Page 78 of 209
Re: Thread for basic questions
Posted: June 13th, 2020, 1:32 am
by Hunting
If one were to create the OCA:LeapLife article, will it be Stable or Chaotic?
Re: Thread for basic questions
Posted: June 15th, 2020, 1:14 am
by muzik
bprentice wrote: ↑June 5th, 2020, 9:59 pm
Is anyone interested in exploring 1D CA rules?
Brian Prentice
This depends on a lot of things. Rule 2166637080 has been shown to have a highly diverse set of spaceships.
Re: Thread for basic questions
Posted: June 16th, 2020, 7:34 am
by Bullet51
What search programs can search for diagonal spaceships in this kind of region?
Code: Select all
x = 32, y = 20, rule = LifeHistory
4B28F$6B26F$5BA2B24F$3B2ABABAB22F$3B2A2BA4B20F$3BA3BA2BA3B18F$3BA4B3A
5B16F$4BA4B3ABA4B14F$2FB2A8B3A4B12F$4F5BA4B4ABA2B10F$6F3BA6BA2BA4B8F$
8F2BA2BA2BA6BA2B6F$10F2BA3B2A3B3A4B4F$12F4BAB2A2BA7B2F$14F4B2A2BAB2A
6B$16F2BA5BA7B$18F2BABA9B$20FA11B$22F10B$24F8B!
The green pattern is a (1,1)c/6 spaceship partial going to NW.
Re: Thread for basic questions
Posted: June 16th, 2020, 8:33 am
by praosylen
Bullet51 wrote: ↑June 16th, 2020, 7:34 am
What search programs can search for diagonal spaceships in this kind of region?
As far as I know, WLS/JLS are the only modern ones that can confine a partial to an arbitrarily-shaped region, although both LSSS and ikpx support floating rows and could in theory find this partial during a search of a broader search space.
Re: Thread for basic questions
Posted: June 16th, 2020, 9:27 am
by wildmyron
It's also possible to extend a partial like this using logic life search, but it would only be feasible to find a completion if it's relatively short. I'd guess that a 10 column extension is doable, perhaps a bit longer if searching for a partial, but currently a new search needs to be run for each partial to be found (this is a limitation of LLS, not the SAT solver it's driving). My intuition from a comparison of JLS and LLS with an experimental completion of Sir Robin partials says that: if there's a short completion then finding it would be quicker with LLS, but without that apriori information it's much more convenient and efficient to search with WLS (or JLS).
Re: Thread for basic questions
Posted: June 18th, 2020, 9:13 am
by gameoflifemaniac
Is a small spaceship with a front spark known?
Re: Thread for basic questions
Posted: June 18th, 2020, 3:12 pm
by dvgrn
gameoflifemaniac wrote: ↑June 18th, 2020, 9:13 am
Is a small spaceship with a front spark known?
Depends on your definition of "small", I guess. Maybe start by looking through the list of
non-monotonic spaceships and see if you think anything there fits the bill.
There are some ships, like the edge-repair spaceships, that don't precisely have front sparks, but they can have spark-like effects on object in front of them and off to one side... and there are things like
37P4H1V0, which has an isolated dot spark along the edge in one phase, but the period is so low that it's kind of hard to find anything useful for the spark to do.
Maybe other people know what the current record-holder is for a spaceship with a useful front spark? What's smaller than
114P6H1V0?
Re: Thread for basic questions
Posted: June 18th, 2020, 3:37 pm
by gameoflifemaniac
dvgrn wrote: ↑June 18th, 2020, 3:12 pm
gameoflifemaniac wrote: ↑June 18th, 2020, 9:13 am
Is a small spaceship with a front spark known?
Depends on your definition of "small", I guess. Maybe start by looking through the list of
non-monotonic spaceships and see if you think anything there fits the bill.
There are some ships, like the edge-repair spaceships, that don't precisely have front sparks, but they can have spark-like effects on object in front of them and off to one side... and there are things like
37P4H1V0, which has an isolated dot spark along the edge in one phase, but the period is so low that it's kind of hard to find anything useful for the spark to do.
Maybe other people know what the current record-holder is for a spaceship with a useful front spark? What's smaller than
114P6H1V0?
After someone said that a smaller sawtooth could be built using a ship burning a blinker fuse, I thought about that a lot.
Re: Thread for basic questions
Posted: June 19th, 2020, 8:11 am
by Hunting
What kind of methuselah-of-specific-cell-count-that-fits-in-a-certain-bounding-box searches should be done with LeapLife?
Re: Thread for basic questions
Posted: June 23rd, 2020, 5:37 pm
by Hdjensofjfnen
Is there a ruletable for Immigration somewhere?
Re: Thread for basic questions
Posted: June 23rd, 2020, 5:52 pm
by dvgrn
Hdjensofjfnen wrote: ↑June 23rd, 2020, 5:37 pm
Is there a ruletable for Immigration somewhere?
What's wrong with
the one in the LifeWiki Rule namespace?
Code:
Select all
x = 36, y = 9, rule = Immigration
24.A$22.A.A$12.2B6.2A12.2B$11.B3.B4.2A12.2B$2A8.B5.B3.2A$2A8.B3.B.2B
4.A.A$10.B5.B7.A$11.B3.B$12.2B!
#C [[ AUTOSTART THUMBNAIL THUMBSIZE 3 ]]
Long story short, if anyone knows of Golly rule tables that
aren't available in the LifeWiki Rule namespace at this point, they should please go ahead and put them there.
Re: Thread for basic questions
Posted: June 27th, 2020, 4:08 am
by gama
Not sure where to put this, but how can I save images in Golly?
Re: Thread for basic questions
Posted: June 27th, 2020, 9:54 am
by dvgrn
gama wrote: ↑June 27th, 2020, 4:08 am
Not sure where to put this, but how can I save images in Golly?
Depends on what you mean by "image". For just single images, maybe just take a screenshot. For communicating here on the forums, it's usually better to copy out an RLE of the pattern (Ctrl+A, Ctrl+C) so that the pattern can be viewed in LifeViewer and copied back into Golly.
For collecting sequences of images, there's giffer.lua (in newer versions of Golly) or giffer.py. Or you can
hack lifeviewer.lua to get much nicer pictures.
Re: Thread for basic questions
Posted: June 30th, 2020, 5:21 pm
by RakeGuy1246
dvgrn wrote: ↑November 18th, 2018, 12:58 pm
People should be able to make whatever metapatterns they want, by cloning the slmake repository and following calcyman's instructions... and much good may it do them, given that it's so impressively tedious to run even a single cell through one metatick.
metaglider.mc.gz
Because of the way this metapixel works, you should be able to build the pattern out of itself and see which meta-metapixels are currently on.
That's the proplem you run into with the OTCA meta-metablinker. You can only see which ones are on and which are off up to the first meta, but since, with this one, the metapixels kill themselves if they're in the "off" state, you can go up as many metas as you want and still be able to clearly see which metapixels in that layer are turned on. If someone wants to build the 0e0p meta-metablinker, be my guest.
Re: Thread for basic questions
Posted: June 30th, 2020, 8:37 pm
by dvgrn
RakeGuy1246 wrote: ↑June 30th, 2020, 5:21 pm
If someone wants to build the 0e0p meta-metablinker, be my guest.
It might be interesting to look at in Golly, actually. A
meta-glider is 7.7MB, so presumably a meta-meta-blinker would be on the order of 15MB. Probably a few megabytes bigger, since the average hashtile index will have another digit or two in it.
I think Golly wouldn't have any trouble zooming and panning a structure like that. However, the best way to
simulate it would definitely be to build the two meta-meta-phases separately with calcyman's tools, drop them into the same universe at a respectful power-of-two distance from each other, and then write a script to shift the view back and forth between them.
Re: Thread for basic questions
Posted: July 3rd, 2020, 11:30 pm
by Hunting
1. Is it possible to adapt knightt for INT rules?
2. How to modify dr to support specific INT rules?
Re: Thread for basic questions
Posted: July 14th, 2020, 5:03 am
by NoAyeBeardo
Can a gun be made using spaceship converters?
Re: Thread for basic questions
Posted: July 14th, 2020, 12:17 pm
by mniemiec
Hunting wrote: ↑July 3rd, 2020, 11:30 pm
1. Is it possible to adapt knightt for INT rules?
2. How to modify dr to support specific INT rules?
I keep hearing mentions of INT rules, but I'm not sure what that term means, and can't find any references to it. Where is it defined?
Re: Thread for basic questions
Posted: July 14th, 2020, 12:43 pm
by praosylen
mniemiec wrote: ↑July 14th, 2020, 12:17 pm
I keep hearing mentions of INT rules, but I'm not sure what that term means, and can't find any references to it. Where is it defined?
It's just an abbreviation for isotropic non-totalistic.
Re: Thread for basic questions
Posted: July 14th, 2020, 12:56 pm
by wildmyron
NoAyeBeardo wrote: ↑July 14th, 2020, 5:03 am
Can a gun be made using spaceship converters?
If the converter input is a glider and the output is a different spaceship then a spaceship gun is made by driving the converter with any glider gun having period greater than the repeat time of the converter. Is that what you were asking about?
Re: Thread for basic questions
Posted: July 14th, 2020, 12:59 pm
by wildmyron
Hunting wrote: ↑July 3rd, 2020, 11:30 pm
1. Is it possible to adapt knightt for INT rules?
Probably, but given Tim Coe's comments about his own coding style I don't expect it would be straightforward.
2. How to modify dr to support specific INT rules?
Also can't help you much with this one, but there is/was a modification for dr to support Just Friends (included in the code and enabled by setting a variable). Looking at the code associated with that variable would be a good starting point.
Re: Thread for basic questions
Posted: July 14th, 2020, 2:22 pm
by mniemiec
mniemiec wrote: ↑July 14th, 2020, 12:17 pm
I keep hearing mentions of INT rules, but I'm not sure what that term means, and can't find any references to it. Where is it defined?
A for awesome wrote: ↑July 14th, 2020, 12:43 pm
It's just an abbreviation for isotropic non-totalistic.
Back in the 1990s, I played with a rule I called Integer Life.
1) Assume a Moore neighborhood, except every cell has an integer value
2) At every step, a cell's neighborhood is the arithmetic sum of the values of its 8 neighbors.
3) If a cell is dead (i.e. value=0), and has a neighborhood of exactly 3n, a new cell of value n is born there; otherwise it stays dead.
4) If a cell is alive (i.e. value=n, n!=0), and has exactly 2n or 3n neighbors, it remains alive; otherwise, it dies. (As it happens, the n!=0 condition is redundant).
5) There is some room for experimentation with regards to what happens with neighborhoods aren't exact multiples of n; I have found that "round down" works the best, i.e. survival in the range [2n...4n); some other more permissive combinations can make a rule that is explosive.
This rule behaves very similarly to life, with some minor exceptions. The exteriors of soups behave much as in Life, but the interiors work very differently; it's a bit like the interior of a star, with high temperatures causing creation of heavier elements. Most stable patterns and oscillators tend to be of one state, but a fair number of hybrids of multiple states (most commonly, 2 in the interior and 1 on the outside) occur
One interesting feature is scaling; any pattern containing cells of states 1, 2, 3, ... is isomorphic to a similar pattern containing cells of states n, 2n, 3n... for any integer n that is not a multiple of 3. One can even extend this from integers to rational numbers whose denominators are not multiples of 3.
Besides supporting most patterns from Life (notably excluding anything involving dead cells with 6 neighbors, as does HighLife, e.g. ship and dead spark coil), there is also a natural 2c/6 orthogonal spaceship:
Re: Thread for basic questions
Posted: July 14th, 2020, 8:46 pm
by Andrew
mniemiec wrote: ↑July 14th, 2020, 2:22 pm
Besides supporting most patterns from Life (notably excluding anything involving dead cells with 6 neighbors, as does HighLife, e.g. ship and dead spark coil), there is also a natural 2c/6 orthogonal spaceship:
Are you sure the above pattern is correct? I've implemented your rule in Golly (using NewCA.lua) but that pattern dies out in 12 gens.
I'm pretty sure my implementation is correct. I see everything else you described; eg. the isomorphism. A ship dies out in 3 gens and a dead spark coil dies out in 8 gens -- is that correct?
EDIT: I found the problem. My test for survival was based on your point 4:
Code: Select all
if (n == 2 * state) or (n == 3 * state) then
-- this live cell survives
I didn't pay enough attention to point 5. When I changed the test to
Code: Select all
if (n >= 2 * state) and (n < 4 * state) then
-- this live cell survives
Now that spaceship works.
Re: Thread for basic questions
Posted: July 15th, 2020, 1:14 am
by mniemiec
Andrew wrote: ↑July 14th, 2020, 8:46 pm
I've implemented your rule in Golly (using NewCA.lua) ... A ship dies out in 3 gens and a dead spark coil dies out in 8 gens -- is that correct?
Yes, those are correct. I didn't think it would be possible to implement this in Golly without a lot of work. I'd be very curious to see your script for it.
One interesting thing is that it's possible to use 1-gliders to make 2-gliders (and thus, by induction n-gliders for any n that's a power of 2). State 4 is easy to achieve by doubling twice; states 5 and higher would probably take a lot more work. Here is a synthesis of the 2c/6 spaceship, and 3 syntheses for 2-gliders. I hope I converted them properly. (You'll have to supply the rule descriptor yourself, of course).
Code: Select all
x = 160, y = 87
99.AA19.A19.A$98.AA19.A.A17.A.A$100.A17.A..A16.A..A$95.3A21.AA18.AA$
97.A$96.A$136.3A$138.A$137.A19.3A$156.A.B$156.ABB$156.A.B$159.A7$11.A
36.A$12.A33.A.A58.AA18.AA$10.3A34.AA57.A.A17.A.A$30.AA18.AA15.3A17.3A
15.A19.A6.A$29.A.A17.A.A16.3A17.3A13.A19.A6.AA$11.3A16.A19.A53.AA18.AA
5.A.A$13.A77.3A$12.A79.B.A$92.BBA$92.B.A$91.A5$157.3B$157.B$158.B4$
102.A$103.A$101.3A8$66.A.A17.A30.A3.A4.A$66.AA17.A.A30.3BA3.A.A$67.A
17.AA32.B.A3.AA$118.3A$47.BB18.BB18.BB38.BB$47.BB18.BB18.BB38.BB$49.A
19.A19.A39.A4$32.AA$32.A.A$28.A3.A$29.AA$28.AA127.3B$157.B$158.B3$71.
3A39.3A$56.AA15.A..AA36.B.A$35.AA18.A.A14.A..A.A20.A14.3BA.A$19.BB15.
AA.BB15.A..BB15.A..BB18.BB11.A3.A..BB$19.BB14.A3.BB18.BB18.BB18.BB18.
BB$21.A19.A19.A19.A19.A19.A$$138.AA$138.AA$4.AA$4.A.A$A3.A$.AA$AA$$
157.3B$157.B$158.B!
Re: Thread for basic questions
Posted: July 15th, 2020, 8:10 am
by Andrew
mniemiec wrote: ↑July 15th, 2020, 1:14 am
I didn't think it would be possible to implement this in Golly without a lot of work. I'd be very curious to see your script for it.
See my latest post in the
NewCA.lua thread. Worth reading some of the early posts in that thread if you haven't used a script based on NewCA. The zip file linked there contains IntegerLife.lua and a few rle files, including one with the above synthesis.