,11,12,13,14}
var a7={0,1,2,3,4,5,6,7,8,9,10,11,12,13,14}
var a8={0,1,2,3,4,5,6,7,8,9,10,11,12,13,14}
#b-states
var b1={1,9}
var b2={1,9}
var b3={1,9}
#u-states
var c1={2,10}
var c2={2,10}
var c3={2,10}
#d-states
var d1={3,11}
var d2={3,11}
var d3={3,11}
#live
var e1={1,2,3,4,5,6,7,8,9,10,11,12,13,14}
var e2={1,2,3,4,5,6,7,8,9,10,11,12,13,14}
var e3={1,2,3,4,5,6,7,8,9,10,11,12,13,14}
#particle (all except interactions)
var f1={1,2,3,4,5,7,8,9,10,11,13}
var f2={1,2,3,4,5,7,8,9,10,11,13}
var f3={1,2,3,4,5,7,8,9,10,11,13}
#dead (for creating interactions)
var g1={0,13}
#become unbonded
9,0,a1,0,a2,0,a3,0,a4,1
10,0,a1,0,a2,0,a3,0,a4,2
11,0,a1,0,a2,0,a3,0,a4,3
9,13,a1,0,a2,0,a3,0,a4,1
10,13,a1,0,a2,0,a3,0,a4,2
11,13,a1,0,a2,0,a3,0,a4,3

#bu,ud interaction gen 3 
#(conserving particles, prefers linear strings)
#overriding “become bonded”
1,f1,f2,0,a1,a2,a3,0,f3,0
2,f1,f2,0,a1,a2,a3,0,f3,0
3,f1,f2,0,a1,a2,a3,0,f3,0

#become bonded
1,f1,a1,a2,a3,a4,a5,a6,a7,9
2,f1,a1,a2,a3,a4,a5,a6,a7,10
3,f1,a1,a2,a3,a4,a5,a6,a7,11

#b-u interaction (101 2)
#gen 1
0,c1,a1,b1,a2,c2,a2,a3,a4,1
1,a1,c1,0,c2,a2,a3,a4,a5,0
0,b1,a1,c1,a2,b2,a2,a3,a4,2
2,a1,b1,0,b2,a2,a3,a4,a5,0

#d-u interaction (101 2)
#gen 1
0,c1,a1,d1,a2,c2,a2,a3,a4,3
3,a1,c1,0,c2,a2,a3,a4,a5,0
0,d1,a1,c1,a2,d2,a2,a3,a4,2
2,a1,d1,0,d2,a2,a3,a4,a5,0

#special interaction
0,f1,a1,a2,0,f2,f3,a3,a4,12

#bud double reaction
0,f1,a1,a2,f2,a3,f3,a4,a5,12

#b-u interaction (101)
#gen 1
12,b1,a1,a2,c1,0,c2,a3,a4,1
12,c1,a1,a2,b1,0,b2,a3,a4,2
#gen 2
6,a1,b1,12,0,c1,a2,a3,a4,0
6,a1,c1,12,0,b1,a2,a3,a4,0

#d-u interaction (101)
#gen 1
12,d1,a1,a2,c1,0,c2,a3,a4,3
12,c1,a1,a2,d1,0,d2,a3,a4,2
#gen 2
6,a1,d1,12,0,c1,a2,a3,a4,0
6,a1,c1,12,0,d1,a2,a3,a4,0

#b-u interaction (2,1)
#gen 1
g1,b1,a1,a2,c1,a3,a4,a5,a6,6
g1,c1,a1,a2,b1,a3,a4,a5,a6,6
#gen 2
6,c1,a1,6,b1,a1,a2,a3,a4,2
6,b1,a1,6,c1,a1,a2,a3,a4,1
b1,6,6,a1,a2,a3,a4,a5,a6,13
c1,6,6,a1,a2,a3,a4,a5,a6,13

#d-u interaction (2,1)
#gen 1
g1,d1,a1,a2,c1,a3,a4,a5,a6,6
g1,c1,a1,a2,d1,a3,a4,a5,a6,6
#gen 2
6,c1,a1,6,d1,a1,a2,a3,a4,2
6,d1,a1,6,c1,a1,a2,a3,a4,3
d1,6,6,a1,a2,a3,a4,a5,a6,13
c1,6,6,a1,a2,a3,a4,a5,a6,13

#b-u interaction (1,1)
#gen 1
g1,9,a1,2,a2,a3,a4,a5,a6,6
g1,1,a1,10,a2,a3,a4,a5,a6,6
#gen 2
6,1,6,10,a1,a2,a3,a4,a5,1
6,2,6,9,a1,a2,a3,a4,a5,2
1,6,a1,6,a2,a3,a4,a5,a6,0
2,6,a1,6,a2,a3,a4,a5,a6,0

#d-u interaction (1,1)
#gen 1
g1,11,a1,2,a2,a3,a4,a5,a6,6
g1,3,a1,10,a2,a3,a4,a5,a6,6
#gen 2
6,3,6,10,a1,a2,a3,a4,a5,3
6,2,6,11,a1,a2,a3,a4,a5,2
3,6,a1,6,a2,a3,a4,a5,a6,0
2,6,a1,6,a2,a3,a4,a5,a6,0

#bug
6,9,a1,10,a2,a3,a4,a5,a6,0

#b-u interaction (0,2)
#gen 1
0,10,a1,0,a3,1,a4,0,a6,6
0,9,a1,0,a3,2,a4,0,a6,6
#gen 2
1,6,0,a1,a2,a3,a4,a5,0,0
2,6,0,a1,a2,a3,a4,a5,0,0
6,1,0,a1,a2,a3,a4,a5,0,1
6,2,0,a1,a2,a3,a4,a5,0,2

#d-u interaction (0,2)
#gen 1
0,10,a1,0,a3,3,a4,0,a6,6
0,11,a1,0,a3,2,a4,0,a6,6
#gen 2
3,6,0,a1,a2,a3,a4,a5,0,0
2,6,0,a1,a2,a3,a4,a5,0,0
6,3,0,a1,a2,a3,a4,a5,0,3
6,2,0,a1,a2,a3,a4,a5,0,2

#b-u interaction (x,2o)
#gen 2
6,0,b1,12,9,10,a1,a2,a3,0
1,12,6,a1,a2,a3,a4,a5,a6,0
12,1,a1,a2,a3,9,10,a4,a5,9
6,0,c1,12,10,9,a1,a2,a3,0
2,12,6,a1,a2,a3,a4,a5,a6,0
12,2,a1,a2,a3,10,9,a4,a5,10

#d-u interaction (x,2o)
#gen 2
6,0,d1,12,11,10,a1,a2,a3,0
3,12,6,a1,a2,a3,a4,a5,a6,0
12,3,a1,a2,a3,11,10,a4,a5,11
6,0,c1,12,10,11,a1,a2,a3,0
2,12,6,a1,a2,a3,a4,a5,a6,0
12,2,a1,a2,a3,10,11,a4,a5,10

#b-u interaction (x,2d)
#gen 2
0,0,2,12,9,10,a1,a2,a3,6
0,0,1,12,10,9,a1,a2,a3,6
#gen 3
12,2,a1,a2,a3,9,10,0,a5,12
12,1,a1,a2,a3,10,9,0,a5,12
12,2,a1,a2,a3,9,10,a4,a5,10
12,1,a1,a2,a3,10,9,a4,a5,9
6,0,c1,12,9,10,f1,a2,a3,14
6,0,b1,12,10,9,f1,a2,a3,14
6,0,c1,12,9,10,a1,a2,a3,0
6,0,b1,12,10,9,a1,a2,a3,0

#d-u interaction (x,2d)
#gen 2
0,0,2,12,11,10,a1,a2,a3,6
0,0,3,12,10,11,a1,a2,a3,6
#gen 3
12,2,a1,a2,a3,11,10,0,a5,12
12,3,a1,a2,a3,10,11,0,a5,12
12,2,a1,a2,a3,11,10,a4,a5,10
12,3,a1,a2,a3,10,11,a4,a5,11
6,0,c1,12,11,10,f1,a2,a3,14
6,0,d1,12,10,11,f1,a2,a3,14
6,0,c1,12,11,10,a1,a2,a3,0
6,0,d1,12,10,3,a1,a2,a3,0

#line protector dies
14,f1,f2,f3,a1,a2,a3,a4,a5,14
14,f1,a1,a2,f1,f2,a3,a4,a5,14
14,a1,a2,a3,a4,a5,a6,a7,a8,0
#shadows die
13,a1,a2,a3,a4,a5,a6,a7,a8,0
#protect corners
6,f1,f2,f3,a1,a2,a3,a4,a5,14
#protect gaps
12,f1,a1,0,f1,f2,a2,a3,a4,14
#kill unwanted interactions
6,a1,a2,a3,a4,a5,a6,a7,a8,0
12,a1,a2,a3,a4,a5,a6,a7,a8,0
@COLORS
0 0 0 0
1 255 0 0
2 0 255 0
3 0 0 255
4 255 255 0
5 0 255 255
6 255 128 0
7 255 255 128
8 128 255 255
9 255 63 63
10 63 255 63
11 63 63 255
12 255 0 255
13 0 0 128
14 128 0 0

Code: Select all

x = 23, y = 10, rule = biology_0_2
2$4.B3.A7.B3.C$3.A2.J3.B4.C2.J3.B$.A6.A4.C6.C$6.J11.J$2.B11.B$8.A11.C
!
Then they left or got interesting in something else. But what they came up with in that short time while not all that far toward what we imagined is enough to make me think that it could be really as interesting as I hoped.

Re: Rule request thread

Posted: July 4th, 2017, 6:20 pm
by muzik
A totalitic or non-totalistic rule that can simulate an infinite amount of c/(odd number) ships, similarly to this rule simulating c/even ships:
AbhpzTa wrote:c/2n p2n spaceships for any integer n>1:

Code: Select all

x = 10, y = 38, rule = B2c3ae4ai56c/S2-kn3-enq4
bo2bo$ob3o$bo2bo5$bo3bo$ob4o$bo3bo5$bo4bo$ob5o$bo4bo5$bo5bo$ob6o$bo5bo
5$bo6bo$ob7o$bo6bo5$bo7bo$ob8o$bo7bo!

Code: Select all

x = 198, y = 30, rule = B2c3ae4ai56c/S2-kn3-enq4
3o3b3o3b3o3b3o3b3o3b3o3b3o3b3o3b3o3b3o3b3o3b3o3b3o3b3o3b3o3b3o3b3o3b3o
3b3o3b3o3b3o3b3o3b3o3b3o3b3o3b3o31b3o9bo$bo5bo5bo5bo5bo5bo5bo5bo5bo5bo
5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo33bo9bobo$bo5bo5bo5bo
5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo33bo
10bo$obo4bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5b
o5bo5bo5bo5bo33bo10bo$bo4bobo4bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5b
o5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo33bo10bo$7bo4bobo4bo5bo5bo5bo5bo5bo5bo
5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo33bo10bo$13bo4bobo4bo
5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo33bo10b
o$19bo4bobo4bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5b
o5bo33bo10bo$25bo4bobo4bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo
5bo5bo5bo5bo33bo9bobo$31bo4bobo4bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo
5bo5bo5bo5bo5bo5bo32b3o9bo$37bo4bobo4bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo
5bo5bo5bo5bo5bo5bo5bo$43bo4bobo4bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo
5bo5bo5bo5bo$49bo4bobo4bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo
$55bo4bobo4bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo$61bo4bobo4bo5b
o5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo$67bo4bobo4bo5bo5bo5bo5bo5bo5bo5b
o5bo5bo5bo5bo5bo$73bo4bobo4bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo$79bo4bo
bo4bo5bo5bo5bo5bo5bo5bo5bo5bo5bo5bo$85bo4bobo4bo5bo5bo5bo5bo5bo5bo5bo
5bo5bo$91bo4bobo4bo5bo5bo5bo5bo5bo5bo5bo5bo$97bo4bobo4bo5bo5bo5bo5bo5b
o5bo5bo$103bo4bobo4bo5bo5bo5bo5bo5bo5bo$109bo4bobo4bo5bo5bo5bo5bo5bo$
115bo4bobo4bo5bo5bo5bo5bo$121bo4bobo4bo5bo5bo5bo$127bo4bobo4bo5bo5bo$
133bo4bobo4bo5bo$139bo4bobo4bo$145bo4bobo$151bo!

Re: Rule request thread

Posted: July 12th, 2017, 6:00 am
by Saka
I would like a script-made rule that runs Life as normal but with icons for all possible sets of neighbors so that it creates a "net" of cells. Here's how it works:
1. A cell is born as state 1

Code: Select all

x = 50, y = 28, rule = LifeHistory
D3.D2.D3.2D.D.2D.3D.D2.D.D.D$2D.2D.D.D.D3.D.D2.D.D.D2.D.D.D$D.D.D.3D.
D.D.D.D2.3D.D2.D2.D$D3.D.D.D2.2D.D.2D.D.D.2D.2D.D3$2D2.3D.3D.D$D.D.D.
D.D3.3D$2D2.D.D.D3.D2.D$D.D.D.D.D3.D2.D15.19F$2D2.3D.D3.D2.D15.F5.F.
3A.F5.F$31.F5.F5AF5.F$31.F5.F5AF5.F$10.D20.F5.F5AF5.F$10.D20.F5.F.3A.
F5.F$10.D20.19F$10.D16.D3.F5.F.3A.F5.F$11.D16.D2.F5.F5AF5.F$11.2D12.
5D.F5.F5AF5.F$12.4D9.D2.D2.F5.F5AF5.F$15.10D2.D3.F5.F.3A.F5.F$31.19F$
31.F5.F.3A.F5.F$31.F5.F5AF5.F$31.F5.F5AF5.F$31.F5.F5AF5.F$31.F5.F.3A.
F5.F$31.19F!
2. The cell detects it's fellow state 1 neighbors and changes to the state with the proper icon:

Code: Select all

x = 19, y = 19, rule = LifeHistory
19F$F5.F5.F5.F$F5.F5.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$
19F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.
F5.F$19F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F5.F5.F$F5.F5.F
5.F$19F!
3. Repeat from step 1, but surviving states also change to state 1:

Code: Select all

x = 19, y = 19, rule = LifeHistory
19F$F5.F5.F5.F$F5.F5.F5.F$F5.F5.F5.F$F5.F5.F5.F$F5.F5.F5.F$19F$F.3A.F
.3A.F.3A.F$F5AF5AF5AF$F5AF5AF5AF$F5AF5AF5AF$F.3A.F.3A.F.3A.F$19F$F5.F
5.F5.F$F5.F5.F5.F$F5.F5.F5.F$F5.F5.F5.F$F5.F5.F5.F$19F!
But if you do the math, all possible neighbors will require 256 states with icons. That, for now, is impossible in Golly. BUT, we can cut down on states by deleting all 1e and 1c states and replacing them with the 2i and 2n states:

Code: Select all

x = 19, y = 19, rule = LifeHistory
19F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.
F5.F$19F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F
2.A2.F5.F$19F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F
$F5.F2.A2.F5.F$19F!
This will cut down on 8 states, making a rule with 246 neighbor states + 1 detect state + Void = 248 states.
PLEASE someone do it, this would be very cool.

Code: Select all

x = 19, y = 19, rule = LifeHistory
19F$F5.F4.AF5.F$F5.F3.A.F5.F$F5.F2.A2.F5.F$F5.F.A3.F5.F$F5.FA4.F5.F$
19F$F4.AF5.F5.F$F3.A.F5.F5.F$F2.A2.F5.F5.F$F2.A2.F5.F5.F$F2.A2.F5.F5.
F$19F$F2.A2.F5.F5.F$F2.A2.F5.F5.F$F2.3AF5AF5AF$F5.F5.F5.F$F5.F5.F5.F$
19F!

Re: Rule request thread

Posted: July 12th, 2017, 10:21 am
by 83bismuth38
i request a stable, totalistic, hexagonal rule with oscillators and ships. :mrgreen:

Re: Rule request thread

Posted: July 12th, 2017, 10:49 am
by BlinkerSpawn
83bismuth38 wrote:i request a stable, totalistic, hexagonal rule with oscillators and ships. :mrgreen:
From the first post:
drc wrote:This is a thread where you can request a rule to be made.
You can't exactly make a totalistic rule unless you allow multiple states.
Making totalistic rules is pretty easy and painless anyway; just set "symmetries:permute" and play around.
Go here if you need a refresher on the .rule format.
Start with 3 states (1 vacuum, 2 ON). If you don't get any satisfactory results, go up to 4.

Re: Rule request thread

Posted: July 12th, 2017, 11:17 am
by wildmyron
83bismuth38 wrote:i request a stable, totalistic, hexagonal rule with oscillators and ships. :mrgreen:
Well, if you insist on asking again, here are 8 rules which all meet this criteria:

B2/S34H
B25/S34H
B26/S34H
B256/S34H
B2/S346H
B25/S346H
B26/S346H
B256/S346H

(Unless I missed checking one of the combinations for explosiveness)
The following ship works in all of them:

Code: Select all

x = 9, y = 11, rule = B2/S34H
bobo$3b3o$b2o$o4bo$4b3o$o3bo$3bobobo$2b2o2b2o$3b2o3bo$4b2obo$6bo!
They all have a range of small common oscillators and most have several other ships at the same speed as the one above.

Re: Rule request thread

Posted: July 12th, 2017, 11:19 am
by 83bismuth38
BlinkerSpawn wrote:
83bismuth38 wrote:i request a stable, totalistic, hexagonal rule with oscillators and ships. :mrgreen:
From the first post:
drc wrote:This is a thread where you can request a rule to be made.
You can't exactly make a totalistic rule unless you allow multiple states.
Making totalistic rules is pretty easy and painless anyway; just set "symmetries:permute" and play around.
Go here if you need a refresher on the .rule format.
Start with 3 states (1 vacuum, 2 ON). If you don't get any satisfactory results, go up to 4.
did i nention it was HEXAGONAL?
edit:
wildmyron wrote:
83bismuth38 wrote:i request a stable, totalistic, hexagonal rule with oscillators and ships. :mrgreen:
Well, if you insist on asking again, here are 8 rules which all meet this criteria:

B2/S34H
B25/S34H
B26/S34H
B256/S34H
B2/S346H
B25/S346H
B26/S346H
B256/S346H

(Unless I missed checking one of the combinations for explosiveness)
The following ship works in all of them:

Code: Select all

x = 9, y = 11, rule = B2/S34H
bobo$3b3o$b2o$o4bo$4b3o$o3bo$3bobobo$2b2o2b2o$3b2o3bo$4b2obo$6bo!
They all have a range of small common oscillators and most have several other ships at the same speed as the one above.
missed this post. thanks anyways.
EDIT 2: it seems that none of them have the ship as a natural occurrence, so t's not very fun ):

Re: Rule request thread

Posted: July 12th, 2017, 11:27 am
by wildmyron
@83bismuth38:

The process for making rule tables for hexagonal rules is the same as for the Moore neighbourhood - just specify "neighborhood: hexagonal" in the @TABLE section and make sure that your transitions have eight elements instead of ten. "Golly -> Help -> File Formats" is a good reference. There are many examples on these forums which you can refer to as well, and if you have problems with something you'd like to do, someone here is more likely to help than to continue doing all the work for you.

Re: Rule request thread

Posted: July 12th, 2017, 11:29 am
by 83bismuth38
i don't do 'RuleLoader' format, too confusing.

Re: Rule request thread

Posted: July 12th, 2017, 7:29 pm
by Saka
83bismuth38 wrote:i don't do 'RuleLoader' format, too confusing.
Tutorials!!!

Re: Rule request thread

Posted: July 14th, 2017, 8:23 am
by Saka
Saka wrote:I would like a script-made rule that runs Life as normal but with icons for all possible sets of neighbors so that it creates a "net" of cells. Here's how it works:
1. A cell is born as state 1

Code: Select all

x = 50, y = 28, rule = LifeHistory
D3.D2.D3.2D.D.2D.3D.D2.D.D.D$2D.2D.D.D.D3.D.D2.D.D.D2.D.D.D$D.D.D.3D.
D.D.D.D2.3D.D2.D2.D$D3.D.D.D2.2D.D.2D.D.D.2D.2D.D3$2D2.3D.3D.D$D.D.D.
D.D3.3D$2D2.D.D.D3.D2.D$D.D.D.D.D3.D2.D15.19F$2D2.3D.D3.D2.D15.F5.F.
3A.F5.F$31.F5.F5AF5.F$31.F5.F5AF5.F$10.D20.F5.F5AF5.F$10.D20.F5.F.3A.
F5.F$10.D20.19F$10.D16.D3.F5.F.3A.F5.F$11.D16.D2.F5.F5AF5.F$11.2D12.
5D.F5.F5AF5.F$12.4D9.D2.D2.F5.F5AF5.F$15.10D2.D3.F5.F.3A.F5.F$31.19F$
31.F5.F.3A.F5.F$31.F5.F5AF5.F$31.F5.F5AF5.F$31.F5.F5AF5.F$31.F5.F.3A.
F5.F$31.19F!
2. The cell detects it's fellow state 1 neighbors and changes to the state with the proper icon:

Code: Select all

x = 19, y = 19, rule = LifeHistory
19F$F5.F5.F5.F$F5.F5.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$
19F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.
F5.F$19F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F5.F5.F$F5.F5.F
5.F$19F!
3. Repeat from step 1, but surviving states also change to state 1:

Code: Select all

x = 19, y = 19, rule = LifeHistory
19F$F5.F5.F5.F$F5.F5.F5.F$F5.F5.F5.F$F5.F5.F5.F$F5.F5.F5.F$19F$F.3A.F
.3A.F.3A.F$F5AF5AF5AF$F5AF5AF5AF$F5AF5AF5AF$F.3A.F.3A.F.3A.F$19F$F5.F
5.F5.F$F5.F5.F5.F$F5.F5.F5.F$F5.F5.F5.F$F5.F5.F5.F$19F!
But if you do the math, all possible neighbors will require 256 states with icons. That, for now, is impossible in Golly. BUT, we can cut down on states by deleting all 1e and 1c states and replacing them with the 2i and 2n states:

Code: Select all

x = 19, y = 19, rule = LifeHistory
19F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.
F5.F$19F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F
2.A2.F5.F$19F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F
$F5.F2.A2.F5.F$19F!
This will cut down on 8 states, making a rule with 246 neighbor states + 1 detect state + Void = 248 states.
PLEASE someone do it, this would be very cool.

Code: Select all

x = 19, y = 19, rule = LifeHistory
19F$F5.F4.AF5.F$F5.F3.A.F5.F$F5.F2.A2.F5.F$F5.F.A3.F5.F$F5.FA4.F5.F$
19F$F4.AF5.F5.F$F3.A.F5.F5.F$F2.A2.F5.F5.F$F2.A2.F5.F5.F$F2.A2.F5.F5.
F$19F$F2.A2.F5.F5.F$F2.A2.F5.F5.F$F2.3AF5AF5AF$F5.F5.F5.F$F5.F5.F5.F$
19F!
pls

Re: Rule request thread

Posted: July 15th, 2017, 10:26 am
by Naszvadi
Created a reporting script about 1 dimensional elementary margolus rules (16 only) and related Wolfram ones. Script is a bare pythons script without any imports.

Spoiler:

Code: Select all

 $ ./1da.py | sed -nre '/1D margolus/,${p;b};/....[^0-9](0|34|48|90|128|165|170|187|200|236|240|243|254|255)[^0-9]/{p;b};/Margolus[][]|^$/{p;b};/1D/p'
1D Wolfram rules (order / rule /
  • / smallest contained margolus rule(s)):
    • 1: 0: [0, 255], Margolus[0, 15]
    • 4: 3: [3, 17, 63, 119], Margolus[1, 7]
    • 6: 5: [5, 95], Margolus[1, 7]
    • 11: 10: [10, 80, 175, 245], Margolus[2, 4, 11, 13]
    • 13: 12: [12, 68, 207, 221], Margolus[2, 4, 11, 13]
    • 16: 15: [15, 85], Margolus[3, 5]
    • 30: 34: [34, 48, 187, 243], Margolus[2, 4, 11, 13]
    • 43: 51: [51], Margolus[3, 5]
    • 48: 60: [60, 102, 153, 195], Margolus[6, 9]
    • 56: 90: [90, 165], Margolus[6, 9]
    • 65: 128: [128, 254]
    • 69: 136: [136, 192, 238, 252], Margolus[8, 14]
    • 78: 160: [160, 250], Margolus[8, 14]
    • 82: 170: [170, 240], Margolus[10, 12]
    • 86: 200: [200, 236]
    • 87: 204: [204], Margolus[10, 12]
    1D margolus rules (order / rule /
    • ):
      1. : 0: [0, 15]
      2. : 1: [1, 7]
      3. : 2: [2, 4, 11, 13]
      4. : 3: [3, 5]
      5. : 6: [6, 9]
      6. : 8: [8, 14]
      7. : 10: [10, 12]
      1D Wolfram rules composed from margolus ones (order / rule / composed Wolfram rule):
      1. : 0: 0
      2. : 1: 236
      3. : 2: 34
      4. : 3: 240
      5. : 4: 48
      6. : 5: 170
      7. : 6: 90
      8. : 7: 200
      9. : 8: 128
      10. : 9: 165
      11. : 10: 170
      12. : 11: 187
      13. : 12: 240
      14. : 13: 243
      15. : 14: 254
      16. : 15: 255
      Notice that rule pairs [128, 254] and [200, 236] the exceptions as mappings, their ranges depend on all 3 neighbours.

      Is there a Turing-complete elementary margolus rule? Should be odd, if exists any, and also [1, 2, 3] the only candidates due to their trivial equivalence classes' partitions.

      Corollary for Wolfram CA: there are strictry less than 88 distinct elementary cellular automata.

Re: Rule request thread

Posted: July 20th, 2017, 12:04 pm
by 83bismuth38
could you tame this rule, keeping these 2 ships:

Code: Select all

x = 8, y = 16, rule = B34-air5/S234-ai
b3o$2obo$o2bo$ob2o$b2o6$5bo$4b3o$b3o3bo$2o3b2o$bo3bo$2bo!

Re: Rule request thread

Posted: July 20th, 2017, 1:30 pm
by AbhpzTa
83bismuth38 wrote:could you tame this rule, keeping these 2 ships:

Code: Select all

x = 8, y = 16, rule = B34-air5/S234-ai
b3o$2obo$o2bo$ob2o$b2o6$5bo$4b3o$b3o3bo$2o3b2o$bo3bo$2bo!
MIN : B3-ky4knwz5-cekn/S23-cky4jnqrwy
MAX : B2n34-ar56eik/S234-ai5ejq6ei

MIN rule is explosive ...

Re: Rule request thread

Posted: July 20th, 2017, 1:54 pm
by Rhombic
AbhpzTa wrote: MIN rule is explosive ...
A very ignificant number of rules surprisingly become LESS explosive when certain conditions are added. How could a script be written to find these?

Re: Rule request thread

Posted: July 20th, 2017, 2:35 pm
by 83bismuth38
AbhpzTa wrote:
83bismuth38 wrote:could you tame this rule, keeping these 2 ships:

Code: Select all

x = 8, y = 16, rule = B34-air5/S234-ai
b3o$2obo$o2bo$ob2o$b2o6$5bo$4b3o$b3o3bo$2o3b2o$bo3bo$2bo!
MIN : B3-ky4knwz5-cekn/S23-cky4jnqrwy
MAX : B2n34-ar56eik/S234-ai5ejq6ei

MIN rule is explosive ...
so is max, and i can't find a nonexplosive rule with it because stating the min and max is incredibly vague
edit: it's too vauge.

Re: Rule request thread

Posted: July 20th, 2017, 2:39 pm
by drc
83bismuth38 wrote: so is max, and i can't find a nonexplosive rule with it because stating the min and max is incredibly vague
oh yeah, let's just have them go through every single combination of extraneous transitions. All approximately 30 of them.

Just incase you don't understand, poking through all of those rules would result in 2^30=1073741824 different strings, and it's very likely none of those are non-explosive. A rule this explosive is hard to tame with that set of transitions. I tried manually and couldn't get anywhere.

Min/Max rules are actually really useful in this sort of poking around for non-explosive rules. Cause you know which transitions to turn on and off.

Re: Rule request thread

Posted: July 20th, 2017, 2:44 pm
by 83bismuth38
drc wrote:
83bismuth38 wrote: so is max, and i can't find a nonexplosive rule with it because stating the min and max is incredibly vague
oh yeah, let's just have them go through every single combination of extraneous transitions. All approximately 30 of them.

Just incase you don't understand, poking through all of those rules would result in 2^30=1073741824 different strings, and it's very likely none of those are non-explosive. A rule this explosive is hard to tame with that set of transitions. I tried manually and couldn't get anywhere.

Min/Max rules are actually really useful in this sort of poking around for non-explosive rules. Cause you know which transitions to turn on and off.
oh ):
so much for taming the first rule i created with an oblique ship...
rip:

Code: Select all

x = 4, y = 5, rule = B3-ky4knwz5-cekn/S23-cky4jnqrwy
b2o$o2bo$3bo$4o$bobo!

Re: Rule request thread

Posted: July 30th, 2017, 9:54 pm
by muzik
Can someone find a rule where there exists all speeds of form 2c/n, where n is odd, similarly to the two rules discovered recently? 2c/3, 2c/5 and 2c/7 need not count.

Here's a possible tail for such a ship family:

Code: Select all

x = 23, y = 4, rule = B2c3ajr4i5-ay6c7/S123eknqr4etz56ac7
o$o2b19o$b3o18bo$3b19o!

Re: Rule request thread

Posted: August 14th, 2017, 4:11 am
by gameoflifemaniac
How about a Larger thatn Life rule emulated with a 16-state rule with icons?
The icons would look like this.

Re: Rule request thread

Posted: August 14th, 2017, 6:06 am
by Saka
Saka wrote:
Saka wrote:I would like a script-made rule that runs Life as normal but with icons for all possible sets of neighbors so that it creates a "net" of cells. Here's how it works:
1. A cell is born as state 1

Code: Select all

x = 50, y = 28, rule = LifeHistory
D3.D2.D3.2D.D.2D.3D.D2.D.D.D$2D.2D.D.D.D3.D.D2.D.D.D2.D.D.D$D.D.D.3D.
D.D.D.D2.3D.D2.D2.D$D3.D.D.D2.2D.D.2D.D.D.2D.2D.D3$2D2.3D.3D.D$D.D.D.
D.D3.3D$2D2.D.D.D3.D2.D$D.D.D.D.D3.D2.D15.19F$2D2.3D.D3.D2.D15.F5.F.
3A.F5.F$31.F5.F5AF5.F$31.F5.F5AF5.F$10.D20.F5.F5AF5.F$10.D20.F5.F.3A.
F5.F$10.D20.19F$10.D16.D3.F5.F.3A.F5.F$11.D16.D2.F5.F5AF5.F$11.2D12.
5D.F5.F5AF5.F$12.4D9.D2.D2.F5.F5AF5.F$15.10D2.D3.F5.F.3A.F5.F$31.19F$
31.F5.F.3A.F5.F$31.F5.F5AF5.F$31.F5.F5AF5.F$31.F5.F5AF5.F$31.F5.F.3A.
F5.F$31.19F!
2. The cell detects it's fellow state 1 neighbors and changes to the state with the proper icon:

Code: Select all

x = 19, y = 19, rule = LifeHistory
19F$F5.F5.F5.F$F5.F5.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$
19F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.
F5.F$19F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F5.F5.F$F5.F5.F
5.F$19F!
3. Repeat from step 1, but surviving states also change to state 1:

Code: Select all

x = 19, y = 19, rule = LifeHistory
19F$F5.F5.F5.F$F5.F5.F5.F$F5.F5.F5.F$F5.F5.F5.F$F5.F5.F5.F$19F$F.3A.F
.3A.F.3A.F$F5AF5AF5AF$F5AF5AF5AF$F5AF5AF5AF$F.3A.F.3A.F.3A.F$19F$F5.F
5.F5.F$F5.F5.F5.F$F5.F5.F5.F$F5.F5.F5.F$F5.F5.F5.F$19F!
But if you do the math, all possible neighbors will require 256 states with icons. That, for now, is impossible in Golly. BUT, we can cut down on states by deleting all 1e and 1c states and replacing them with the 2i and 2n states:

Code: Select all

x = 19, y = 19, rule = LifeHistory
19F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.
F5.F$19F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F
2.A2.F5.F$19F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F$F5.F2.A2.F5.F
$F5.F2.A2.F5.F$19F!
This will cut down on 8 states, making a rule with 246 neighbor states + 1 detect state + Void = 248 states.
PLEASE someone do it, this would be very cool.

Code: Select all

x = 19, y = 19, rule = LifeHistory
19F$F5.F4.AF5.F$F5.F3.A.F5.F$F5.F2.A2.F5.F$F5.F.A3.F5.F$F5.FA4.F5.F$
19F$F4.AF5.F5.F$F3.A.F5.F5.F$F2.A2.F5.F5.F$F2.A2.F5.F5.F$F2.A2.F5.F5.
F$19F$F2.A2.F5.F5.F$F2.A2.F5.F5.F$F2.3AF5AF5AF$F5.F5.F5.F$F5.F5.F5.F$
19F!
pls
Extra pls

Re: Rule request thread

Posted: August 16th, 2017, 4:21 pm
by spaceman00
Can someone do a rule like this?
Would this rule be feasible to do?

Its 3d conway game of life but its 2d with height being represented with colors. The binary representation of the hex representation of of the colors, tell what cells at what height are alive or dead. 0 means dead 1 means alive. so the color of 000000000000000000000000 (thats 24 zeroes) means all cells are dead but a color of 000000000000000000000001 (23 zeroes and the last number is a one), means the bottom cell is alive and all cells at the top of it are dead.

By using all rgb colors, you will have a height of 24, because 2^24 = 256*256*256.

3d game of life is a little confuse to watch, with height and their states being represented by color, you maybe would make it become less confuse.

Re: Rule request thread

Posted: August 16th, 2017, 5:00 pm
by BlinkerSpawn
spaceman00 wrote:Can someone do a rule like this?
Would this rule be feasible to do?

Its 3d conway game of life but its 2d with height being represented with colors. The binary representation of the hex representation of of the colors, tell what cells at what height are alive or dead. 0 means dead 1 means alive. so the color of 000000000000000000000000 (thats 24 zeroes) means all cells are dead but a color of 000000000000000000000001 (23 zeroes and the last number is a one), means the bottom cell is alive and all cells at the top of it are dead.

By using all rgb colors, you will have a height of 24, because 2^24 = 256*256*256.

3d game of life is a little confuse to watch, with height and their states being represented by color, you maybe would make it become less confuse.
This rule would have 16777216 states.
Golly currently only supports rules with up to 256 states.
However, you *could* do a simple 1-bit RGB to get height 3, and possibly have the rule act on a 4D torus to get superstring-type action, if desired.
All times are UTC-04:00
Page 11 of 44

Powered by phpBB® Forum Software © phpBB Limited