ConwayLife.com - A community for Conway's Game of Life and related cellular automata
Home  •  LifeWiki  •  Forums  •  Download Golly

3c/7 othogonal and 2c/9 diagonal spaceships

For discussion of specific patterns or specific families of patterns, both newly-discovered and well-known.

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby Sokwe » December 21st, 2015, 6:02 am

wildmyron wrote:Codeholic's table of Known spaceships by width has some fairly small numbers on the row for (1,0)c/8.

The numbers for (1,0)c/7 are also fairly small. In fact, width-13 (1,0)c/7 odd-symmetric doesn't take more than about an hour with gfind and reaches a somewhat promising depth of 291 (approx. 41 rows). The longest partial patterns all had the same front end:
x = 11, y = 41, rule = B3/S23
2bobobobo$bo2bobo2bo$5ob5o$o2bo3bo2bo$4bobo$o3bobo3bo$b3o3b3o2$5bo$2ob
2ob2ob2o$3o5b3o$2b3ob3o2$bob2ob2obo$bo7bo$4bobo$4bobo$2bobobobo$2bobob
obo$3bo3bo$3b5o$3bo3bo$2b3ob3o$2bobobobo$2bo5bo$3bo3bo$4b3o$3bo3bo$3bo
3bo$2bo5bo$b2o5b2o$4bobo$bobo3bobo2$b4ob4o$3bo3bo2$b2o5b2o$o2bo3bo2bo$
bobo3bobo$2bo5bo!


I am currently running the following searches
  • (3,0)c/8 odd-symmetric, width 21 (with knight2)
  • (3,0)c/8 gutter, width 23 (with knight2) Edit: Done.
  • (1,1)c/6, width 9 (with Paul Tooke's gfind modification) Edit: Done. Started width-10 search

If anyone completes a search not yet covered in the table, either update the table (with a comment on what program was used), or post the result on the forums.

Edit: the (1,1)c/6 width-9 search completed: negative result, as expected. The longest partials start with a wing-like structure that probably isn't useful:
x = 25, y = 30, rule = B3/S23
o3bo13bo$b4o13bo2bo$b3o14bob3o$b2ob2o12bobo2b2o$b2ob2o11b2obob2o$b2o$
3bo14bo3bo$bo16bo$bobobo12bo3bo$2bo2b2o12b2obo$3bob2o11bo2bo$5b2o11bob
o$3bob2o11bob2o$5bo15b2o$22bobo$4bo17bobo$5b2o16b2o$3b3o15b3o$3b2o16b
2o$3bo17bo$3bo17bo$2bo17bo$2bo17bo$o4bo12bo4bo$ob3o13bob3o$obo15bobo2$
3b2o16b2o$2b2o16b2o$2bo17bo!

I've started the width-10 search, but I don't expect it to find anything.
-Matthias Merzenich
Sokwe
Moderator
 
Posts: 1255
Joined: July 9th, 2009, 2:44 pm

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby codeholic » December 21st, 2015, 8:17 pm

@moebius I'd like to use an opportunity to ask you a question. I have been always wondering, how you found a period-16 spaceship, that carries your name, back in 1995? That's something that is still beyond my imagination. Or was it a partial result that luckily transformed into a smoking ship?
Ivan Fomichev
User avatar
codeholic
Moderator
 
Posts: 1140
Joined: September 13th, 2011, 8:23 am
Location: Hamburg, Germany

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby Scorbie » December 21st, 2015, 9:33 pm

If I remember correctly, the Coe ship is a weakly connected partial of a 2c/4 spaceship.
EDIT: This one:
x = 15, y = 20, rule = B3/S23
2b2o$2ob2o$4o8b2o$b2o2b2ob4ob2o$5bo3b5o$10b3o3$10b4o$9bo3bo$13bo$12bo$
9b2o2$5bob2o3b2o$7bob3ob2o$2bo2bo2b6o$2ob2o4b4o$4o$b2o!
So sadly it can't be generalized...
Last edited by Scorbie on December 22nd, 2015, 6:57 pm, edited 2 times in total.
Best wishes to you, Scorbie
User avatar
Scorbie
 
Posts: 1350
Joined: December 7th, 2013, 1:05 am

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby wildmyron » December 22nd, 2015, 10:57 am

Sokwe wrote:If anyone completes a search not yet covered in the table, either update the table (with a comment on what program was used), or post the result on the forums.

Should there also be some indication when a ship is known but not known to be minimal - whilst still keeping the results for negative searches? For example: 2c/6, even symmetry (though I think we can push this up to known ship at width 18 now); most of the 2c/8 row; c/6 diagonal; etc.

wildmyron wrote:I'm currently running 'gfind o8n1vl112' just to have a look at the even symmetry partials as well. If no one else is running them yet, I'll run the l128 searches over the next week or two.

No surprises there - the search was unsuccessful. Took about 10 hours. I'm running the odd and even symmetry searches at l128 now. Best partials:
x = 164, y = 31, rule = B3/S23
100bo3bo4bo3bo$51bo4b2o4bo41b2o2b2o$b3o6b3o39b10o38bo3bo4bo3bo40b2o2b
2o$b2o8b2o39bo8bo38bo2bo6bo2bo40b6o$3b3o2b3o43b6o41bo10bo37b2o10b2o$4b
2o2b2o42bo3b2o3bo39bo10bo37b3o8b3o$3bo6bo41bobo4bobo40b2o6b2o39b3o6b3o
$b5o2b5o39bob2o2b2obo40bob2o2b2obo40bob2o2b2obo$bo4b2o4bo40bobo2bobo
42bobo2bobo42bobo2bobo$2b2obo2bob2o$3b8o43b2o2b2o44b2o2b2o44b2o2b2o$6b
2o45b3o2b3o42b3o2b3o42b3o2b3o$4b2o2b2o$4b2o2b2o42bo3b2o3bo40bo3b2o3bo
40bo3b2o3bo$3b3o2b3o$b2ob2o2b2ob2o41bob2obo44bob2obo44bob2obo$o3b2o2b
2o3bo39b2o4b2o42b2o4b2o42b2o4b2o$bo4b2o4bo41bob2obo44bob2obo44bob2obo$
5b4o45b2o2b2o44b2o2b2o44b2o2b2o$2b3o4b3o42bo4bo44bo4bo44bo4bo$3b2o4b2o
43bo4bo44bo4bo44bo4bo$53b3o2b3o42b3o2b3o42b3o2b3o$4bo4bo$3bo6bo43bo4bo
44bo4bo44bo4bo$bob2o4b2obo38b2ob2o2b2ob2o38b2ob2o2b2ob2o38b2ob2o2b2ob
2o$3b2o4b2o40b2o2bo2bo2b2o38b2o2bo2bo2b2o38b2o2bo2bo2b2o$o12bo40bo4bo
44bo4bo44bo4bo$obo8bobo37b2o8b2o38b2o8b2o38b2o8b2o$4b2o2b2o44bo4bo44bo
4bo44bo4bo$2bobo4bobo40b2o6b2o40b2o6b2o40b2o6b2o$3bo6bo42bo6bo42bo6bo
42bo6bo!


@Sokwe: I appreciate you keeping the table up to date. You can use my real name (Arie Paap) if you are so inclined.
wildmyron
 
Posts: 790
Joined: August 9th, 2013, 12:45 am

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby Scorbie » December 22nd, 2015, 12:22 pm

wildmyron wrote:@Sokwe: I appreciate you keeping the table up to date. You can use my real name (Arie Paap) if you are so inclined.
Hmm. That name seems familiar... Do I know you?
Best wishes to you, Scorbie
User avatar
Scorbie
 
Posts: 1350
Joined: December 7th, 2013, 1:05 am

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby moebius » December 22nd, 2015, 2:54 pm

Codeholic,

I would run searches and produce giant armadas of 2c/4 ships. I would tend to run them then and blow them up. One time the period 16 smoker emerged after trimming the tail of some ship. It was basically total luck.

Wildmyron,

The problem with the width 16 2c/6 searches is they always miss stuff. I ran a bunch of different searches trying to cover the search tree, but I could never be sure I wasn't missing something. For example, I was looking through my turtle armadas with their variety of tags I never ran across the 9th one in your pictures. Your list of ships does not include the following ones, or anything very close:

x = 60, y = 39, rule = B3/S23
6bo2bo18bo2bo$4bobo2bobo14bobo2bobo14bobo2bobo$5bo4bo16bo4bo15b2o4b2o$
5bo4bo16bo4bo17bo2bo2$5b2o2b2o16b2o2b2o15b2o4b2o2$4bobo2bobo14bobo2bob
o14bo6bo$bobo2bo2bo2bobo8bobo2bo2bo2bobo$2obob2o2b2obob2o6b2obob2o2b2o
bob2o9b3o4b3o$2obobo4bobob2o6b2obobo4bobob2o11bo4bo$3bobo4bobo12bobo4b
obo11b2obo4bob2o$2b2obob2obob2o10b2obob2obob2o10bo10bo$3bo2bo2bo2bo12b
o2bo2bo2bo11bo3bo2bo3bo$3bo2bo2bo2bo12bo2bo2bo2bo13b2o4b2o$4b2o4b2o14b
2o4b2o12bobo6bobo$44b3o10b3o$4b2o4b2o14b2o4b2o11b2o10b2o$3bo8bo10bo4bo
2bo4bo9bobo6bobo$5bo4bo11bobob3o2b3obobo8bo2bo4bo2bo$2bo2bo4bo2bo9b3o
2b4o2b3o10b2o6b2o$bo3bob2obo3bo13bo2bo$2b5o2b5o15b2o16b2o6b2o$3bob2o2b
2obo36bob2obo$6bo2bo15b2o6b2o10bo2b3o2b3o2bo$26b8o10bo5bo2bo5bo$45b3ob
2o2b2ob3o$3b2o6b2o13bo6bo15bo4bo$4b8o14bo6bo16bo2bo$28bo2bo16b2ob2ob2o
$4bo6bo14b2ob2ob2o15b2o2b2o$4bo6bo15b2o2b2o14bobo4bobo$6bo2bo15bobo4bo
bo12b2o6b2o$4b2ob2ob2o13b2o6b2o12b2obo2bob2o$5b2o2b2o14b2obo2bob2o16b
2o$3bobo4bobo16b2o$3b2o6b2o$3b2obo2bob2o$7b2o!


Some more rigorous enumeration of all the possible tags and tags on tags for the width 16 2c/6 period 3 parasitics will have to be performed before we can conclusively say that period 6 cases do not occur.

Have a happy day,

-Tim Coe
moebius
 
Posts: 26
Joined: December 10th, 2015, 9:07 am

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby Sokwe » December 22nd, 2015, 3:42 pm

The (3,0)c/8 width-23 gutter search finished with a negative result. Here is the final partial pattern reported:
x = 23, y = 32, rule = B3/S23
2o2bo3bo5bo3bo2b2o$4b2o11b2o$2bobo2b2o5b2o2bobo$2b2obo11bob2o$bo3bo2bo
5bo2bo3bo$2bo2bo2bo5bo2bo2bo$3b2ob3o5b3ob2o$2b2o2b2o7b2o2b2o$b2o3b3o5b
3o3b2o$2b2o2b2o7b2o2b2o$bo3bo3b2ob2o3bo3bo$b2obo2bobo3bobo2bob2o$b2obo
bobo5bobobob2o$2bo3b2o7b2o3bo$2bo3bo9bo3bo$2bo17bo$6b3o5b3o$4b2ob2o5b
2ob2o2$4bo2bobo3bobo2bo$3bo6bobo6bo$6bo3bobo3bo$7b3o3b3o$3bo4bo5bo4bo$
3bo15bo$3b2obo9bob2o$5b2o9b2o$4b3o9b3o$ob3o13b3obo$o3bo13bo3bo$bobo15b
obo$2bo17bo!


wildmyron wrote:Should there also be some indication when a ship is known but not known to be minimal - whilst still keeping the results for negative searches?

I can't think of a way to do this without the table becoming confusing or getting too complicated.

wildmyron wrote:I think we can push this up to known ship at width 18 now

Personally, I'm not convinced that any current search programs are capable of showing this. gfind might be able to, but I haven't explored its duplicate row elimination code enough to convince myself that it does (in fact, I sort of suspect it does not).
-Matthias Merzenich
Sokwe
Moderator
 
Posts: 1255
Joined: July 9th, 2009, 2:44 pm

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby codeholic » December 23rd, 2015, 3:15 am

moebius wrote:I would run searches and produce giant armadas of 2c/4 ships. I would tend to run them then and blow them up. One time the period 16 smoker emerged after trimming the tail of some ship. It was basically total luck.

Thanks for your answer!

Does anyone know if someone tried automated search for large period spaceships and puffers based on this principle? I. e. 1) find a spaceship or a partial; 2) trim the end of the ship/partial row by row and evolve the results.
Ivan Fomichev
User avatar
codeholic
Moderator
 
Posts: 1140
Joined: September 13th, 2011, 8:23 am
Location: Hamburg, Germany

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby Sokwe » December 23rd, 2015, 3:54 am

codeholic wrote:Does anyone know if someone tried automated search for large period spaceships and puffers based on this principle? I. e. 1) find a spaceship or a partial; 2) trim the end of the ship/partial row by row and evolve the results.

Paul Tooke used this technique to find most of the known c/2 and 2c/5 puffer engines. I think he had developed some tools to filter the output to some extent, but I don't remember him posting any of them. Unfortunately, he hasn't been active for 2 years, so you might not get anything more useful than that. I suppose you could try looking through his posts for more information.
-Matthias Merzenich
Sokwe
Moderator
 
Posts: 1255
Joined: July 9th, 2009, 2:44 pm

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby towerator » December 23rd, 2015, 5:42 am

Is that stupid to suggest to "take it from the other side" and trying to find back ends, then progressing in the sense of direction until the front is complete?
This is game of life, this is game of life!
Loafin' ships eaten with a knife!
towerator
 
Posts: 328
Joined: September 2nd, 2013, 3:03 pm

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby Sokwe » December 23rd, 2015, 8:28 am

towerator wrote:Is that stupid to suggest to "take it from the other side" and trying to find back ends, then progressing in the sense of direction until the front is complete?

There is a "reverse search order" mode for gfind, and WLS/JLS can always be set to search from the back to the front of the spaceship. Unfortunately, these searches tend to take much longer than the standard front-to-back searches of the same width. Part of the problem might be that there are more potential back ends than there are potential front ends for most (all?) spaceship speeds.

Edit: The (1,1)c/6 width-10 search finished. It took approximately 11 hours. Regrettably, it didn't get any deeper than the width-9 search, so I have no new partials to report.

I am currently running the following searches:
  • (3,0)c/8 odd-symmetric, width-21
  • (1,0)c/6 odd-symmetric, width-17
  • (1,1)c/7, width-9
-Matthias Merzenich
Sokwe
Moderator
 
Posts: 1255
Joined: July 9th, 2009, 2:44 pm

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby towerator » December 24th, 2015, 5:08 am

Is it possible to download this program somewhere?
This is game of life, this is game of life!
Loafin' ships eaten with a knife!
towerator
 
Posts: 328
Joined: September 2nd, 2013, 3:03 pm

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby Sokwe » December 24th, 2015, 8:17 am

towerator wrote:Is it possible to download this program somewhere?

There are a lot of programs that can search for spaceships. Information about many search programs can be found here. I'm not sure which one you are asking about, so I'll list all of the spaceship search programs I know of that are available:

  • WinLifeSearch (Windows-specific) by Jason Summers; usage is fairly straightforward. Based on Lifesrc. Nicolay Beluchenko made a modified version with additional features, available here (binary only).
  • JavaLifeSearch by Karel Suhajda; Like WinLifeSearch, but programmed in Java and cross-platform. Slower but with more features.
  • gfind by David Eppstein; command-line program to search for thin spaceships. Very fast. Scripts forum discussion here.
  • gfind-pt by Paul Tooke; modifications to gfind. Paul says only use it as a guide to modifying gfind, and don't use it directly, but none of us listen.
  • Knight2 by Tim Coe; only works for certain spaceship speeds, but often faster than gfind. A work in progress, and still buggy. Scripts forum discussion here.
  • lifelocallookahead by Andrew J. Wade. Seems interesting, but I don't know much about it.
  • afind by Eugene Langvagen; Similar to (and somewhat slower than) gfind. Has some additional features. Scripts forum discussion here and here.
  • Lifesrc by David Bell; An old program, and the basis for WLS. I'm not very familiar with it. Requires curses or ncurses library.
  • gsearch by David Eppstein. Unlike the other programs in this list, gsearch tests random patterns to see if they become spaceships. Uses David Eppsteins spaceship database.
-Matthias Merzenich
Sokwe
Moderator
 
Posts: 1255
Joined: July 9th, 2009, 2:44 pm

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby Sokwe » December 25th, 2015, 3:21 am

The (3,0)c/8 odd-symmetric, width-21 search finished without finding a spaceship. Here is the final partial reported:
x = 21, y = 28, rule = B3/S23
obobo3bobobo3bobobo$bo3bobobobobobo3bo$2o3bob2o3b2obo3b2o$7b2obob2o$bo
bo4bobobo4bobo$b2ob2o9b2ob2o$4b3obo3bob3o$bobob3o5b3obobo$3b2o2bo5bo2b
2o$3b2o11b2o$6b3o3b3o$bo6bo3bo6bo$bobo4bo3bo4bobo$2bo4bo5bo4bo$7bo5bo$
b3o2b2o5b2o2b3o$4b2o2bo3bo2b2o$5b4o3b4o$b2o2b4o3b4o2b2o$3bo3bo5bo3bo$
5bo9bo$4b2obo5bob2o$6b2o5b2o$5b3o5b3o$bob3o9b3obo$bo3bo9bo3bo$2bobo11b
obo$3bo13bo!

I didn't keep close track of the time for this search, and I had to restart it once, but I think it took less than three days. It should be reasonably possible to search at least one cell wider for each type of (3,0)c/8, but I don't intend to do it right now.

The (1,0)c/6 search also finished with a negative result. Here is the longest partial:
x = 17, y = 81, rule = B3/S23
b2o3bobobo3b2o$2bo3b5o3bo$6b5o$3b2ob5ob2o$3bo2bobobo2bo$3bo9bo$3bo2bo
3bo2bo$4b9o$8bo$6b2ob2o$3bob2obob2obo$2b2ob2o3b2ob2o$2bo2bo5bo2bo$2bo
2bo5bo2bo$2b2ob2obob2ob2o$2bo2bo2bo2bo2bo$b2ob3o3b3ob2o2$3bo3b3o3bo$4b
o2b3o2bo$2bo2bo2bo2bo2bo$2bo3bo3bo3bo$5b7o$2bo2bob3obo2bo$bobo9bobo$bo
6bo6bo$bobo4bo4bobo$2bo5bo5bo$bo13bo$2b2o9b2o2$3bobo5bobo$6bo3bo$3bob
7obo$3bo3bobo3bo$4bo2b3o2bo$4b2o5b2o$2bo2bo5bo2bo$2bobo7bobo$2b3o7b3o
2$bo3bo5bo3bo$6o5b6o$3bo9bo$4b2o5b2o$o3bo7bo3bo$2bob2o5b2obo$2b2obo5bo
b2o$3bobo5bobo$4bo7bo$3b2o7b2o$2bob3o3b3obo$b2obo2bobo2bob2o$2b2o2b2ob
2o2b2o$5b3ob3o$4bo7bo$2bo3bo3bo3bo$2b2o3bobo3b2o$2b2o2bo3bo2b2o$5bob3o
bo$5bob3obo$5b7o$b2obobo3bobob2o$2bob3o3b3obo$2bo2bobobobo2bo$2b2o3bob
o3b2o$5bobobobo$3bob2o3b2obo$3bo2bo3bo2bo$3bo2bo3bo2bo$5b2obob2o$6b5o
3$5bo5bo$5bo5bo$5b3ob3o$6b2ob2o2$7bobo$8bo!

If someone can use this partial to find a width-19 ship, it could save us the trouble of running gfind at that width.

I am currently running the following searches:
  • (1,0)c/6 asymmetric, width-10
  • (1,1)c/7, width-9
-Matthias Merzenich
Sokwe
Moderator
 
Posts: 1255
Joined: July 9th, 2009, 2:44 pm

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby Sokwe » January 5th, 2016, 4:35 am

Here is a width-19 odd-symmetric (1,0)c/6 spaceship:
x = 19, y = 146, rule = B3/S23
6bo5bo$5bo2bobo2bo$4bo3bobo3bo$4bo3bobo3bo$4bob2o3b2obo$4bo4bo4bo$3bo
4bobo4bo$3bo11bo$5b2obobob2o$9bo$3b2o9b2o$3b3o7b3o$3bo3bo3bo3bo$4bo2bo
3bo2bo$4bo9bo$6b3ob3o$4bobo5bobo$4bo9bo$4bo9bo$6b3ob3o$6bo2bo2bo$5b2ob
3ob2o$4b2o7b2o$4b2o2bobo2b2o$2bo13bo$2bo2b4ob4o2bo$5bo2bobo2bo$bobo2b
2o3b2o2bobo$2bo3bobobobo3bo$7b5o$8b3o$9bo$6b3ob3o$3bo3bo3bo3bo$6bo5bo$
2bob2o7b2obo$2bo3b2o3b2o3bo$7bo3bo$bo5bo3bo5bo$bo3bo7bo3bo$o2bo11bo2bo
$bo4b2o3b2o4bo$2bo4b2ob2o4bo$4b3o5b3o3$2b3o9b3o$2b4o7b4o$2bo3bo5bo3bo$
6bo5bo$2bo13bo$2bo13bo$b2o13b2o$b2o13b2o$b2o2bo7bo2b2o$2b2o11b2o$3bo3b
o3bo3bo$4bo3bobo3bo$4bo3bobo3bo$3b2obobobobob2o$4bo9bo$5b2o5b2o$6b3ob
3o$9bo$7bobobo$8b3o2$7b2ob2o$7b2ob2o$4b2o7b2o$5b2obobob2o$6b3ob3o$6bo
5bo2$4b3o5b3o$4b2o7b2o$3bo3b2ob2o3bo$b4o9b4o$9bo$bo2b2obo3bob2o2bo$7bo
3bo2$3b3o7b3o3$5b3o3b3o$4bob3ob3obo$3bo4bobo4bo$4bobobobobobo$6bobobob
o$6bobobobo$4bobobobobobo$3b2o3bobo3b2o$3b2obobobobob2o2$3bobo7bobo$5b
ob2ob2obo$4b3o5b3o3$6bo5bo$5b2o5b2o$3b4o5b4o$6bo5bo$3bob3o3b3obo$4bo2b
2ob2o2bo$5bobo3bobo$3b3obobobob3o$4bo2b2ob2o2bo$8bobo$2bo13bo$2b2o2bo
5bo2b2o$7b2ob2o$4b3o5b3o$3b2obo5bob2o$2bo13bo$bo2bo3bobo3bo2bo$2bobob
3ob3obobo2$7bo3bo$5b4ob4o$4bo3bobo3bo$3bo11bo$3bo11bo2$3b2o2bo3bo2b2o$
2b3o2bo3bo2b3o$b2o4bo3bo4b2o$6bo5bo$5b2o5b2o$5bobo3bobo$8bobo$5bo2bobo
2bo$5b3o3b3o$4b3o5b3o$2b2obo7bob2o2$bo15bo$2bobo9bobo$4bo9bo$5bo7bo2$
4bob3ob3obo$4bo3bobo3bo$3bo3bo3bo3bo$5bo7bo!

It was found by extending a width-17 partial result at width-19 for a while, and then reducing the width back down to 17 for the tail. There are possibly shorter ships of this type.

I also completed the (1,1)c/7 width-9 and (1,0)c/6 width-10 asymmetric searches with no results.
-Matthias Merzenich
Sokwe
Moderator
 
Posts: 1255
Joined: July 9th, 2009, 2:44 pm

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby HartmutHolzwart » January 5th, 2016, 4:44 am

Congratulations! c/6 ships seem to be pretty rare.

Do you continue your current search to find some more? In a sense this looks like the c/5 case 10 years ago. So if only enough time is devoted to searches, we will eventually find something unexpected that will drive further research.

One of the parts looks like a starting point for a c/6 grey ship. Not that I have hopes to see progress in that direction any time soon... But still!

Kind regards,
Hartmut
HartmutHolzwart
 
Posts: 387
Joined: June 27th, 2009, 10:58 am
Location: Germany

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby Sokwe » January 5th, 2016, 4:51 am

HartmutHolzwart wrote:Do you continue your current search to find some more?

The search that found this ship is still running, as well as a few width-19 searches. I have been working on this for over a week, and this is the first ship that I found.
-Matthias Merzenich
Sokwe
Moderator
 
Posts: 1255
Joined: July 9th, 2009, 2:44 pm

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby Scorbie » January 5th, 2016, 6:07 am

Congrats for the discovery! Reminds me of the Totem Pole (the spaceship) :)
HartmutHolzwart wrote:Do you continue your current search to find some more? In a sense this looks like the c/5 case 10 years ago. So if only enough time is devoted to searches, we will eventually find something unexpected that will drive further research.
Do you mean the Snail? (It sure does look like the Snail(again, the spaceship))
Best wishes to you, Scorbie
User avatar
Scorbie
 
Posts: 1350
Joined: December 7th, 2013, 1:05 am

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby Sokwe » January 15th, 2016, 1:08 am

Here's a partial result found in an attempt to find a "fully p6" 3c/6 spaceship:
x = 60, y = 27, rule = B3/S23
34b2o9bo$32b2o2b2o2b2obo3bo7b2o$26bo2b4o4bobobo6bo8bo$26bo2b3o7b2ob2o
2bobo2bo2bo$4bo2bo17bo6b2obo2b3o10bo2b3o$3b2o2bo3b2o7b2o3b3o2b3o2bobob
o4bo3b2obo2b2o$2bo2b2ob5o6bob2o8bo4b2obobo2bobo3bo3b2o$bo2bo3bo2b2o3b
2obobob3ob5o5b3o4b3o4bo4bobo$2ob2o3b3o5b3o2b2o5b2o9b2o10bobob5o$o3bob
2o6bo3b4o2bo6bo2b2o5bo5b2o3bo$3obob3o3b2o2bob2obob3o5bo10bobo8b2o$9bob
5o5b2o5b2o2bob3o21b2o$b7obob2o2bo2b3o2b5o4bobo18bo2bo$o32bo19bo2bobo$b
7obob2o2bo2b3o2b5o4bobo18bo2bo$9bob5o5b2o5b2o2bob3o21b2o$3obob3o3b2o2b
ob2obob3o5bo10bobo8b2o$o3bob2o6bo3b4o2bo6bo2b2o5bo5b2o3bo$2ob2o3b3o5b
3o2b2o5b2o9b2o10bobob5o$bo2bo3bo2b2o3b2obobob3ob5o5b3o4b3o4bo4bobo$2bo
2b2ob5o6bob2o8bo4b2obobo2bobo3bo3b2o$3b2o2bo3b2o7b2o3b3o2b3o2bobobo4bo
3b2obo2b2o$4bo2bo17bo6b2obo2b3o10bo2b3o$26bo2b3o7b2ob2o2bobo2bo2bo$26b
o2b4o4bobobo6bo8bo$32b2o2b2o2b2obo3bo7b2o$34b2o9bo!

I found it by extending a width-19 partial at width-29. I also searched more thoroughly at width-25 bilaterally symmetric, but the only fully p6 front end was the width-19 in the above partial.

I also searched for a width-15 asymmetric 3c/6 ship with gfind, but the search only reached a depth of 73 (about 12 rows) before finishing. I am currently running a width-16 search.
-Matthias Merzenich
Sokwe
Moderator
 
Posts: 1255
Joined: July 9th, 2009, 2:44 pm

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby wildmyron » January 15th, 2016, 2:29 am

moebius wrote:Wildmyron,

The problem with the width 16 2c/6 searches is they always miss stuff. I ran a bunch of different searches trying to cover the search tree, but I could never be sure I wasn't missing something. For example, I was looking through my turtle armadas with their variety of tags I never ran across the 9th one in your pictures. Your list of ships does not include the following ones, or anything very close:

x = 60, y = 39, rule = B3/S23
6bo2bo18bo2bo$4bobo2bobo14bobo2bobo14bobo2bobo$5bo4bo16bo4bo15b2o4b2o$
5bo4bo16bo4bo17bo2bo2$5b2o2b2o16b2o2b2o15b2o4b2o2$4bobo2bobo14bobo2bob
o14bo6bo$bobo2bo2bo2bobo8bobo2bo2bo2bobo$2obob2o2b2obob2o6b2obob2o2b2o
bob2o9b3o4b3o$2obobo4bobob2o6b2obobo4bobob2o11bo4bo$3bobo4bobo12bobo4b
obo11b2obo4bob2o$2b2obob2obob2o10b2obob2obob2o10bo10bo$3bo2bo2bo2bo12b
o2bo2bo2bo11bo3bo2bo3bo$3bo2bo2bo2bo12bo2bo2bo2bo13b2o4b2o$4b2o4b2o14b
2o4b2o12bobo6bobo$44b3o10b3o$4b2o4b2o14b2o4b2o11b2o10b2o$3bo8bo10bo4bo
2bo4bo9bobo6bobo$5bo4bo11bobob3o2b3obobo8bo2bo4bo2bo$2bo2bo4bo2bo9b3o
2b4o2b3o10b2o6b2o$bo3bob2obo3bo13bo2bo$2b5o2b5o15b2o16b2o6b2o$3bob2o2b
2obo36bob2obo$6bo2bo15b2o6b2o10bo2b3o2b3o2bo$26b8o10bo5bo2bo5bo$45b3ob
2o2b2ob3o$3b2o6b2o13bo6bo15bo4bo$4b8o14bo6bo16bo2bo$28bo2bo16b2ob2ob2o
$4bo6bo14b2ob2ob2o15b2o2b2o$4bo6bo15b2o2b2o14bobo4bobo$6bo2bo15bobo4bo
bo12b2o6b2o$4b2ob2ob2o13b2o6b2o12b2obo2bob2o$5b2o2b2o14b2obo2bob2o16b
2o$3bobo4bobo16b2o$3b2o6b2o$3b2obo2bob2o$7b2o!


Some more rigorous enumeration of all the possible tags and tags on tags for the width 16 2c/6 period 3 parasitics will have to be performed before we can conclusively say that period 6 cases do not occur.

Have a happy day,

-Tim Coe


My apologies for not responding earlier, and for prematurely making an unjustified claim. Thank you for posting counterexamples which I believe highlight the deficiency in gfind. I note that the first turtle tag from the first ship you showed, on its own, is also missing from my results:

x = 14, y = 22, rule = B3/S23
3b2o4b2o$2bo8bo$4bo4bo$bo2bo4bo2bo$o3bob2obo3bo$b5o2b5o$2bob2o2b2obo$
5bo2bo3$2b2o6b2o$3b8o2$3bo6bo$3bo6bo$5bo2bo$3b2ob2ob2o$4b2o2b2o$2bobo
4bobo$2b2o6b2o$2b2obo2bob2o$6b2o!


Sokwe wrote:
wildmyron wrote:I think we can push this up to known ship at width 18 now

Personally, I'm not convinced that any current search programs are capable of showing this. gfind might be able to, but I haven't explored its duplicate row elimination code enough to convince myself that it does (in fact, I sort of suspect it does not).

Indeed, The counterexamples above show that there are indeed valid search results which are lost due to the duplicate elimination, even with hashing disabled. I think it is illustrative that all cases are tagalongs where there is one row between the ship and tagalong which is empty for all phases of one period (p3 in this case).

I have previously considered an alteration to wls/jls which may allow a complete enumeration of all solutions and partials up to a given length within a practical timeframe. Minimal details here. Essentially it would prevent jls from exploring the combinatoric realm of non-interacting solutions, although as described it wouldn't reject all of them.

wildmyron wrote:
wildmyron wrote:I'm currently running 'gfind o8n1vl112' just to have a look at the even symmetry partials as well. If no one else is running them yet, I'll run the l128 searches over the next week or two.

No surprises there - the search was unsuccessful. Took about 10 hours. I'm running the odd and even symmetry searches at l128 now.

Update:
(1,0)c/8 even-symmetric width 16 - still ongoing, mainly due to the machine it's running on being switched off over the holidays.
(1,0)c/8 odd-symmetric width 15 - I didn't really get far on this search before suspending it, if someone else wants to run it, feel free.
wildmyron
 
Posts: 790
Joined: August 9th, 2013, 12:45 am

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby Sokwe » January 15th, 2016, 5:33 am

wildmyron wrote:I have previously considered an alteration to wls/jls which may allow a complete enumeration of all solutions and partials up to a given length within a practical timeframe. Minimal details here. Essentially it would prevent jls from exploring the combinatoric realm of non-interacting solutions, although as described it wouldn't reject all of them.

I think this is a property of Nicolay Beluchenko's modification of WLS (assuming I understand you correctly). The modified WLS is available here.
-Matthias Merzenich
Sokwe
Moderator
 
Posts: 1255
Joined: July 9th, 2009, 2:44 pm

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby velcrorex » January 15th, 2016, 6:11 pm

Update:
(1,0)c/8 even-symmetric width 16 - still ongoing, mainly due to the machine it's running on being switched off over the holidays.


I completed that search with partial here. Though it certainly wouldn't be bad to have someone corroborate the search.
-Josh Ball.
User avatar
velcrorex
 
Posts: 339
Joined: November 1st, 2009, 1:33 pm

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby Sokwe » January 15th, 2016, 10:30 pm

velcrorex wrote:I completed that search with partial here. Though it certainly wouldn't be bad to have someone corroborate the search.

I had forgotten about that. Are there any other searches you have done that are not included in the table? Also, was this search done directly with Paul Tooke's modified gfind (I presume so since you posted a partial result)?
-Matthias Merzenich
Sokwe
Moderator
 
Posts: 1255
Joined: July 9th, 2009, 2:44 pm

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby wildmyron » January 16th, 2016, 6:22 am

Sokwe wrote:
wildmyron wrote:I have previously considered an alteration to wls/jls which may allow a complete enumeration of all solutions and partials up to a given length within a practical timeframe. Minimal details here. Essentially it would prevent jls from exploring the combinatoric realm of non-interacting solutions, although as described it wouldn't reject all of them.

I think this is a property of Nicolay Beluchenko's modification of WLS (assuming I understand you correctly). The modified WLS is available here.

Wow, that is very much like what I was thinking of. I recall reading of it being used for excluding gliders from c/4 searches but I didn't realise that that feature was generalised to all searches. I'm still trying to work out what some of the options do and precisely how the search space is pruned, but I'm running a 2c/6 search at width 16 with it now to see what comes out.

velcrorex wrote:I completed that search with partial here. Though it certainly wouldn't be bad to have someone corroborate the search.

Like Sokwe I'd forgotten about that thread. Is it reasonable to presume you haven't run the odd-symmetric search at the same level (width 15)?

I'm not sure that using the same code is the best corroboration but I'll let the search run to completion anyway.
wildmyron
 
Posts: 790
Joined: August 9th, 2013, 12:45 am

Re: 3c/7 othogonal and 2c/9 diagonal spaceships

Postby velcrorex » January 16th, 2016, 8:27 am

Like Sokwe I'd forgotten about that thread. Is it reasonable to presume you haven't run the odd-symmetric search at the same level (width 15)?

I'm not sure that using the same code is the best corroboration but I'll let the search run to completion anyway.


I have not done the odd symmetric search. I used a version of gfind which included Paul Tooke's edits as well as some tinkering on my own, so it wouldn't be the exact same code.
-Josh Ball.
User avatar
velcrorex
 
Posts: 339
Joined: November 1st, 2009, 1:33 pm

PreviousNext

Return to Patterns

Who is online

Users browsing this forum: gmc_nxtman, vyznev and 7 guests