For discussion of specific patterns or specific families of patterns, both newly-discovered and well-known.
Kazyan
Posts: 908
Joined: February 6th, 2014, 11:02 pm

Aidan Pierce recently released collisrc. It's intended for automatically finding synthesis components, but will happily report any still life that shares a cell with the starting still life. At current search depths, this is a feature, not a bug.

Anyway, collisrc finds. Here's a reduction of this 18-bit still life's synthesis from 6 gliders to 5:

Code: Select all

x = 16, y = 26, rule = B3/S23
3bo$bobo$2b2o6$13b2o$12bo2bo$12bo2bo$13b2o11$3o$2bo7bo$bo8b2o$9bobo!
Tanner Jacobi

Ian07
Posts: 393
Joined: September 22nd, 2018, 8:48 am

Here's a collection of some reductions posted by Exa on Discord (note: this is not a complete list):

Code: Select all

x = 121, y = 140, rule = B3/S23
17$31bo33b2o$29b2o33bobo$30b2o32bo$65bo$62b3o$61bo$60bobo$19bo41bo$18b obo61bobo$17bo2bo62b2o$13b2o2bobo63bo$13bo4bo31bo$14b4o32b2o$49bobo42b
o5bo$16b2o75bobobobobo$17bo11bo64b2ob2o2bo$14b3o11b2o71bobo$14bo13bobo
71b2o22$12bo$13b2o45b2o35b2o$12b2o46b2o35bobo$99bo$60b4o35b2o$59bo3bo
37bo$58bobo38b2o$58bobo38bo$59bo37bobo$21b2o73bobo$21bobo2b2o69bo$22bo
2bo2bo$23b2o2bobo$28b2o2$50bo$11b2o37b2o$12b2o35bobo17b2o$11bo57bobo
19b3o$69bo23bo$92bo10$103bo$102bo$13b2o2b2o83b3o$13bo4bo$14b4o38b2o$
55bobo$14b2o40bo$14bo38b3o$12bobo37bo$12b2o38b4o30b2o3bo$55bo31bo2bobo$54bo11bo20bobobo$54b2o9b2o19b2ob2o$27b3o35bobo19bo$27bo57bobo$28bo56b
2o12$53bo47b2o$52bobo15b2o29bobo$51bobo17bo30bo$51bo18bo32b3o$16b2o32b 2o15bobo36bo$15bobo49b2o34b3o$12bo3bo33b2o50bo$12b4o33bo2bo14b2o32bobo
$49bo2bo13bo2bo31b2o$12b2o36b2o14bo2bo20b3o$11bo2bo52b2o23bo$12b2o11b
2o64bo$25bobo17b2o$25bo18bobo$46bo$75b2o$75bobo$75bo!
EDIT: Part 2. Exa has already posted all of these and the above on Catagolue.

Code: Select all

x = 147, y = 109, rule = B3/S23
12$49b2o53bo$50bo5bo43b2obobo$50bobo2bobo41bobobobo$51b2o3bo42bo4bo25b
o$18b2o33b3o40b2obo26b2obobo$18b2o33bo41bo2bo28bobobo$14bo36bobo41bobo 26b3o3bo$9b2o2bobo35b2o43bo26bo$9bo2bobo107bobo$11b2o108bo2bo$12bo109b 2o$11bo$11b2o15b3o$28bo87b3o$29bo57b3o28bo$89bo27bo$88bo$61b2o$61bobo$
61bo9$10bo$11bo$9b3o23b2o70bo$34bo2bo68bobo$34b4o64b2obobo$101bobobo
27bo$34b4o27bo35bo3bo26bo$33bo4bo27bo33b2o3b2o25b3o$34bo2bo26b3o$33b2o
2b2o66b2o$105b2o$15bo64bo$14bobo62bobo$15b2o58b2obobo$47bo26bobobo50bo$13b4o29b2o26bo3bo49bobo$12bo4bo28bobo24b2o3b2o44b2obobo$12b2ob2o106bo
bobo$13bobo62b2o16bo26bo3bo$13bobo62b2o16b2o24b2o3b2o$14bo80bobo$127b
2o$127b2o11$46bo$46bobo$14bo31b2o$10b2obobo$10b2obobo$13bobobo$10b3o3b
2o$10bo$8bobo$8b2o2$40bo$39bobo$40bobo$41b2o2$41b2o$41bo2bo$17b2o24b2o
$17bobo$17bo!
This program really does show amazing potential; I can only imagine how many new syntheses we'll get once it's been ported to C++. Maybe we can integrate it into apgsearch?

Saka
Posts: 3138
Joined: June 19th, 2015, 8:50 pm
Location: In the kingdom of Sultan Hamengkubuwono X

A fun program. Some stuff I posted on discord, but some are probably known and most are probably not very useful.
Big S -> Table on block

Code: Select all

x = 17, y = 33, rule = B3/S23
12bo$11bo$11b3o2$16bo$14b2o$15b2o6$4b2o$3bo2bo$3bob2o$2obo$o2bo$b2o13$
11bo$10b2o$10bobo!
Big S -> Claw with Tail

Code: Select all

x = 19, y = 30, rule = B3/S23
16bobo$16b2o$17bo10$4b2o$3bo2bo$3bob2o$2obo$o2bo$b2o9$12b2o$8bo3bobo$7b2o3bo$7bobo!
Big S -> 11-loop

Code: Select all

x = 20, y = 26, rule = B3/S23
4b2o$3bo2bo$3bob2o$2obo$o2bo$b2o6$12b3o$12bo$13bo2$17b2o$17bobo$11b3o 3bo$11bo$12bo4$17b3o$17bo$18bo!
Big S -> Big S in a different position and flipped

Code: Select all

x = 18, y = 31, rule = B3/S23
15bo$15bobo$15b2o5$4b2o$3bo2bo$3bob2o$2obo$o2bo$b2o5$12b2o$11b2o$13bo 4$12bo$11b2o$11bobo3$13b2o$13bobo$13bo!  Big S -> Beehive on Table Code: Select all x = 17, y = 21, rule = B3/S23 9bo$7b2o$8b2o7$4b2o$3bo2bo$3bob2o$2obo$o2bo$b2o2$14b2o$14bobo$7b2o5bo$6b2o$8bo!

Big S -> Loaf on Table

Code: Select all

x = 17, y = 22, rule = B3/S23
13bo$11b2o$12b2o6$4b2o$3bo2bo$3bob2o$2obo$o2bo$b2o2$14b3o$14bo$15bo2$
8b3o$8bo$9bo!
Hat on 2 blocks -> xs23_69b8b96z033033

Code: Select all

x = 21, y = 22, rule = B3/S23
7b2ob2o$7b2ob2o2$7b2ob2o$8bobo$8bobo$9bo6$2o$b2o$o13b2o$14bobo$14bo3$18b3o$18bo$19bo!  Hat on 2 blocks -> xs19_gbb8brz123 Code: Select all x = 14, y = 17, rule = B3/S23 6b2ob2o$6b2ob2o2$6b2ob2o$7bobo$7bobo$8bo3$2bo$2b2o$bobo3$bo10b2o$b2o8b 2o$obo10bo!
Airy Clave White It Nay

Code: Select all

x = 17, y = 10, rule = B3/S23
o$2bo22b2o$12b2ob2o$11bobobobo$13bobo7bo$22bo$22b3o$20bo$19b2o$19bobo ! x₁=ηx V ⃰_η=c²√(Λη) K=(Λu²)/2 Pₐ=1−1/(∫^∞_t₀(p(t)ˡ⁽ᵗ⁾)dt) $$x_1=\eta x$$ $$V^*_\eta=c^2\sqrt{\Lambda\eta}$$ $$K=\frac{\Lambda u^2}2$$ $$P_a=1-\frac1{\int^\infty_{t_0}p(t)^{l(t)}dt}$$ http://conwaylife.com/wiki/A_for_all Aidan F. Pierce hkoenig Posts: 99 Joined: June 20th, 2009, 11:40 am ### Re: Thread for your Collisrc finds You might want to consider further reducing these collisions. That way you can more easily weed out the duplicates and redundancies. Code: Select all x=127, y=80 47b2o42b2o$48bo42bo$10b2o33b3o40b2obo$10b2o33bo41bo2bo27bo$6bo36bobo41bob o26b3o$5bobo35b2o43bo26bo$6bo107bobo$113bo2bo$114b2o2$20b3o$20bo87b3o$21b
o57b3o28bo$81bo27bo$80bo$53b2o$53bobo$53bo9$2bo$3bo$b3o3$27bo97bo$27b3o27b
o39b2o25bo$30bo27bo38b2o25b3o$29bo26b3o$29b2o66b2o$97b2o$7bo$6bobo$7b2o58b 2o$39bo26bobo52bo$38b2o26bo53bobo$38bobo24b2o52bobo$119bo$88bo29b2o$88b2o$87b
obo13$38bo$38bobo$38b2o2$4b2o$5bo$2b3o$2bo$obo$2o2$32bo$31bobo$32bobo$33b 2o4$9b2o$9bobo$9bo!

Also, the results of single Gliders interacting with all objects up to 17 bits are known.

Posts: 1930
Joined: November 8th, 2014, 8:48 pm
Location: Getting a snacker from R-Bee's

hkoenig wrote:You might want to consider further reducing these collisions. That way you can more easily weed out the duplicates and redundancies.

Code: Select all

x=127, y=80
common edge cases

A better idea might be to check, for each set of gliders that produces a result, whether the gliders can be divided into two sets that both produce results individually.
If not, you've got a primitive converter.
If so, you have two independent converters.
The question is, how hard is it to determine whether two reactions on different SLs are the same converter?
LifeWiki: Like Wikipedia but with more spaceships. [citation needed]

GUYTU6J
Posts: 728
Joined: August 5th, 2016, 10:27 am
Location: 中国

x = 10, y = 9, rule = B3/S23