Small Spaceship Syntheses

[Chris857]

31P8H4V0 down 1G, to 35G

Code: Select all

x = 54, y = 34, rule = B3/S23
9bo3b2o$8bobobo2bo$9b2obob2o$11bobo$11bobo$10b2ob2o$11bobo$11bobo$12b
o14$38b2o$38bobo$38bo2$b2o$obo49bo$2bo48b2o$51bobo2$b2o$2b2o$bo!

[Disaster16439]

How are the big syntheses found? For example, in the recent Lobster synthesis, there seems to be a bunch of crash-random-gliders-into-random-objects-to-provide-just-the-right-thing-type shenanigans, that don’t seem to be possible, even with predecessor searchers like JLS. Especially with the LOM+2G activation of the “wings”.

[iNoMed]

Lobster reduced to 123G by building the edgy xs24s in a much cheaper manner that also ends up reducing the synthesis cost of the xs24 in and of itself (13G to 10G):

Code: Select all

x = 339, y = 112, rule = B3/S23
204bobo$205b2o$205bo18$238bo24bo$236bobo22b2o$237b2o23b2o2$308bo$309b
2o$308b2o6$314bo$253bo61b2o$253bobo58b2o$253b2o2$314b2o$315b2o7b2ob2o$
314bo7bo2bob2o$322b2obo$259bo47bo17bob2o$258bobo46b2o13b2obob2o$259bo
46bobo14bobo$262bo60bobo$262bo61bo$262bo$259bo$258bobo55bo4b2o11b2ob2o
$225b2o32bo55bobo2bo2bo10b2ob2o$224bobo89b2o2bobo$226bo94bo10b7o$2bo4b
2o35bobo50bo4b2o11b2ob2o131bo4b2o11b2ob2o57bo6bo$bobo2bo2bo10b3o21b2o
50bobo2bo2bo10b2ob2o130bobo2bo2bo10b2ob2o41bo11bo4b3obo$2b2o2bobo36bo
51b2o2bobo147b2o2bobo56bobo3b2o4bobo5bob2o$7bo10bo5bo77bo10b7o136bo10b
7o40b2o3bo2bo2bo2bo$17bobo3bobo86bo6bo146bo6bo46bobo3b2o$bo11bo4bo5bo
71bo11bo4b3obo132bo11bo4b3obo42b2o5bo10bo$obo3b2o4bobo80bobo3b2o4bobo
5bob2o130bobo3b2o4bobo5bob2o41bo2bo13bobo$2o3bo2bo2bo2bo80b2o3bo2bo2bo
2bo139b2o3bo2bo2bo2bo51b2obo7b2o4bo$6bobo3b2o18bo68bobo3b2o146bobo3b2o
54bo4bo3bobo$2o5bo23bo63b2o5bo126b2o18b2o5bo10bo49bobobobo3bo$o2bo27b
3o61bo2bo131b2o17bo2bo13bobo49b2ob2o$b2obo7b2o82b2obo7b2o120bo20b2obo
7b2o4bo$3bo4bo3bobo83bo4bo3bobo9bo4bo127bo4bo3bobo$3bobobobo3bo84bobob
obo3bo10b2o2b2o127bobobobo3bo$4b2ob2o90b2ob2o14bobo2bobo127b2ob2o4$
132bo$131b2o$114b2o15bobo$115b2o$114bo3$44b3o$44bo$45bo5$4bo$4b2o$3bob
o3$15b3o$17bo$16bo18$65b2o$64b2o$66bo!

[Goldtiger997]

Slightly simplified final step which I believe should constitute a reduction (I got rid of the beacon):

Code: Select all

x = 83, y = 94, rule = B3/S23
20bobo$21b2o$21bo4$44bo$44bobo$44b2o$20bo16bo12bo$18bobo14b2o12bo$19b
2o10bo4b2o11b3o$o28bobo$b2o27b2o$2o2$61bo$39bo19b2o$38bo21b2o$38b3o2$
10bo19bobo31bobo$11b2o17b2o32b2o$10b2o19bo33bo2$16bobo$17b2o$17bo$40b
2ob2o$12bo25bo2bob2o$2bo7bobo25b2obo$3b2o6b2o19bo8bob2o26bo$2b2o27bobo
4b2obob2o26bobo$32bobo4bobo29b2o$9bobo21bobo3bobo$10b2o22bo5bo$10bo19b
o44bo$29bobo42bo$30bobo4b2o11b2ob2o19b3o$31bobo2bo2bo10b2ob2o$32b2o2bo
bo$27b2o8bo10b7o$27b2o18bo6bo8bo$31bo11bo4b3obo10bobo$30bobo3b2o4bobo
5bob2o9b2o$26b2o2b2o3bo2bo2bo2bo26bobo$25bobo8bobo3b2o27b2o$26bo3b2o5b
o10bo23bo$30bo2bo13bobo$31b2obo7b2o4bobo$33bo4bo3bobo4bobo9bo$33bobobo
bo3bobo4bo8b2o8b3o$34b2ob2o5bobo13b2o7bo$5b2o13b2o23bo24bo$6b2o12b2o
18b2o$5bo30b2o2b2o$35bobo$36bo2$7b2o$6bobo72b2o$8bo71b2o$36bo34b3o8bo$
35b2o34bo$35bobo18b2o14bo$55b2o$57bo$14b2o5b2o$13bobo6b2o$8b3o4bo5bo
15b3o3b3o$10bo26bo5bo7b3o6bo$9bo28bo5bo6bo7b2o$52bo6bobo$17b3o$19bo$
11b3o4bo20b3o$13bo14b3o8bo$12bo17bo9bo$29bo2$69bo$10bo57b2o$10b2o56bob
o$9bobo8$28b2o$29b2o$28bo!

How are the big syntheses found? For example, in the recent Lobster synthesis, there seems to be a bunch of crash-random-gliders-into-random-objects-to-provide-just-the-right-thing-type shenanigans, that don’t seem to be possible, even with predecessor searchers like JLS. Especially with the LOM+2G activation of the “wings”.

My technique for these sorts of syntheses is fairly ad hoc, though these days it generally involves running a lot of searches with LLS and QuFince. I run LLS predecessor searches where some region towards the centre of the spaceship is initially stable and the population is constrained. I then sift through those results manually to look for options which look the least space-dusty.

Regarding the LOM+2G activation of the wings, I first found this mostly correct predecessor for what needed happening with LLS:

Code: Select all

x = 12, y = 13, rule = B3/S23
b2ob2o$b2ob2o2$b7o$o6bo$b3obo$3bob2o$10bo$7bo3bo$bo5bobobo$obo5bo2bo$b
obo$2bo! [[ STOP 10 ]] 

That looked promising enough to throw into a 4G QuFince search. That search produced the mentioned LOM+2G collision, which was conveniently edgy enough to avoid colliding with some of the recipes I had found for activating the middle part of the Lobster.

[iNoMed]

The above should bring it to 119G as well, following these steps:

Code: Select all

x = 166, y = 94, rule = B3/S23
103bobo$104b2o$104bo4$127bo$127bobo$127b2o$103bo16bo12bo$101bobo14b2o
12bo$102b2o10bo4b2o11b3o$83bo28bobo$84b2o27b2o$83b2o2$144bo$122bo19b2o
$121bo21b2o$121b3o2$93bo19bobo31bobo$94b2o17b2o32b2o$93b2o19bo33bo2$
99bobo$100b2o$100bo$26b2ob2o92b2ob2o$24bo2bob2o64bo25bo2bob2o$24b2obo
57bo7bobo25b2obo$18bo8bob2o55b2o6b2o19bo8bob2o26bo$17bobo4b2obob2o54b
2o27bobo4b2obob2o26bobo$18bobo4bobo87bobo4bobo29b2o$19bobo3bobo64bobo
21bobo3bobo$20bo5bo66b2o22bo5bo$16bo76bo19bo44bo$15bobo94bobo42bo$16bo
bo4b2o11b2ob2o72bobo4b2o11b2ob2o19b3o$17bobo2bo2bo10b2ob2o73bobo2bo2bo
10b2ob2o$18b2o2bobo90b2o2bobo$13b2o8bo10b7o69b2o8bo10b7o$13b2o18bo6bo
69b2o18bo6bo8bo$17bo11bo4b3obo75bo11bo4b3obo10bobo$16bobo3b2o4bobo5bob
2o73bobo3b2o4bobo5bob2o9b2o$12b2o2b2o3bo2bo2bo2bo78b2o2b2o3bo2bo2bo2bo
26bobo$11bobo8bobo3b2o78bobo8bobo3b2o27b2o$12bo3b2o5bo10bo74bo3b2o5bo
10bo23bo$16bo2bo13bobo77bo2bo13bobo$17b2obo7b2o4bobo77b2obo7b2o4bobo$
19bo4bo3bobo4bobo78bo4bo3bobo4bobo9bo$19bobobobo3bobo4bo79bobobobo3bob
o4bo8b2o8b3o$20b2ob2o5bobo84b2ob2o5bobo13b2o7bo$o30bo56b2o13b2o23bo24b
o$b2o23b2o61b2o12b2o18b2o$2o20b2o2b2o60bo30b2o2b2o$7b3o11bobo94bobo$7b
o14bo96bo$8bo$90b2o$89bobo72b2o$91bo71b2o$119bo34b3o8bo$118b2o34bo$
118bobo18b2o14bo$138b2o$140bo$97b2o5b2o$96bobo6b2o$91b3o4bo5bo15b3o3b
3o$93bo26bo5bo7b3o6bo$92bo28bo5bo6bo7b2o$135bo6bobo$100b3o$102bo$94b3o
4bo20b3o$96bo14b3o8bo$95bo17bo9bo$112bo2$152bo$93bo57b2o$93b2o56bobo$
92bobo8$111b2o$112b2o$111bo!

[vilc]

118G, with a slightly cheaper cleanup.

Code: Select all

x = 83, y = 99, rule = B3/S23
20bobo$21b2o$21bo4$44bo$44bobo$44b2o$20bo16bo12bo$18bobo14b2o12bo$19b
2o10bo4b2o11b3o$o28bobo$b2o27b2o$2o2$61bo$39bo19b2o$38bo21b2o$38b3o2$
10bo19bobo31bobo$11b2o17b2o32b2o$10b2o19bo33bo2$16bobo$17b2o$17bo$40b
2ob2o$12bo25bo2bob2o$2bo7bobo25b2obo$3b2o6b2o19bo8bob2o26bo$2b2o27bobo
4b2obob2o26bobo$32bobo4bobo29b2o$9bobo21bobo3bobo$10b2o22bo5bo$10bo19b
o44bo$29bobo42bo$30bobo4b2o11b2ob2o19b3o$31bobo2bo2bo10b2ob2o$32b2o2bo
bo$27b2o8bo10b7o$27b2o18bo6bo8bo$31bo11bo4b3obo10bobo$30bobo3b2o4bobo
5bob2o9b2o$26b2o2b2o3bo2bo2bo2bo26bobo$25bobo8bobo3b2o27b2o$26bo3b2o5b
o10bo23bo$30bo2bo13bobo$31b2obo7b2o4bobo$33bo4bo3bobo4bobo9bo$33bobobo
bo3bobo4bo8b2o8b3o$34b2ob2o5bobo13b2o7bo$5b2o13b2o23bo24bo$6b2o12b2o
18b2o$5bo30b2o2b2o$35bobo$36bo2$7b2o$6bobo72b2o$8bo71b2o$36bo34b3o8bo$
35b2o34bo$35bobo18b2o14bo$55b2o$57bo$14b2o5b2o$13bobo6b2o$8b3o4bo5bo
15b3o3b3o$10bo26bo5bo7b3o6bo$9bo28bo5bo6bo7b2o$52bo6bobo$17b3o$19bo$
11b3o4bo20b3o$13bo14b3o8bo$12bo17bo9bo$29bo3$10bo$10b2o$9bobo13$21b2o$
22b2o$21bo!

[Chris857]

55P9H3V0 down 4G, to 193G

Code: Select all

x = 198, y = 45, rule = B3/S23
40bobo111bobo$40b2o113b2o$41bo113bo3$36bo123bo$8bo26bo125bo26bo$7bobo
25b3o121b3o25bobo$8bo179bo$5bo5bo173bo5bo$4bobo3bobo171bobo3bobo$5bo2b
o2bo173bo2bo2bo$6b2ob2o175b2ob2o$8bo179bo$8bo179bo$6b2ob2o175b2ob2o$b
o5bobo5bo165bo5bobo5bo$obo2bobobobo2bobo163bobo2bobobobo2bobo$bo2bobo
3bobo2bo165bo2bobo3bobo2bo$5bo5bo173bo5bo$6b3o177b3o$9bo179bo6b2o$8bo
bo177bobo3bo2bo$8bo2bo176bo2bo2b2o$9b2o178b2o14$33bo129bo$32b2o129b2o
$32bobo127bobo2$41b2o111b2o$41bobo109bobo$41bo113bo!

And turtle down 1G, to 224G (note: looks like it might have already known of this reduction, but isn't reflected in the whole spaceship synth? I just resubmitted all of its steps to see if that forces an update)

Code: Select all

x = 38, y = 45, rule = B3/S23
obo$b2o$bo3bobo$6b2o$6bo9$31b2o$31bo$32b3o$34bo2bo$35b3o$29b2ob3o$30b
obo2bo$30bobo2bo$29b2ob3o$35b3o$34bo2bo$34b2o11$7bo$7b2o$6bobo5$6b2o$
7b2o$6bo!

Turtle edit: and just like that, it actually costs 220G.