Database of All Completed and Ongoing *find Searches

For discussion of specific patterns or specific families of patterns, both newly-discovered and well-known.
User avatar
testitemqlstudop
Posts: 1367
Joined: July 21st, 2016, 11:45 am
Location: in catagolue
Contact:

Database of All Completed and Ongoing *find Searches

Post by testitemqlstudop » January 22nd, 2019, 4:29 pm

Well,
testitemqlstudop wrote:Oh f*** it, 3092376453120 calculations wasted.

I ran three separate instances of ntzfind, all with width 9, one doing asymmetric, one doing odd-symmetric, and one doing even-symmetric.

We should really have a database of what's searched...
So I decided to start this thread.

All searches are tabulated in order of widest last search tried at User:Sokwe/Spaceship_searches

Let's not waste any more computing power!
Last edited by testitemqlstudop on March 3rd, 2019, 9:20 pm, edited 6 times in total.

User avatar
Freywa
Posts: 877
Joined: June 23rd, 2011, 3:20 am
Location: Singapore
Contact:

Re: Database of All Completed and Ongoing *find Searches

Post by Freywa » January 22nd, 2019, 4:53 pm

You do note that there's already articles on LifeWiki on precisely this topic, right?
Princess of Science, Parcly Taxel

Code: Select all

x = 31, y = 5, rule = B2-a/S12
3bo23bo$2obo4bo13bo4bob2o$3bo4bo13bo4bo$2bo4bobo11bobo4bo$2bo25bo!

User avatar
testitemqlstudop
Posts: 1367
Joined: July 21st, 2016, 11:45 am
Location: in catagolue
Contact:

Re: Database of All Completed and Ongoing *find Searches

Post by testitemqlstudop » January 22nd, 2019, 5:22 pm

Wait what? No there isn't, i looked at all 147 LifeWiki results for "search"!

User avatar
Ian07
Moderator
Posts: 891
Joined: September 22nd, 2018, 8:48 am
Location: New Jersey, US

Re: Database of All Completed and Ongoing *find Searches

Post by Ian07 » January 22nd, 2019, 5:50 pm

testitemqlstudop wrote:Wait what? No there isn't, i looked at all 147 LifeWiki results for "search"!
Here it is: http://www.conwaylife.com/wiki/LifeWiki ... tatus_Page

User avatar
2718281828
Posts: 738
Joined: August 8th, 2017, 5:38 pm

Re: Database of All Completed and Ongoing *find Searches

Post by 2718281828 » January 22nd, 2019, 5:51 pm

Freywa wrote:You do note that there's already articles on LifeWiki on precisely this topic, right?
If testitemqlstudop looked for all 'Search' pages then this one was found:
http://www.conwaylife.com/wiki/LifeWiki ... tatus_Page


However, I think testitemqlstudop is looking for ongoing searches. I think this idea isn't too bad as all 'new' searches will need always at least a couple of days. I think the widest 3c/7 search took about a year to complete. And it makes sense that we are not all running the same search at the same time but distribute it.

Moreover, I think the wiki is not up to date, I think there are some (negative) results in viewtopic.php?f=2&t=2031 that are not in the table.

User avatar
testitemqlstudop
Posts: 1367
Joined: July 21st, 2016, 11:45 am
Location: in catagolue
Contact:

Re: Database of All Completed and Ongoing *find Searches

Post by testitemqlstudop » January 22nd, 2019, 6:02 pm

Ian07 wrote:
testitemqlstudop wrote:Wait what? No there isn't, i looked at all 147 LifeWiki results for "search"!
Here it is: http://www.conwaylife.com/wiki/LifeWiki ... tatus_Page
That hasn't been updated since March!

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

Re: Database of All Completed and Ongoing *find Searches

Post by Sokwe » January 22nd, 2019, 6:49 pm

testitemqlstudop wrote:Wait what? No there isn't, i looked at all 147 LifeWiki results for "search"!
A standard search on the wiki only looks in the main namespace, but the table is in the LifeWiki namespace.

After you do a search on the wiki, look just below the search bar. There should be a link labeled "Everything". Click that link to ensure that you get every page containing your search terms.
testitemqlstudop wrote:
That hasn't been updated since March!
Very little has changed since March. I'll add the few exceptions soon.

There is also a page for completed searches using various programs in case some program is found to have a bug. It should be noted that my old zfind-v2.0/s code has bugs that could potentially cause errors for high speed searches. As a result, the high speed results in the table that rely on early zfind searches might be invalid (I eventually intend to hunt these down and double-check them with Tom Rokicki's zfind fix).
-Matthias Merzenich

User avatar
Billabob
Posts: 158
Joined: April 2nd, 2015, 5:28 pm

Re: Database of All Completed and Ongoing *find Searches

Post by Billabob » January 22nd, 2019, 7:14 pm

Actually, I've been running higher-period searches with ntzfind 3.0 for quite a long time without reporting anything. I haven't had a single positive result, but you may find my spreadsheet of negatives useful:
partials.zip
Spaceship search table + partials
(24.05 KiB) Downloaded 473 times
Most of the searches took a negligible amount of time - a few hours at most. If I find promising partials I save them and search again at a higher width - at the moment I have an ntzfind instance that's been running for 16.2 days, although I've forgotten what it's for.
▄▀
▀▀▀

AforAmpere
Posts: 1334
Joined: July 1st, 2016, 3:58 pm

Re: Database of All Completed and Ongoing *find Searches

Post by AforAmpere » January 22nd, 2019, 8:25 pm

Negative:

Code: Select all

./ntzfind3_1 B3/S23 p11 k1 x1 w6 a
./ntzfind3_1 B3/S23 p11 k2 x1 w6 a
./ntzfind3_1 B3/S23 p11 k3 x1 w6 a
./ntzfind3_1 B3/S23 p11 k4 x1 w7 a
Filling in some gaps in the above. Should I update the wiki?
I manage the 5S project, which collects all known spaceship speeds in Isotropic Non-totalistic rules. I also wrote EPE, a tool for searching in the INT rulespace.

Things to work on:
- Find (7,1)c/8 and 9c/10 ships in non-B0 INT.
- EPE improvements.

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

Re: Database of All Completed and Ongoing *find Searches

Post by Sokwe » January 22nd, 2019, 8:37 pm

Billabob wrote:you may find my spreadsheet of negatives useful
Thanks, although you have recorded some things differently from the main table. On the wiki, a cell is marked green if there is a known spaceship of that width/speed combination and it has been proven that no thinner spaceships of that type exist. Your table, on the other hand, includes for example a green "48" for (0,5)c/15 asymmetric ships. I'm certain that you haven't proved the nonexistence of a width-47 p15 c/3 orthogonal ship, considering this ship exists:

Code: Select all

x = 30, y = 62, rule = B3/S23
8b2o3b2o$5b2ob2o3b2ob2o$2b4o2bobobobo2b4o$bo4bo2b2ob2o2bo4bo$2b2o6bobo
6b2o$7bobo3bobo$5b3ob5ob3o$4b2obo7bob2o$3bo2bo2bobobo2bo2bo$9bobobo$3b
ob2ob2obob2ob2obo$2b6o3bo3b6o$2b2ob2obo5bob2ob2o2$6b3obobob3o$5bo4bobo
4bo$5bo3b2ob2o3bo$3bo15bo$3bo4bo5bo4bo$5b2ob7ob2o$7bobobobobo$4bo13bo$
2b3o2b2ob3ob2o2b3o$bo2bob2o2b3o2b2obo2bo$o3bob5ob5obo3bo$obobo13bobobo
$b2obob3o5b3obob2o$2b2obobo7bobob2o$4b3obob3obob3o$bo5bo2b3o2bo5bo$6bo
2bo3bo2bo$6b2o7b2o$5bob3o3b3obo$5bo11bo$5bobo2b3o2bobo$2b2obobo7bobob
2o$2b2obobo2bobo2bobob2o$5bobob2ob2obobo$4bobob2obob2obobo$2b3o13b3o$
9b5o$2b2o7bo7b2o$3bo15bo$3bo6b3o6bo$3bob2o2b5o2b2obo$7bob5obo$8bo5bo2$
b2ob4o7b4ob2o$b2o17b2o$b4o3b7o3b4o$3bo15bo$4b2o11b2o2$24bo$22b3ob3o$
21b2o6bo$16bo3bo2bo3b2o$11b2o2b4o$10bo7bo$10bo4bobo2bo$19bo!
Billabob wrote:at the moment I have an ntzfind instance that's been running for 16.2 days, although I've forgotten what it's for.
zfind should echo your options back to you at the start of the search. If you send the output to a file you can simply look at the top of the file to see what the search is. For any potentially long search you should always send the output to a file. To do so, simply add "> output.txt" or ">> output.txt" after your zfind parameters. Obviously, you can change the name of the output file if you wish (you wouldn't want to overwrite your previous output).
-Matthias Merzenich

User avatar
Billabob
Posts: 158
Joined: April 2nd, 2015, 5:28 pm

Re: Database of All Completed and Ongoing *find Searches

Post by Billabob » January 22nd, 2019, 8:59 pm

Sokwe wrote:
Billabob wrote:you may find my spreadsheet of negatives useful
Thanks, although you have recorded some things differently from the main table. On the wiki, a cell is marked green if there is a known spaceship of that width/speed combination and it has been proven that no thinner spaceships of that type exist. Your table, on the other hand, includes for example a green "48" for (0,5)c/15 asymmetric ships. I'm certain that you haven't proved the nonexistence of a width-47 p15 c/3 orthogonal ship, considering this ship exists:
Sorry, I didn't know that. Thanks for the heads up. I think when I started in March I didn't like the look of the empty rows, so I just filled them in with the first spaceships I could find in jslife. Please note there will be some outdated data among the lower periods because I've only been adding my own results.
Sokwe wrote:
Billabob wrote:at the moment I have an ntzfind instance that's been running for 16.2 days, although I've forgotten what it's for.
zfind should echo your options back to you at the start of the search. If you send the output to a file you can simply look at the top of the file to see what the search is. For any potentially long search you should always send the output to a file. To do so, simply add "> output.txt" or ">> output.txt" after your zfind parameters. Obviously, you can change the name of the output file if you wish (you wouldn't want to overwrite your previous output).
Excellent, thank you for the tip. I feel a bit silly for throwing away all those partials now! Might consider re-running some of the searches just to capture more partials - in fact I might tackle 6c/13 width 8 (then maybe 9) if nobody has any objections. Width 7 took 66 seconds, so 9 shouldn't take longer than a few weeks.
▄▀
▀▀▀

AforAmpere
Posts: 1334
Joined: July 1st, 2016, 3:58 pm

Re: Database of All Completed and Ongoing *find Searches

Post by AforAmpere » January 22nd, 2019, 9:44 pm

Sokwe, is it alright for others to update the status table? I'd like to put in some stuff.

EDIT, also negative:

Code: Select all

./ntzfind3_1 B3/S23 p11 k3 x1 w7 a
./ntzfind3_1 B3/S23 p11 k4 x1 w8 a
./ntzfind3_1 B3/S23 p12 k1 x1 w5 a
./ntzfind3_1 B3/S23 p12 k5 x1 w8 a
./ntzfind3_1 B3/S23 p12 k1 x1 w5 a
./ntzfind3_1 B3/S23 p12 k2 x1 w6 a
./ntzfind3_1 B3/S23 p12 k3 x1 w6 a
./ntzfind3_1 B3/S23 p12 k4 x1 w7 a
I manage the 5S project, which collects all known spaceship speeds in Isotropic Non-totalistic rules. I also wrote EPE, a tool for searching in the INT rulespace.

Things to work on:
- Find (7,1)c/8 and 9c/10 ships in non-B0 INT.
- EPE improvements.

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

Re: Database of All Completed and Ongoing *find Searches

Post by Sokwe » January 22nd, 2019, 10:43 pm

AforAmpere wrote:Sokwe, is it alright for others to update the status table? I'd like to put in some stuff.
It's generally alright, however...
AforAmpere wrote:Negative:

Code: Select all

./ntzfind3_1 B3/S23 p11 k1 x1 w6 a
./ntzfind3_1 B3/S23 p11 k2 x1 w6 a
./ntzfind3_1 B3/S23 p11 k3 x1 w6 a
./ntzfind3_1 B3/S23 p11 k4 x1 w7 a
Filling in some gaps in the above. Should I update the wiki?
I'm not sure how Tom implemented the "x1" search feature in ntzfind, but if it's anything like my modification, it wouldn't actually cover every possible pattern of the given width. For that reason I avoid using zfind results for these entries.

Instead, please record those searches here.
-Matthias Merzenich

User avatar
testitemqlstudop
Posts: 1367
Joined: July 21st, 2016, 11:45 am
Location: in catagolue
Contact:

Re: Database of All Completed and Ongoing *find Searches

Post by testitemqlstudop » January 24th, 2019, 1:00 pm

Anyways, ntzfind finished a width-10 c/3 search in less than a second (it's not on the wiki, soo....)

none exsist, best partial:

Code: Select all

x = 10, y = 16, rule = B3/S23
6bo$2b2obobo$2b2o4bo$2b3o3bo$5bo$2b2o3b2o$2b2obo3bo$4bo3bo$2o3bo$2o2bo
bo$bob2ob2o$b3ob3o$o6bo$ob4o$2b4o$7b2o!
I don't have enough ram for width-11, but I guess I'll die trying...

AforAmpere
Posts: 1334
Joined: July 1st, 2016, 3:58 pm

Re: Database of All Completed and Ongoing *find Searches

Post by AforAmpere » January 25th, 2019, 8:28 pm

How did you set up ikpx testitemqlstudop? Are you on Windows or not? Also, I am currently running the previously done zfind searches with the latest version.
I manage the 5S project, which collects all known spaceship speeds in Isotropic Non-totalistic rules. I also wrote EPE, a tool for searching in the INT rulespace.

Things to work on:
- Find (7,1)c/8 and 9c/10 ships in non-B0 INT.
- EPE improvements.

User avatar
testitemqlstudop
Posts: 1367
Joined: July 21st, 2016, 11:45 am
Location: in catagolue
Contact:

Re: Database of All Completed and Ongoing *find Searches

Post by testitemqlstudop » January 25th, 2019, 8:46 pm

AforAmpere wrote:How did you set up ikpx testitemqlstudop? Are you on Windows or not? Also, I am currently running the previously done zfind searches with the latest version.
If you are on windows, use WSL (fast) or Cygwin (slow). I use linux.

Do "git clone https://gitlab.com/apgoucher/metasat" - no quotes OR download the archive and extract.
Go to the metasat directory (if you used git, do "cd metasat")
Make sure you have python 2 installed.
Go into terminal (cmd on windows, you know if you're on linux) and run "python ikpx.py".
Follow instructions from there.

If you have questions, feel free to ask me

AforAmpere
Posts: 1334
Joined: July 1st, 2016, 3:58 pm

Re: Database of All Completed and Ongoing *find Searches

Post by AforAmpere » January 25th, 2019, 8:53 pm

When I tried that command, I got this:

Code: Select all

Cloning into 'metasat'...
fatal: unable to access 'https://gitlab.com/apgoucher/metasat/': error setting certificate verify locations:
  CAfile: /etc/pki/tls/certs/ca-bundle.crt
  CApath: none
I am pretty inexperienced with git, what did I do?

I tried something else, and I still got an error:

Code: Select all

git clone git@gitlab.com:apgoucher/metasat
Cloning into 'metasat'...
git@gitlab.com: Permission denied (publickey).
fatal: Could not read from remote repository.

Please make sure you have the correct access rights
and the repository exists.
I manage the 5S project, which collects all known spaceship speeds in Isotropic Non-totalistic rules. I also wrote EPE, a tool for searching in the INT rulespace.

Things to work on:
- Find (7,1)c/8 and 9c/10 ships in non-B0 INT.
- EPE improvements.

AforAmpere
Posts: 1334
Joined: July 1st, 2016, 3:58 pm

Re: Database of All Completed and Ongoing *find Searches

Post by AforAmpere » January 25th, 2019, 9:06 pm

I installed it after downloading ikpx from the site, but iglucose does not compile properly.

Code: Select all

Installing iglucose...
installers/add_iglucose.sh: line 7: cd: iglucose/core: No such file or directory
/metasat-master/scripts/iglucose.sh: line 10: compiled/iglucose: No such file or directory
Exit status: 127
Traceback (most recent call last):
  File "ikpx.py", line 1170, in <module>
    clmain()
  File "ikpx.py", line 1166, in clmain
    do_everything(psets, params, directory, encoding=args.encoding, loadhead=args.loadhead, cmd=args.command)
  File "ikpx.py", line 897, in do_everything
    raise ValueError('iglucose exited with status %d' % status)
ValueError: iglucose exited with status 127
I manage the 5S project, which collects all known spaceship speeds in Isotropic Non-totalistic rules. I also wrote EPE, a tool for searching in the INT rulespace.

Things to work on:
- Find (7,1)c/8 and 9c/10 ships in non-B0 INT.
- EPE improvements.

User avatar
calcyman
Moderator
Posts: 2932
Joined: June 1st, 2009, 4:32 pm

Re: Database of All Completed and Ongoing *find Searches

Post by calcyman » January 25th, 2019, 9:16 pm

It has submodules, so you need to actually clone it:

Code: Select all

git clone https://gitlab.com/apgoucher/metasat.git
Note the .git extension you were missing before.
What do you do with ill crystallographers? Take them to the mono-clinic!

User avatar
testitemqlstudop
Posts: 1367
Joined: July 21st, 2016, 11:45 am
Location: in catagolue
Contact:

Re: Database of All Completed and Ongoing *find Searches

Post by testitemqlstudop » January 25th, 2019, 9:19 pm

AforAmpere wrote:I installed it after downloading ikpx from the site, but iglucose does not compile properly.

Code: Select all

Installing iglucose...
installers/add_iglucose.sh: line 7: cd: iglucose/core: No such file or directory
/metasat-master/scripts/iglucose.sh: line 10: compiled/iglucose: No such file or directory
Exit status: 127
Traceback (most recent call last):
  File "ikpx.py", line 1170, in <module>
    clmain()
  File "ikpx.py", line 1166, in clmain
    do_everything(psets, params, directory, encoding=args.encoding, loadhead=args.loadhead, cmd=args.command)
  File "ikpx.py", line 897, in do_everything
    raise ValueError('iglucose exited with status %d' % status)
ValueError: iglucose exited with status 127
Wahht? Oh, a new version, I see
You have to go to https://gitlab.com/apgoucher/iglucose and download it into the iglucose folder, and also for https://github.com/arminbiere/lingeling
then run.

Or just do what calcyman said, and then "git submodule init" + "git submodule update".

AforAmpere
Posts: 1334
Joined: July 1st, 2016, 3:58 pm

Re: Database of All Completed and Ongoing *find Searches

Post by AforAmpere » January 25th, 2019, 9:34 pm

Doing the test input I was given a while a go, it still does not output anything. Maybe it just can't work on Windows.
I manage the 5S project, which collects all known spaceship speeds in Isotropic Non-totalistic rules. I also wrote EPE, a tool for searching in the INT rulespace.

Things to work on:
- Find (7,1)c/8 and 9c/10 ships in non-B0 INT.
- EPE improvements.

User avatar
testitemqlstudop
Posts: 1367
Joined: July 21st, 2016, 11:45 am
Location: in catagolue
Contact:

Re: Database of All Completed and Ongoing *find Searches

Post by testitemqlstudop » January 25th, 2019, 10:03 pm

Maybe calcyman knows what's wrong?

Are you getting an iglucose error? It should be platform-independent, after all.

AforAmpere
Posts: 1334
Joined: July 1st, 2016, 3:58 pm

Re: Database of All Completed and Ongoing *find Searches

Post by AforAmpere » January 25th, 2019, 10:05 pm

testitemqlstudop wrote: Are you getting an iglucose error?
No errors. It outputs everything up to the kill SIGINT line, and then doesn't output anything else, even though it should be a quick search.
I manage the 5S project, which collects all known spaceship speeds in Isotropic Non-totalistic rules. I also wrote EPE, a tool for searching in the INT rulespace.

Things to work on:
- Find (7,1)c/8 and 9c/10 ships in non-B0 INT.
- EPE improvements.

User avatar
calcyman
Moderator
Posts: 2932
Joined: June 1st, 2009, 4:32 pm

Re: Database of All Completed and Ongoing *find Searches

Post by calcyman » January 26th, 2019, 7:12 am

AforAmpere wrote:
testitemqlstudop wrote: Are you getting an iglucose error?
No errors. It outputs everything up to the kill SIGINT line, and then doesn't output anything else, even though it should be a quick search.
It's probably a problem with os.mkfifo() on your platform. Are you using Cygwin or WSL?

In fact, has anyone successfully managed to get ikpx running on Cygwin? There seems to be an issue regarding named pipes: https://www.cygwin.com/ml/cygwin/2016-01/msg00085.html
What do you do with ill crystallographers? Take them to the mono-clinic!

User avatar
2718281828
Posts: 738
Joined: August 8th, 2017, 5:38 pm

Re: Database of All Completed and Ongoing *find Searches

Post by 2718281828 » January 26th, 2019, 9:16 am

There is no camelship of width 12 (not really surprising), done using ./knight2 -p 8 -x 3 -y 1 -n -w 12 it took about a day to complete. So I think width 13 is feasible in about a week. But I won't run it soon. But looking at the partial it would be extremely surprising if there is a ship with width < 16.

Best reported (end) partial:

Code: Select all

x = 10, y = 16, rule = B3/S23
7b2o$3o2bobo$2obo$o2b4o$b2o$2bob2ob2o$6bo$2b2o2bobo$bo2bob3o$bo4bobo$
3bo4b2o$4b3o2bo$4bo$5bo3bo$5bob2o$6bo!
not really great.

Post Reply