Online Database of Unsuccessful GFIND Searches

For scripts to aid with computation or simulation in cellular automata.
User avatar
ad_ca
Posts: 16
Joined: January 28th, 2010, 5:35 am
Contact:

Online Database of Unsuccessful GFIND Searches

Post by ad_ca » February 11th, 2010, 4:59 am

Well, I don't actually have one, but really would like to see something like this; probably here.

:idea: 'a negative result is still a result' :idea:

One of the easiest (lol not really) ways to become famous in the Game of Life is to have the 'gfind' program (found here) running somewhere in the background searching for new spaceship velocities like c/7, 3c/8 etc. But no one actually wants to repeat the searches others have already accomplished, so some sort of collection of such negative results is definitely needed.

For example, I remember a letter by Paul Tooke saying:
At 3c/7 I completed asymmetric up to width 11, glide symmetric up to widths
11&12, bilaterally symmetric up to widths 22 (even), 23 (odd) and 25 (odd
with a central 'gutter' of OFF cells).
...
At 3c/8 and 4c/9 I completed the three bilaterally symmetric modes up to
widths 17, 18 & 19.
(this corresponds to

Code: Select all

gfind b3s23/o7n3agv/l154
and

Code: Select all

gfind b3s23/o7n3uw/l168
)

But it was a long time ago;
What is the current status of these searches?

Sokwe
Moderator
Posts: 2645
Joined: July 9th, 2009, 2:44 pm

Re: Online Database of Unsuccessful GFIND Searches

Post by Sokwe » February 11th, 2010, 9:00 pm

This would be a good thing to keep track of, as I have often wondered what searches have already been completed. I wouldn't think such a database would need to be limited to unsuccessful searches. For example, one could find all of the bilaterally-symmetric, period-4, c/4 orthogonal spaceships with a width of 13 (here are the shortest ones; I haven't yet gone through all of them):

Code: Select all

x = 192, y = 70, rule = B3/S23
180b2o7b2o2$180bo9bo$179b2o9b2o2$165bo13b2o9b2o$165bo16bo5bo$165bo13bo
2bo5bo2bo$183bo3bo$163b2ob2o15bo3bo$163b2ob2o13bo7bo$182bobobobo$143b
2ob2o15b2ob2o15b2ob2o$142bo5bo13bo5bo15bobo$144bobo17bobo17bobo$141bo
2bobo2bo11bo2bobo2bo11b2obobob2o$144bobo17bobo13bobobobobobo$141b2obob
ob2o11b2obobob2o10b3obobob3o$142bobobobo13bobobobo11b2o2bobo2b2o$142b
3ob3o13b3ob3o13b3ob3o$123b2ob2o13b2o5b2o11b2o5b2o11b2o5b2o$123b2ob2o
12bobo2bo2bobo9bobo2bo2bobo9bobo2bo2bobo$120b2o2bobo2b2o12b2ob2o15b2ob
2o15b2ob2o$121bo2bobo2bo10bo9bo9bo9bo9bo9bo$124bobo14b9o11b9o11b9o$
121bob2ob2obo$120b5ob5o14bo19bo19bo$85bo34b2o7b2o8b2o4bo4b2o7b2o4bo4b
2o7b2o4bo4b2o$85bo13b3o7b3o9b2obobob2o15bo19bo19bo$103bo3bo16bobo13bo
9bo9bo9bo9bo9bo$83b2ob2o11bo4bobo4bo12bobo13b2o7b2o9b2o7b2o9b2o7b2o$
100bo9bo9bob2o3b2obo12b5o15b5o15b5o$63bo3bo15b2ob2o13b2obobob2o31bob5o
bo11bob5obo11bob5obo$64bobo14b3o3b3o13bo3bo12bo9bo14bo19bo19bo$63bo3bo
11b4o5b4o10b2o3b2o11b2o2bobo2b2o11b2o3b2o13b2o3b2o13b2o3b2o$40b2o7b2o
12b2ob2o11bo11bo11bo3bo12b3obobob3o12bo3bo15bo3bo15bo3bo$61b2obobob2o
10b2obo3bob2o10bobo3bobo31bobo3bobo11bobo3bobo11bobo3bobo$40bo9bo10b2o
bobob2o12bo2bo2bo12b2o5b2o12b3ob3o12b2o5b2o11b2o5b2o11b2o5b2o$39b2o9b
2o12bobo16bobobo11b3o7b3o10b2o3b2o10b3o7b3o7b3o7b3o7b3o7b3o$64bobo18bo
13b2o9b2o27b2o9b2o7b2o9b2o7b2o9b2o$25bo13b2o9b2o8bo3bobo3bo9bobo5bobo
14bo19bo19bo19bo19bo$24b3o15bo5bo12b2obobob2o12bo5bo15b3o17b3o17b3o17b
3o17b3o$20b4obob4o8bo2bo5bo2bo12bobo16bo3bo12b4obob4o9b4obob4o9b4obob
4o9b4obob4o9b4obob4o$20b2ob2ob2ob2o8bo3bo3bo3bo11b2ob2o14bo5bo11b2ob2o
b2ob2o9b2ob2ob2ob2o9b2ob2ob2ob2o9b2ob2ob2ob2o9b2ob2ob2ob2o$21bo2bobo2b
o13bo3bo16bobo12b3o7b3o9bo2bobo2bo11bo2bobo2bo11bo2bobo2bo11bo2bobo2bo
11bo2bobo2bo$24bobo13bo2bo3bo2bo11bobobobo35bobo17bobo17bobo17bobo17bo
bo$21bo2bobo2bo11bo2bobo2bo11bo2bobo2bo13bo3bo13bo2bobo2bo11bo2bobo2bo
11bo2bobo2bo11bo2bobo2bo11bo2bobo2bo$22b2o3b2o13b2o3b2o13b2o3b2o13b2o
3b2o13b2o3b2o13b2o3b2o13b2o3b2o13b2o3b2o13b2o3b2o2$2b2o3b2o13b2o3b2o
13b2o3b2o13b2o3b2o13b2o3b2o13b2o3b2o13b2o3b2o13b2o3b2o13b2o3b2o13b2o3b
2o$2b2o3b2o13b2o3b2o13b2o3b2o13b2o3b2o13b2o3b2o13b2o3b2o13b2o3b2o13b2o
3b2o13b2o3b2o13b2o3b2o$3bo3bo15bo3bo15bo3bo15bo3bo15bo3bo15bo3bo15bo3b
o15bo3bo15bo3bo15bo3bo$3b2ob2o15b2ob2o15b2ob2o15b2ob2o15b2ob2o15b2ob2o
15b2ob2o15b2ob2o15b2ob2o15b2ob2o2$2bobobobo13bobobobo13bobobobo13bobob
obo13bobobobo13bobobobo13bobobobo13bobobobo13bobobobo13bobobobo$bo2bob
o2bo11bo2bobo2bo11bo2bobo2bo11bo2bobo2bo11bo2bobo2bo11bo2bobo2bo11bo2b
obo2bo11bo2bobo2bo11bo2bobo2bo11bo2bobo2bo$2ob5ob2o9b2ob5ob2o9b2ob5ob
2o9b2ob5ob2o9b2ob5ob2o9b2ob5ob2o9b2ob5ob2o9b2ob5ob2o9b2ob5ob2o9b2ob5ob
2o$bo7bo11bo7bo11bo7bo11bo7bo11bo7bo11bo7bo11bo7bo11bo7bo11bo7bo11bo7b
o$o9bo9bo9bo9bo9bo9bo9bo9bo9bo9bo9bo9bo9bo9bo9bo9bo9bo9bo9bo$2o7b2o9b
2o7b2o9b2o7b2o9b2o7b2o9b2o7b2o9b2o7b2o9b2o7b2o9b2o7b2o9b2o7b2o9b2o7b2o
2$bo2b3o2bo11bo2b3o2bo11bo2b3o2bo11bo2b3o2bo11bo2b3o2bo11bo2b3o2bo11bo
2b3o2bo11bo2b3o2bo11bo2b3o2bo11bo2b3o2bo$bobo3bobo11bobo3bobo11bobo3bo
bo11bobo3bobo11bobo3bobo11bobo3bobo11bobo3bobo11bobo3bobo11bobo3bobo
11bobo3bobo$bobo3bobo11bobo3bobo11bobo3bobo11bobo3bobo11bobo3bobo11bob
o3bobo11bobo3bobo11bobo3bobo11bobo3bobo11bobo3bobo$4bobo17bobo17bobo
17bobo17bobo17bobo17bobo17bobo17bobo17bobo$2bobobobo13bobobobo13bobobo
bo13bobobobo13bobobobo13bobobobo13bobobobo13bobobobo13bobobobo13bobobo
bo2$2bobobobo13bobobobo13bobobobo13bobobobo13bobobobo13bobobobo13bobob
obo13bobobobo13bobobobo13bobobobo$3bo3bo15bo3bo15bo3bo15bo3bo15bo3bo
15bo3bo15bo3bo15bo3bo15bo3bo15bo3bo$3bo3bo15bo3bo15bo3bo15bo3bo15bo3bo
15bo3bo15bo3bo15bo3bo15bo3bo15bo3bo!
This wouldn't help in terms of finding new spaceship speeds, but would be useful in enumerating spaceships within certain parameters, or finding small spaceships.
ad_ca wrote:What is the current status of these searches?
I think I completed unsuccessful searches for bilaterally-symmetric 3c/7 spaceships of width 25 and 26 a couple of years ago, but I'm not entirely sure. I suppose they would be worth checking again (unless someone can verify that those widths have been searched). It would be nice if others would post parameters for their unsuccessful searches.

On a related note, has anyone written a script for post-processing of a gfind search to eliminate duplicate spaceships and output everything in a single file?
-Matthias Merzenich

User avatar
Nathaniel
Site Admin
Posts: 861
Joined: December 10th, 2008, 3:48 pm
Location: New Brunswick, Canada
Contact:

Re: Online Database of Unsuccessful GFIND Searches

Post by Nathaniel » February 11th, 2010, 11:39 pm

This actually would be a nice distributed computing-type project. Have gfind periodically upload results and partial results to a central online database, that somehow catalogs and keeps track of what has and hasn't been completed. I haven't looked at the gfind source code too much, but it seems like a reasonable enough task.

Before I take on something like that that I want to finally fix a couple other things around the site and finish uploading LifeLine scans to the LifeWiki. Has anyone here written C scripts that interact with web servers before? I'm fine and dandy with the web server end of things, but my C knowledge is still reasonably amateur.

Alegend
Posts: 31
Joined: October 5th, 2009, 3:20 pm

Re: Online Database of Unsuccessful GFIND Searches

Post by Alegend » February 13th, 2010, 12:51 pm

Well, I think that we should give gfind a GUI, for MUCH easier use. :mrgreen:

I'm good with console C++, by the way! Okay, I'm not THAT good.

User avatar
ad_ca
Posts: 16
Joined: January 28th, 2010, 5:35 am
Contact:

Re: Online Database of Unsuccessful GFIND Searches

Post by ad_ca » February 13th, 2010, 2:06 pm

gfind is written in pure C, it's cross-platform and its command line is very simple. I don't think a GUI for gfind would actually be useful, unless it would be shipped with some extra features like initial row setting similar to that of ofind (and I do actually want a GUI for ofind, though it's a much more limited program than, for example, WLS).

Yeah, distributed computing might be good (especially because of its BFS nature and also a need for a lot of RAM), but I'm not capable of doing this either.
On a related note, has anyone written a script for post-processing of a gfind search to eliminate duplicate spaceships and output everything in a single file?
I *have* seen it somewhere and even used it, but I don't remember where to find it. I'll try to find it out soon. :oops:

Alegend
Posts: 31
Joined: October 5th, 2009, 3:20 pm

Re: Online Database of Unsuccessful GFIND Searches

Post by Alegend » February 13th, 2010, 2:57 pm

Well, I can't really use the command-line version. LOL

User avatar
ad_ca
Posts: 16
Joined: January 28th, 2010, 5:35 am
Contact:

Re: Online Database of Unsuccessful GFIND Searches

Post by ad_ca » February 13th, 2010, 3:13 pm

Why? What's wrong with it?

Sokwe
Moderator
Posts: 2645
Joined: July 9th, 2009, 2:44 pm

Re: Online Database of Unsuccessful GFIND Searches

Post by Sokwe » February 13th, 2010, 11:53 pm

Nathaniel wrote:This actually would be a nice distributed computing-type project.
Indeed, a distributed effort might actually be able to find something in a reasonable amount of time, provided that it could be split so that multiple branches of the same search could be run simultaneously on different computers (I haven't looked at gfind's algorithm much, so I am not sure how this would be done efficiently). Also, does anyone know how gfind handles repeatable tagalongs such as the following c/4 tagalong?

Code: Select all

x = 13, y = 83, rule = B3/S23
3b2o2$4bo$4b2o2$4b2o$7bo$4bo2bo$4bo3bo$8bo$8bo$8b3o$5b3o$12bo$10bobo$
7b2o$7bo2bo$7bo2bo$10bo2$10b2o2$10bo$9b2o2$9b2o$7bo$7bo2bo$6bo3bo$6bo$
6bo$4b3o$7b3o$2bo$2bobo$6b2o$4bo2bo$4bo2bo$4bo2$3b2o2$4bo$4b2o2$4b2o$
7bo$4bo2bo$4bo3bo$8bo$8bo$8b3o$5b3o$12bo$10bobo$7b2o$7bo2bo$7bo2bo$10b
o2$10b2o2$10bo$9b2o2$9b2o$7bo$7bo2bo$6bo3bo$7bo$6b2o3bo$8b2obo$8bo$2b
2ob2o$5b2obo$bo3bo2b3o$bo3bobobo$5bob2o$2b2obo2$3o4bobo$3o5bo$bo6bo!
Alegend wrote:I can't really use the command-line version.
After compiling the program, all you need to do is run it from a command-line interpreter with the desired parameters. If you want to look up the parameters that gfind takes, run gfind with "c" as the only parameter. At what point are you having trouble?
-Matthias Merzenich

Alegend
Posts: 31
Joined: October 5th, 2009, 3:20 pm

Re: Online Database of Unsuccessful GFIND Searches

Post by Alegend » February 14th, 2010, 1:31 pm

Well, it's just, I don't like the command line, and plus, I didn't start it from the command prompt.

User avatar
ad_ca
Posts: 16
Joined: January 28th, 2010, 5:35 am
Contact:

Re: Online Database of Unsuccessful GFIND Searches

Post by ad_ca » February 15th, 2010, 2:49 pm

Well, it's just, I don't like the command line
A good command-line tool is much better than a bad GUI :roll:
Sorry for offtopic. I guess I should just write some how-to on gfind searching.

I'm running a 4c/9 search right now at l180 (i.e. for asymmetrical ships at width 10, and symmetrical 19-20-21), and it doesn't seem to be too difficult (though it's most probably unsuccessful).

User avatar
velcrorex
Posts: 339
Joined: November 1st, 2009, 1:33 pm

Re: Online Database of Unsuccessful GFIND Searches

Post by velcrorex » March 2nd, 2010, 10:11 pm

This would be a fantastic idea.

As for the 3/7c ship searches mentioned in the first post, I've completed 14 and 15 glide reflect and 24 even with wls with no results.
-Josh Ball.

Paul Tooke
Posts: 111
Joined: May 19th, 2010, 7:35 am
Location: Cambridge, UK

Re: Online Database of Unsuccessful GFIND Searches

Post by Paul Tooke » May 19th, 2010, 10:08 am

To add to the 3c/7 results posted earlier I have since completed friutless gfind searches with o7n3l182uvw and o7n3l196u. That last search took nearly a year to complete, so I won't be banging my head against this particular wall any time soon!

For the benefit of folks unfamiliar with gfinds options these correspond to width 26 even, 27 odd and 27 odd with a central column of dead cells. That last search would also have found any width 13 asymmetric spaceships.

Alegend
Posts: 31
Joined: October 5th, 2009, 3:20 pm

Re: Online Database of Unsuccessful GFIND Searches

Post by Alegend » July 3rd, 2012, 4:13 pm

n1o9l72a gives me nothing. Increasing the level to 90 is also fruitless, but it takes much longer. I'll try it with up to width 7.

thunk
Posts: 170
Joined: October 3rd, 2015, 8:50 pm
Location: Central USA

Re: Online Database of Unsuccessful GFIND Searches

Post by thunk » March 9th, 2016, 1:41 am

In light of the recent discovery of the copperhead, I've decided to do a broad search in gfind (and zfind) for spaceships of high period. Here are my negative results so far (that aren't already reported on codeholic's page).

The text below will be updated occasionally as searches are conducted.

Negative results:
--(1,1)c/10 asymmetric, diagonal width 6. (g)
--(1,0)c/7 even, width 16; odd, width 15. (z)
--(1,0)c/10 odd, width 11. (z)
--(1,0)c/9 even, width 14. (z)

In progress:
--(1,0)c/7 width 17 gutter. (g)
--(1,0)c/8 width 15 odd. (z)
--(1,0)c/9 width 13 odd. (z)
--(3,0)c/11 width 13 odd, 14 even. (z)
Last edited by thunk on March 11th, 2016, 1:37 am, edited 8 times in total.
"What's purple and commutes?
The Evanston Express."

Sokwe
Moderator
Posts: 2645
Joined: July 9th, 2009, 2:44 pm

Re: Online Database of Unsuccessful GFIND Searches

Post by Sokwe » March 9th, 2016, 2:13 am

thunk wrote:Here are my negative results so far (that aren't already reported on codeholic's page)...

--(1,1)c/10 asymmetric, width 6.
I want to point out that a standard gfind search for a diagonal spaceship will use a "diagonal width", while the spaceship search status page uses an "orthogonal width". To see the difference, here are patterns of diagonal width 10 and orthogonal width 10 respectively:

Code: Select all

x = 40, y = 20, rule = B3/S23
4bo25b10o$3b3o24b10o$2b5o23b10o$b7o22b10o$9o21b10o$10o20b10o$b10o19b
10o$2b10o18b10o$3b10o17b10o$4b10o16b10o$5b10o15b10o$6b10o14b10o$7b10o
13b10o$8b10o12b10o$9b10o11b10o$10b9o11b10o$11b7o12b10o$12b5o13b10o$13b
3o14b10o$14bo15b10o!
Also, for orthogonal ships, I would avoid searching for asymmetric ships directly. Instead, search for gutter-symmetric ships at the same search width. This should find any asymmetric ships of that search width, while also potentially finding bilaterally-symmetric ships.

For diagonal searches, the gutter search won't cover the asymmetric case, so it might still be worthwhile to do the asymmetric search.

I am currently running the following searches:
  • (2,0)c/10 gutter-symmetric width-13
  • (2,0)c/10 even glide-symmetric width-12
  • (4,0)c/10 even bilaterally-symmetric width-16
-Matthias Merzenich

thunk
Posts: 170
Joined: October 3rd, 2015, 8:50 pm
Location: Central USA

Re: Online Database of Unsuccessful GFIND Searches

Post by thunk » March 9th, 2016, 3:16 am

Sokwe wrote: I want to point out that a standard gfind search for a diagonal spaceship will use a "diagonal width", while the spaceship search status page uses an "orthogonal width". To see the difference, here are patterns of diagonal width 10 and orthogonal width 10 respectively:

Also, for orthogonal ships, I would avoid searching for asymmetric ships directly. Instead, search for gutter-symmetric ships at the same search width. This should find any asymmetric ships of that search width, while also potentially finding bilaterally-symmetric ships.

For diagonal searches, the gutter search won't cover the asymmetric case, so it might still be worthwhile to do the asymmetric search.
Ah, thank you.
"What's purple and commutes?
The Evanston Express."

thunk
Posts: 170
Joined: October 3rd, 2015, 8:50 pm
Location: Central USA

Re: Online Database of Unsuccessful GFIND Searches

Post by thunk » March 10th, 2016, 4:31 pm

zdr's "zfind" (what I'm calling it) does indeed seem to be faster at high periods than gfind, finishing the width 16 even c/7 search in less than 3 hours (gfind ran for about two days, and was far from complete). Here is the longest partial reported:

Code: Select all

x = 16, y = 46, rule = B3/S23
2bo10bo$bobo8bobo$bo2bo6bo2bo$2b2o8b2o2$bo12bo$2bo10bo$b2o3b4o3b2o$bo
3b6o3bo2$5b2o2b2o$5b2o2b2o$5b2o2b2o$5b2o2b2o$5bob2obo$5bo4bo$4b2o4b2o$
5b2o2b2o$6bo2bo3$6b4o$6b4o$6b4o$4bo6bo$4bo6bo$2b2obo4bob2o$b2o4b2o4b2o
$b2obo6bob2o$4bo6bo$2b12o$bobob2o2b2obobo$3bo8bo$3b2obo2bob2o$2bo4b2o
4bo$5bo4bo$5bob2obo$3bo8bo$2bobob4obobo2$2b2ob6ob2o$5bob2obo$b2ob2ob2o
b2ob2o$b2o3bo2bo3b2o$2o2bo6bo2b2o$obob2ob2ob2obobo!
EDIT: Here's a rather long partial from the width 15 odd search (in progress). There's probably a double loafer around here somewhere.

Code: Select all

x = 16, y = 61, rule = B3/S23
4bo5bo$3bobo3bobo$2bo2bo3bo2bo$3b2o5b2o2$5bo3bo$3b4ob4o2$3bobo3bobo$6b
obo$3b2o5b2o$4bo5bo$5bo3bo$5bo3bo$6b3o$5bo3bo$4bo5bo$4bobobobo$4b3ob3o
$5bo3bo$5b5o$5bo3bo$4bobobobo$4bobobobo$3b4ob4o$2bo3bobo3bo$3bo2bobo2b
o$6bobo$5b2ob2o$2b2ob2ob2ob2o$2b2o7b2o2$bobo7bobo$bo2bo5bo2bo$bo2b2o3b
2o2bo$4b3ob3o$4b2obob2o$6b3o2$6bobo$5bo3bo$7bo$5bobobo$b2o2b2ob2o2b2o$
3bo7bo$o2b2obobob2o2bo$obobo5bobobo$4b2o3b2o$2bo4bo4bo$2b2ob5ob2o$3bo
7bo$3bobobobobo$b3ob5ob3o$b2o4bo4b2o$b2o9b2o$b3o7b3o$2o2b2o3b2o2b2o$2b
obo5bobo2$2o2b2o3b2o2b2o$2bo9bo!
Last edited by thunk on March 10th, 2016, 9:10 pm, edited 2 times in total.
"What's purple and commutes?
The Evanston Express."

User avatar
muzik
Posts: 5614
Joined: January 28th, 2016, 2:47 pm
Location: Scotland

Re: Online Database of Unsuccessful GFIND Searches

Post by muzik » March 10th, 2016, 4:37 pm

thunk wrote:

Code: Select all

x = 16, y = 46, rule = B3/S23
2bo10bo$bobo8bobo$bo2bo6bo2bo$2b2o8b2o2$bo12bo$2bo10bo$b2o3b4o3b2o$bo
3b6o3bo2$5b2o2b2o$5b2o2b2o$5b2o2b2o$5b2o2b2o$5bob2obo$5bo4bo$4b2o4b2o$
5b2o2b2o$6bo2bo3$6b4o$6b4o$6b4o$4bo6bo$4bo6bo$2b2obo4bob2o$b2o4b2o4b2o
$b2obo6bob2o$4bo6bo$2b12o$bobob2o2b2obobo$3bo8bo$3b2obo2bob2o$2bo4b2o
4bo$5bo4bo$5bob2obo$3bo8bo$2bobob4obobo2$2b2ob6ob2o$5bob2obo$b2ob2ob2o
b2ob2o$b2o3bo2bo3b2o$2o2bo6bo2b2o$obob2ob2ob2obobo!
You've probably noticed this already but the two loaves at the top of the ship are being perturbed by the same "ladle spark" (best name!) as in the normal loafer. This makes me think that this could actually be developed into a tagalong for the loafer.

drc
Posts: 1664
Joined: December 3rd, 2015, 4:11 pm

Re: Online Database of Unsuccessful GFIND Searches

Post by drc » March 10th, 2016, 4:41 pm

muzik wrote: You've probably noticed this already but the two loaves at the top of the ship are being perturbed by the same "ladle spark" (best name!) as in the normal loafer. This makes me think that this could actually be developed into a tagalong for the loafer.
Considering it doesn't have a name yet, that's a good name for it (maybe dipper_?)

User avatar
muzik
Posts: 5614
Joined: January 28th, 2016, 2:47 pm
Location: Scotland

Re: Online Database of Unsuccessful GFIND Searches

Post by muzik » March 10th, 2016, 4:55 pm

drc wrote:Considering it doesn't have a name yet, that's a good name for it (maybe dipper_?)
The only thing is, it's a parent of this common spark

Code: Select all

x = 3, y = 2, rule = B3/S23
b2o$3o!

Sokwe
Moderator
Posts: 2645
Joined: July 9th, 2009, 2:44 pm

Re: Online Database of Unsuccessful GFIND Searches

Post by Sokwe » March 10th, 2016, 6:04 pm

thunk wrote:zdr's "zfind" (what I'm calling it) does indeed seem to be faster at high periods than gfind, finishing the width 16 even c/7 search in less than 3/4 hours (gfind ran for about two days, and was far from complete).
zfind seems to stop after a certain amount of time based on the "l" parameter. As a result, the it won't be able to eliminate all c/7 width-16 ships unless the "l" parameter is set high enough.

Look at the third-to-last line in the output. If it says "end" then I think the search is complete, but if it says "done", the search only stopped, because the "l" parameter was set too low.

I'm still trying to work my way through the code. It's difficult, because there are no comments and all temporary variables are stored in an array instead of being given unique names.
-Matthias Merzenich

thunk
Posts: 170
Joined: October 3rd, 2015, 8:50 pm
Location: Central USA

Re: Online Database of Unsuccessful GFIND Searches

Post by thunk » March 10th, 2016, 6:07 pm

Sokwe wrote:zfind seems to stop after a certain amount of time based on the "l" parameter. As a result, the it won't be able to eliminate all c/7 width-16 ships unless the "l" parameter is set high enough.

Look at the third-to-last line in the output. If it says "end" then I think the search is complete, but if it says "done", the search only stopped, because the "l" parameter was set too low.

I'm still trying to work my way through the code. It's difficult, because there are no comments and all temporary variables are stored in an array instead of being given unique names.
The c/7 w16 and c/10 w11 runs both finished with "end". "done" also prints if there is a ship.
"What's purple and commutes?
The Evanston Express."

Sokwe
Moderator
Posts: 2645
Joined: July 9th, 2009, 2:44 pm

Re: Online Database of Unsuccessful GFIND Searches

Post by Sokwe » March 10th, 2016, 6:16 pm

thunk wrote:The c/7 w16 and c/10 w11 runs both finished with "end". "done" also prints if there is a ship.
Thanks.

I think the "l" parameter is p*(the maximum length of the spaceship), where p is the period.
-Matthias Merzenich

User avatar
praosylen
Posts: 2443
Joined: September 13th, 2014, 5:36 pm
Location: Pembina University, Home of the Gliders
Contact:

Re: Online Database of Unsuccessful GFIND Searches

Post by praosylen » March 10th, 2016, 8:31 pm

Just ran an unsuccessful zfind search on 2c/11 width-12 even-symmetric (it outputted "end", so I think that's definite). Took approximately 45 minutes; here are two partials that looked reasonably good:

Code: Select all

x = 38, y = 20, rule = B3/S23
5b2o21bo6bo$28bo6bo$bo8bo16b3o4b3o$bo8bo$bo3b2o3bo$3bo4bo$2b3o2b3o$2bo
bo2bobo17b3o4b3o$3bo4bo20bo4bo$bo2bo2bo2bo16b2o6b2o$2o8b2o15b2o6b2o$2o
bo4bob2o$bobo4bobo17bo6bo$b3o4b3o18bo4bo$2bo6bo20bo2bo$29bo4bo$4b4o19b
o2bo2bo2bo$26b3ob4ob3o$28bob4obo$26b3ob4ob3o!
Edit: Same negative result with 3c/11, but there's a very promising front end (two of them) on this partial:

Code: Select all

x = 12, y = 24, rule = B3/S23
2bo6bo$2bo6bo$bobo4bobo4$b3o4b3o3$b3o4b3o$b3o4b3o$2ob2o2b2ob2o3$2obo4b
ob2o$bo3b2o3bo$2b2o4b2o$bo2b4o2bo$2bo6bo$3b2o2b2o$2b2o4b2o$obo2b2o2bob
o$ob3o2b3obo$b2o6b2o!
former username: A for Awesome
praosylen#5847 (Discord)

The only decision I made was made
of flowers, to jump universes to one of springtime in
a land of former winter, where no invisible walls stood,
or could stand for more than a few hours at most...

thunk
Posts: 170
Joined: October 3rd, 2015, 8:50 pm
Location: Central USA

Re: Online Database of Unsuccessful GFIND Searches

Post by thunk » March 10th, 2016, 8:53 pm

A for awesome wrote:Just ran an unsuccessful zfind search on 2c/11 width-12 even-symmetric (it outputted "end", so I think that's definite). Took approximately 45 minutes; here are two partials that looked reasonably good]
Ah, thank you. I also ran the coprime period 11 and 12 orthogonals (1c/11, 2c/11, 3c/11, 4c/11, 5c/11, 1c/12, 5c/12) mostly to search depth 5 (w5 asymmetric, 9 odd, 10 even, 11 odd) with gfind and zfind.

The exceptions are 1c/11 and 1c/12 asymmetric and gutter (to search depth 4 only).
"What's purple and commutes?
The Evanston Express."

Post Reply