Page 2 of 2

Re: ikpx 2.2

Posted: January 21st, 2021, 10:55 pm
by Dylan Chen
if any one want re-find Sir Robin. maybe this two backup bin files would help.


Code: Select all

-rw-r--r-- 1 root root 16M Jan 20 17:41 BeforeSirRobin_2_1_6_width_34_even.bin.zx
-rw-r--r-- 1 root root 17M Jan 22 09:45 AfterSirRobin_2_1_6_width_34_even.bin.zx

Code: Select all

https://slate.textile.io/ipfs/bafybeifwzvgo2egqqbu6mfearbgjifbjyodnxp6oqqgwhs245vlmu6yb6m
https://slate.textile.io/ipfs/bafybeidutce67n44tq2lydmlj2xjtsuuvb2h4s2pvpnxvbau4bqicjpok4
decompress it and run it with ikpx2


edit

and another knight head partial.

Code: Select all

x = 32, y = 51, rule = B3/S23
11bo$10b3o3bobobo$9b2o2b2o2bobob2o$10b3obobo4b3o$10b2o2b2o5b3o$10bob3o
3bo2b3o$10bo2bo4b2ob3o$5b2obo3bo5b5o$4b3o4b3o6bo$2b2o2bo2bob4o$2b2o$o
4bo2b6o$7o$b2o4bob2o2b3o$9b2obo$3bo4bo3bo3bo$7b2o3bo$6bo6b2ob2o$7b2o7b
obo$18bo$14b2o$13bo3b2o$13bo$12bo2bo$11b3o2b2o$17b3o$13bo2bo2b2o$23bob
o$18b2o2bobobo$19b2obobobo$20bobobob2o$19bo3bob2o$21bob4obo$21bobobobo
$21bob2obo$18b2o$19bobo4b2o$19bo$19bo$18b2ob2o$18b2o$17bo4bo$17bob3obo
$18b7o2bo$19bo3b3ob2o$24b2obob2o2$27bo3bo$25bob4o$24b3o$26bo!

Re: ikpx 2.2

Posted: January 25th, 2021, 6:52 pm
by creeperman7002
Can there be a scoring system for partials (score = depth + breadth)? That could be used to determine the best partial in a search.
Another way to determine the best partial could be to see how long-lived the first few rows of the partial are.

Re: ikpx 2.2

Posted: January 28th, 2021, 8:31 am
by Bullet51
Is it possible to extend partials from the tail of the spaceship, e.g.

Code: Select all

x = 10, y = 20, rule = B3/S23
4bobo$2bo$2b2o2bo$3b3o$7b2o$bo5b3o$obo2b2o$obob3o2bo$2bo2bob2o$2o2bobo
$3b2o$bo4b3o$bo4b3o$3o3b2o$2b2ob2o$4b2o$5bo2$5b2o$4bo!
?

Re: ikpx 2.2

Posted: February 10th, 2021, 1:54 am
by GUYTU6J
When searching for a knightship in Glimmering Garden, it reported the following issue:

Code: Select all

Warning: SAT solver 'kissat' reached decision limit
Can I just ignore it?

Re: ikpx 2.2

Posted: February 10th, 2021, 2:22 pm
by goldenratio
GUYTU6J wrote:
February 10th, 2021, 1:54 am
When searching for a knightship in Glimmering Garden, it reported the following issue:

Code: Select all

Warning: SAT solver 'kissat' reached decision limit
Can I just ignore it?
From Calcyman on discord:
calcyman wrote:To avoid threads becoming blocked with particularly difficult SAT problems, ikpx2 imposes a decision limit (which usually equates to about 15 minutes) after which the SAT solver gives up and the thread works on a different problem instead. This was in the original version of ikpx, but ikpx2 is polite enough to loudly tell you when this happens. :)

If you're just interested in finding spaceships, then you can ignore these messages. If you want to populate some 'spaceship status page' on the LifeWiki to say that a certain width has no spaceships, then those messages are important (because they warn you that the search may have been non-exhaustive).
He also said how to disable the decision limit:
calcyman wrote:Yes, comment out the calls to kissat_set_decision_limit and ccadical_limit in src/solver.hpp and recompile.

Re: ikpx 2.2

Posted: March 26th, 2021, 12:00 pm
by LaundryPizza03
What is the width of the valid rows in this (7,1)c/8 partial? I suspect that ikpx2 may not be powerful enough to find this speed, as the partial in question is the narrowest possible at the tip.

Code: Select all

x = 12, y = 10, rule = B2aci3aj4cijkrtz5ajkqy6kn78/S2ak3ejkqr4cijqtwz5-aenr6ace7e8
2bob2obobobo$2obo5b3o$bo3b4obo$3bo2b2o2b2o$6bo2b3o$3bo2bobobo$bo2bo3bo
$4bobo$b2o3b4o$5b5o!