Thread for your unsure discoveries

For discussion of specific patterns or specific families of patterns in Conway's Game of Life, both newly-discovered and well-known.
User avatar
dvgrn
Moderator
Posts: 12030
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Thread for your unsure discoveries

Post by dvgrn »

I6_I6 wrote: August 31st, 2025, 3:46 pm
Resu wrote: August 28th, 2025, 1:10 pm Boat flipper, probably known...
I found a similar reaction that can turn a boat...
"Turning" vs. "flipping" is mostly a matter of interpretation when it comes to boats. These two are pretty much the same reaction as far as the boat is concerned, just with different outputs:

Code: Select all

x = 43, y = 15, rule = B3/S23
2b2o28b2o$2bobo27bobo$3bo29bo2$3bo29bo$2b3o27b3o$bo3bo26b3o$3ob3o$bo4b
o22b9o$b3ob3o17b2obo2bobobo2bob2o$2bo3bo17bobo4bobobo4bobo$3b3o18bobo
2bobobobobo2bobo$4bo24bob2ob2obo$26b2o11b2o$27b5o3b5o!
User avatar
I6_I6
Posts: 1044
Joined: July 26th, 2025, 8:44 pm
Location: Here, there, somewhere, anywhere, everywhere.
Contact:

Re: Thread for your unsure discoveries

Post by I6_I6 »

dvgrn wrote: September 1st, 2025, 9:38 am ...
"Turning" vs. "flipping" is mostly a matter of interpretation when it comes to boats. These two are pretty much the same reaction as far as the boat is concerned, just with different outputs:

Code: Select all

x = 43, y = 15, rule = B3/S23
2b2o28b2o$2bobo27bobo$3bo29bo2$3bo29bo$2b3o27b3o$bo3bo26b3o$3ob3o$bo4b
o22b9o$b3ob3o17b2obo2bobobo2bob2o$2bo3bo17bobo4bobobo4bobo$3b3o18bobo
2bobobobobo2bobo$4bo24bob2ob2obo$26b2o11b2o$27b5o3b5o!
You're right, I just realized that the core reaction is essentially the same.

Code: Select all

#C [[ THEME Golly ]]
x = 27, y = 15, rule = LifeHistory
8.A$A6.A.A$3A4.BA2B.B2D$3.A4.2B.2B2DB$2.2A2.3B.6B2.3B$2.20B$4.19B$4.2B
C10BD4B$4.2B2C10BD4B$4.B2C11B2D3B$4.13B2D4B$5.12BD3B.B2A$6.13B3.BA.A$
6.3B.B3.B10.A$25.2A!
RuleEdit: A .rule file editing tool
User avatar
Resu
Posts: 817
Joined: May 23rd, 2025, 10:47 am
Location: UTC+14:00
Contact:

Re: Thread for your unsure discoveries

Post by Resu »

I6_I6 wrote: September 1st, 2025, 10:13 am You're right, I just realized that the core reaction is essentially the same.
What is the smallest active region that can create that reaction?

Code: Select all

x = 31, y = 13, rule = C
8.2X2.3X.3X.X.X$8.X.X.X3.X3.X.X$8.X.X.3X.3X.X.X$8.2X2.X5.X.X.X$8.X.X.
3X.3X.3X$M2.M$4.M$M3.M$.4M$27.2M$27.M.M$29.M$29.2M! [[ AUTOSTART GPS 10 ]]
User avatar
I6_I6
Posts: 1044
Joined: July 26th, 2025, 8:44 pm
Location: Here, there, somewhere, anywhere, everywhere.
Contact:

Re: Thread for your unsure discoveries

Post by I6_I6 »

Resu wrote: September 1st, 2025, 10:40 am
I6_I6 wrote: September 1st, 2025, 10:13 am You're right, I just realized that the core reaction is essentially the same.
What is the smallest active region that can create that reaction?
This is the smallest I found, with a population of 12:

Code: Select all

x = 7, y = 8, rule = B3/S23
2b2o$2bobo$3bo2$2b3o$2b3o$o5bo$obobobo!
It's basically the front part of the LOM.

Code: Select all

#C [[ THEME Golly ]]
x = 27, y = 15, rule = LifeHistory
8.A$A6.A.A$3A4.BA2B.B2D$3.A4.2B.2B2DB$2.2A2.3B.6B2.3B$2.20B$4.19B$4.2B
C10BD4B$4.2B2C10BD4B$4.B2C11B2D3B$4.13B2D4B$5.12BD3B.B2A$6.13B3.BA.A$
6.3B.B3.B10.A$25.2A!
RuleEdit: A .rule file editing tool
hotdogPi
Moderator
Posts: 2275
Joined: August 12th, 2020, 8:22 pm

Re: Thread for your unsure discoveries

Post by hotdogPi »

I6_I6 wrote: September 1st, 2025, 2:18 pm
Resu wrote: September 1st, 2025, 10:40 am
I6_I6 wrote: September 1st, 2025, 10:13 am You're right, I just realized that the core reaction is essentially the same.
What is the smallest active region that can create that reaction?
This is the smallest I found, with a population of 12:

Code: Select all

x = 7, y = 8, rule = B3/S23
2b2o$2bobo$3bo2$2b3o$2b3o$o5bo$obobobo!
It's basically the front part of the LOM.
5 cells

Code: Select all

x = 4, y = 7, rule = B3/S23
2o$obo$bo3$3o$o2bo!
User:HotdogPi/My discoveries

Periods discovered:

All evens ≤128 except 52,58,78,82,92,94,98,104,118,122

5-15,㉕-㉛,㉟㊺,51,63,65,73,75
1㊳㊵㊹㊼㊽,54,56,72,74,80,90,92
217,240,300,486,576

Guns: 20,21,32,54,55,57,114,117,124,126
SKOPs: 32,74,76,102,196
User avatar
I6_I6
Posts: 1044
Joined: July 26th, 2025, 8:44 pm
Location: Here, there, somewhere, anywhere, everywhere.
Contact:

Re: Thread for your unsure discoveries

Post by I6_I6 »

hotdogPi wrote: September 1st, 2025, 2:37 pm 5 cells

Code: Select all

x = 4, y = 7, rule = B3/S23
2o$obo$bo3$3o$o2bo!
So we just need a pi very close to the boat to have a very quick and clean reaction! (hooray)
I still have no idea as to how this is useful in any way, but I guess that's why it's in this thread.

Code: Select all

#C [[ THEME Golly ]]
x = 27, y = 15, rule = LifeHistory
8.A$A6.A.A$3A4.BA2B.B2D$3.A4.2B.2B2DB$2.2A2.3B.6B2.3B$2.20B$4.19B$4.2B
C10BD4B$4.2B2C10BD4B$4.B2C11B2D3B$4.13B2D4B$5.12BD3B.B2A$6.13B3.BA.A$
6.3B.B3.B10.A$25.2A!
RuleEdit: A .rule file editing tool
User avatar
johamwit
Posts: 117
Joined: September 12th, 2021, 6:58 am

Re: Thread for your unsure discoveries

Post by johamwit »

I6_I6 wrote: September 1st, 2025, 2:47 pm
hotdogPi wrote: September 1st, 2025, 2:37 pm ...
So we just need a pi very close to the boat to have a very quick and clean reaction! (hooray)
...
Or an assisted glider collision, as in:

Code: Select all

x = 87, y = 43, rule = LifeHistory
5$17.2E$17.E.E$18.E3$53.E$45.E6.E.E$45.3E5.E$48.E$47.2E3$51.E$51.2E$
8.2E40.2E$9.2E$8.E20.2E32.2E$29.E.E31.E.E$16.2E11.E35.E$17.E47.2E$14.
3E$14.E!
( Shown here with an unrelated R-to-G converter ) :wink:

Edit:

Taking advantage of the collision symmetry ( and a byproduct boat-to-hook ) :

Code: Select all

x = 0, y = 0, rule = LifeHistory
9.2E$9.E.E32.2E$10.E33.E.E$45.E4$9.3D$9.D2.D31.3D$10.3D31.D2.D$45.3D
4$2E9.E$.2E7.E.E22.2E$E10.2E8.2E13.2E$21.E.E11.E20.2E$21.E21.E12.E.E$
42.E.E11.E$43.2E!
Last edited by johamwit on September 3rd, 2025, 5:33 am, edited 1 time in total.
Oh my God! -- it's full of search space!
User avatar
I6_I6
Posts: 1044
Joined: July 26th, 2025, 8:44 pm
Location: Here, there, somewhere, anywhere, everywhere.
Contact:

Re: Thread for your unsure discoveries

Post by I6_I6 »

johamwit wrote: September 1st, 2025, 10:11 pm Or an assisted glider collision, as in:

Code: Select all

x = 87, y = 43, rule = LifeHistory
5$17.2E$17.E.E$18.E3$53.E$45.E6.E.E$45.3E5.E$48.E$47.2E3$51.E$51.2E$
8.2E40.2E$9.2E$8.E20.2E32.2E$29.E.E31.E.E$16.2E11.E35.E$17.E47.2E$14.
3E$14.E!
( Shown here with an unrelated R-to-G converter ) :wink:
Reductions:

Code: Select all

x = 60, y = 21, rule = LifeHistory
9.2E34.2E$9.E.E33.E.E$10.E35.E12$2E34.2E$.2E34.2E$E20.2E13.E20.2E$21.
E.E33.E.E$11.2E8.E25.E9.E$11.2E34.E$47.E!
The blinker is not rephased.

Code: Select all

#C [[ THEME Golly ]]
x = 27, y = 15, rule = LifeHistory
8.A$A6.A.A$3A4.BA2B.B2D$3.A4.2B.2B2DB$2.2A2.3B.6B2.3B$2.20B$4.19B$4.2B
C10BD4B$4.2B2C10BD4B$4.B2C11B2D3B$4.13B2D4B$5.12BD3B.B2A$6.13B3.BA.A$
6.3B.B3.B10.A$25.2A!
RuleEdit: A .rule file editing tool
User avatar
PK22
Posts: 618
Joined: January 25th, 2025, 11:38 am
Location: United Kingdom
Contact:

Re: Thread for your unsure discoveries

Post by PK22 »

I found a Blockic 1G seed for a Big S and a block while testing a script:

Code: Select all

x = 29, y = 25, rule = B3/S23
27bo$26bo$26b3o16$b2o$b2o4$2o7b2o$2o7b2o!
The core reaction (HF + 2 blocks) seems very simple, so someone might have discovered it before.
User:PK22
Learning miscellaneous programming languages.
User avatar
qqd
Posts: 622
Joined: September 10th, 2022, 4:24 pm
Location: In a superposition of multiple different locations.

Re: Thread for your unsure discoveries

Post by qqd »

Potential dependent reflector version of iNoMed's p60 honey farm shuttle (which I independently rediscovered):

Code: Select all

x = 61, y = 47, rule = LifeHistory
10.A11.A$9.A.A9.A.A$7.3A.A9.A.3A$6.A4.A.2A3.2A.A4.A$6.ABABA.A.A3.A.A.
ABABA$7.2A.A.A7.A.A.2A4.2A$.4B2.4BABA5.2AB3AB4.A$.3B2A4.2BA3B.A3B2A4.
BA.A$2ABA2B6.4BA3BA4.B.B2A$2AB.A.BA3.9BAB2.4B$.B3.ABAB.4BA5BA7B$6.A2B
.2BA7B2ABA4B$5.7BA2BA4B5A4B$4.B2A2B.2BA2BA2B4AB2A.4B$5.2A3.4B2A5B2AB
3.4B$10.6B2AB2AB6.4B$11.5BABABAB7.4B$13.8B9.4B$13.4B3AB10.4B$14.B.6B
10.4B$15.BC5B11.3BA$15.8B11.3BA$17.7B11.3AB$19.B.4B11.4B.B$22.B3A11.
7B$23.A3B11.8B$24.A3B11.5BCB$25.4B10.6B.B$26.4B10.B3A4B$27.4B9.8B$28.
4B7.BABABA5B$29.4B6.B2AB2A6B$30.4B3.B2A5B2A4B3.2A$31.4B.2AB4A2BA2BA2B
.2B2AB$32.4B5A4BA2BA7B$33.4BAB2A7BA2B.2BA$33.7BA5BA4B.BABA3.B$33.4B2.
BA9B3.AB.A.B2A$31.2AB.B4.A3BA4B6.2BAB2A$30.A.AB4.2A3BA.3BA2B4.2A3B$
30.A4.B3AB2A5.ABA4B2.4B$29.2A4.2A.A.A7.A.A.2A$34.ABABA.A.A3.A.A.ABABA
$34.A4.A.2A3.2A.A4.A$35.3A.A9.A.3A$37.A.A9.A.A$38.A11.A!
[[ STOP 60 ]]
iNoMed wrote: November 18th, 2022, 5:18 am Here's a pair of p60 HFH oscillators using the recently-found honey farm catalyst.
...
and the other is an entirely manually-found mechanism:

Code: Select all

x = 33, y = 15, rule = B3/S23
5b2o6b2o11b2o$5bo2bo4b2o9bo2bo2$14bo$5bob2o5bo9b2obo$2o2bobo7bo11bobo
2b2o$o4bo21bo4bo$4bo7bo7bo7bo$bo2bo6bobo5bobo6bo2bo$7b2obobo7bobob2o$
6bobobobobo3bobobobobo$6bo4bob2o3b2obo4bo$7b3obo9bob3o$9bobo9bobo$10bo
11bo!
EDIT: MWSS reaction based on the above (not repeatable):

Code: Select all

x = 61, y = 47, rule = LifeHistory
10.A11.A$9.A.A9.A.A$7.3A.A9.A.3A$6.A4.A.2A3.2A.A4.A$6.ABABA.A.A3.A.A.
ABABA$7.2A.A.A7.A.A.2A$.4B2.4BABA5.2AB3AB$.3B2A4.2BA3B.A3B2A4.B$2ABA2B
6.4BA3BA4.B.B$2AB.A.BA3.9BAB2.4B$.B3.ABAB.4BA5BA7B.2A$6.A2B.2BA7B2ABA
4B.2A$5.7BA2BA4B5A4B$4.B2A2B.2BA2BA2B4AB2A.4B$5.2A3.4B2A5B2AB3.4B$10.
6B2AB2AB6.4B$11.5BABABAB7.4B$13.8B9.4B$13.4B3AB10.4B$14.B.6B10.4B$15.
BC3B3A10.4B$15.5BA2B11.4B$17.4BA2B11.4B$19.B.4B11.4B.B$22.4B11.2BA4B$
23.4B11.2BA5B$24.4B10.3A3BCB$25.4B10.6B.B$26.4B10.B3A4B$27.4B9.8B$28.
4B7.BABABA5B$29.4B6.B2AB2A6B$30.4B3.B2A5B2A4B3.2A$31.4B.2AB4A2BA2BA2B
.2B2AB$32.4B5A4BA2BA7B$30.2A.4BAB2A7BA2B.2BA$30.2A.7BA5BA4B.BABA3.B$33.
4B2.BA9B3.AB.A.B2A$33.B.B4.A3BA4B6.2BAB2A$33.B4.2A3BA.3BA2B4.2A3B$35.
B3AB2A5.ABA4B2.4B$35.2A.A.A7.A.A.2A$34.ABABA.A.A3.A.A.ABABA$34.A4.A.2A
3.2A.A4.A$35.3A.A9.A.3A$37.A.A9.A.A$38.A11.A!
Currently writing a utility in Lua that may be helpful for faster manual pattern manipulation.
User avatar
PK22
Posts: 618
Joined: January 25th, 2025, 11:38 am
Location: United Kingdom
Contact:

Re: Thread for your unsure discoveries

Post by PK22 »

While testing a script to find Blonkic (using only blocks and blinkers) 1G seeds, I found this reaction:

Code: Select all

#C Blonk-tie descendant + 2 blocks > HWSS + beehive
x = 23, y = 30, rule = B3/S23
21bo$20bo$20b3o15$2bo$2bo$2bo4$2o$2o4$2b2o$2b2o!
Adding an extra block (placement found manually) leads to a 4 object 1G seed for a HWSS:

Code: Select all

x = 23, y = 34, rule = B3/S23
$20bobo$20b2o$21bo15$b3o5$2o$2o4$2b2o$2b2o3$2o$2o!
I believe it may be the smallest known HWSS seed - the one listed on the wiki (also with 4 objects) has a population of 21 cells and bounding box 14x14 (196), while my example has a population of 15 cells and bounding box 4x16 (64) in its minimum phase. It also has the advantage of being Blonkic.

Is it the smallest known HWSS seed, and is it new?
EDIT: octo3obj has one with bounding box 10x7 (70) and population 18, but it is still larger than my example:

Code: Select all

x = 14, y = 7, rule = B3/S23
7b2o$6bo2bo$7b2o2b3o$2o9bo$2o10bo$4b2ob2o$4b2ob2o!
User:PK22
Learning miscellaneous programming languages.
User avatar
rabbit
Posts: 244
Joined: March 4th, 2024, 6:00 am
Location: Stuck inside a magician's hat. it hurts so bad

Re: Thread for your unsure discoveries

Post by rabbit »

PK22 wrote: September 4th, 2025, 11:44 am Adding an extra block (placement found manually) leads to a 4 object 1G seed for a HWSS:

Code: Select all

x = 23, y = 34, rule = B3/S23
$20bobo$20b2o$21bo15$b3o5$2o$2o4$2b2o$2b2o3$2o$2o!
Using a blinker to suppress the ash instead of a block leads to a population reduction to 14 cells, while keeping the minimum phase's bounding box the same:

Code: Select all

x = 11, y = 21, rule = B3/S23
8bobo$8b2o$9bo3$b3o5$2o$2o4$2b2o$2b2o2$o$o$o!
That's the bunny.
Check out my OCA exploration script, crule.lua!
User avatar
dvgrn
Moderator
Posts: 12030
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Thread for your unsure discoveries

Post by dvgrn »

PK22 wrote: September 4th, 2025, 11:44 am Is it the smallest known HWSS seed, and is it new?
EDIT: octo3obj has one with bounding box 10x7 (70) and population 18, but it is still larger than my example.
It's probably new, since collision enumerations haven't been done for long thin bounding boxes, only for square bounding boxes like 11x11 or 12x12.

There are three octo3obj results for a clean HWSS, and one of them has a single-phase bounding box of 10x6 = 60 cells, slightly smaller than 4x16=64, and also a 14-cell population:

Code: Select all

x = 11, y = 8, rule = B3/S23
4b2o$4b2o3bo$9bo$3o6bo$5b2o$5b2o2b2o$8b2o$10bo!
The all-phases bounding box is 10x7, but the all-phases bounding box of the new 14-cell HWSS seed is considerably bigger than that.
User avatar
qqd
Posts: 622
Joined: September 10th, 2022, 4:24 pm
Location: In a superposition of multiple different locations.

Re: Thread for your unsure discoveries

Post by qqd »

Has it been proven that any m x n rectangle completely populated by live cells cannot be a Garden of Eden? If not, here is a pattern that can show you how to prove it (Technically only works if m>1 and n>1, but it has already been proven that any 1xn pattern must have at least one parent, Theorem 1.2 in the book):

Code: Select all

x = 171, y = 12, rule = B3/S23
120b2ob2ob2o12b2ob2ob2obo9bob2ob2ob2obo$o2bo2bo13bo2bo2bo2bo11bo2bo2bo
2bo9bo2bo2bo13bo2bo2bo2bo11bo2bo2bo2bo$2ob2ob2o12b2ob2ob2obo9bob2ob2ob
2obo9b2ob2ob2o12b2ob2ob2obo9bob2ob2ob2obo9bo2bo2bo13bo2bo2bo2bo11bo2bo
2bo2bo$120b2ob2ob2o12b2ob2ob2obo9bob2ob2ob2obo$o2bo2bo13bo2bo2bo2bo11b
o2bo2bo2bo9bo2bo2bo13bo2bo2bo2bo11bo2bo2bo2bo$2ob2ob2o12b2ob2ob2obo9bo
b2ob2ob2obo9b2ob2ob2o12b2ob2ob2obo9bob2ob2ob2obo9bo2bo2bo13bo2bo2bo2bo
11bo2bo2bo2bo$120b2ob2ob2o12b2ob2ob2obo9bob2ob2ob2obo$o2bo2bo13bo2bo2b
o2bo11bo2bo2bo2bo9bo2bo2bo13bo2bo2bo2bo11bo2bo2bo2bo$2ob2ob2o12b2ob2ob
2obo9bob2ob2ob2obo9b2ob2ob2o12b2ob2ob2obo9bob2ob2ob2obo9bo2bo2bo13bo2b
o2bo2bo11bo2bo2bo2bo$120b2ob2ob2o12b2ob2ob2obo9bob2ob2ob2obo$60bo2bo2b
o13bo2bo2bo2bo11bo2bo2bo2bo$120bo2bo2bo13bo2bo2bo2bo11bo2bo2bo2bo!
[[ STOP 1 ]]
These are nine predecessors to completely filled rectangles where m mod 3 = 2,0 and 1 and n mod 3 = 2,0 and 1.
Last edited by qqd on October 5th, 2025, 1:09 am, edited 1 time in total.
Currently writing a utility in Lua that may be helpful for faster manual pattern manipulation.
User avatar
PK22
Posts: 618
Joined: January 25th, 2025, 11:38 am
Location: United Kingdom
Contact:

Re: Thread for your unsure discoveries

Post by PK22 »

I found two integer constructions which are smaller than the ones listed on the wiki:
Mirrored table (737261 < 33108318):

Code: Select all

x = 21, y = 5, rule = B3/S23
3ob3ob3ob3ob3obo$2bo3bo3bo3bobo3bo$2bob3o3bob3ob3obo$2bo3bo3bobo3bobob
o$2bob3o3bob3ob3obo!
Dead spark coil (796143 < 4108303132):

Code: Select all

x = 21, y = 5, rule = B3/S23
3ob3ob3obobobob3o$2bobobobo3bobobo3bo$2bob3ob3obob3ob3o$2bo3bobobobo3b
o3bo$2bob3ob3obo3bob3o!
It is possible they might have been discovered before, given that they only have six digits each.
User:PK22
Learning miscellaneous programming languages.
User avatar
qqd
Posts: 622
Joined: September 10th, 2022, 4:24 pm
Location: In a superposition of multiple different locations.

Re: Thread for your unsure discoveries

Post by qqd »

Thessalonic 2x pulse divider, using an interesting reaction between a Herschel and a block (The preceding BFx59H is necessary for this to work):

Code: Select all

x = 32, y = 29, rule = B3/S23
31bo$29b3o$28bo$28b2o3$15b2o$15b2o$2o5b2o$obo5bo18b2o$2bo5bobo16b2o$2b
2o5b2o6$bo3b2o15bo$b2o2b2o14bobo$2b2o17bo2bo$b2o20bo5$26b2o$26bo$27b3o
$29bo!
Currently writing a utility in Lua that may be helpful for faster manual pattern manipulation.
User avatar
b-engine
Posts: 3762
Joined: October 26th, 2023, 4:11 am
Location: Somewhere on where Earth At
Contact:

Re: Thread for your unsure discoveries

Post by b-engine »

qqd wrote: September 18th, 2025, 11:04 am Thessalonic 2x pulse divider, using an interesting reaction between a Herschel and a block (The preceding BFx59H is necessary for this to work):
In this case a snake is suffice for the BFx59H:

Code: Select all

x = 70, y = 29, rule = B3/S23
30bo38bo$28b3o36b3o$27bo38bo$27b2o37b2o3$14b2o37b2o$14b2o37b2o$b2o3b2o
32b2o3b2o$2bo4bo18b2o13bo4bo18b2o$bo5bobo16b2o12bo5bobo16b2o$b2o5b2o30b
2o5b2o6$o3b2o15bo17bo3b2o15bo$2o2b2o14bobo16b2o2b2o14bobo$b2o17bo2bo16b
2o17bo2bo$2o20bo16b2o19b2o5$25b2o37b2o$25bo38bo$26b3o36b3o$28bo38bo!
User avatar
qqd
Posts: 622
Joined: September 10th, 2022, 4:24 pm
Location: In a superposition of multiple different locations.

Re: Thread for your unsure discoveries

Post by qqd »

b-engine wrote: September 21st, 2025, 3:37 am
qqd wrote: September 18th, 2025, 11:04 am Thessalonic 2x pulse divider, using an interesting reaction between a Herschel and a block (The preceding BFx59H is necessary for this to work):
In this case a snake is suffice for the BFx59H:

Code: Select all

x = 70, y = 29, rule = B3/S23
30bo38bo$28b3o36b3o$27bo38bo$27b2o37b2o3$14b2o37b2o$14b2o37b2o$b2o3b2o
32b2o3b2o$2bo4bo18b2o13bo4bo18b2o$bo5bobo16b2o12bo5bobo16b2o$b2o5b2o30b
2o5b2o6$o3b2o15bo17bo3b2o15bo$2o2b2o14bobo16b2o2b2o14bobo$b2o17bo2bo16b
2o17bo2bo$2o20bo16b2o19b2o5$25b2o37b2o$25bo38bo$26b3o36b3o$28bo38bo!
I just realized one of the blocks was unneccesary:

Code: Select all

x = 70, y = 29, rule = B3/S23
30bo38bo$28b3o36b3o$27bo38bo$27b2o37b2o5$b2o3b2o32b2o3b2o$2bo4bo18b2o
13bo4bo18b2o$bo5bobo16b2o12bo5bobo16b2o$b2o5b2o30b2o5b2o6$o3b2o15bo17b
o3b2o15bo$2o2b2o14bobo16b2o2b2o14bobo$b2o17bo2bo16b2o17bo2bo$2o20bo16b
2o19b2o5$25b2o37b2o$25bo38bo$26b3o36b3o$28bo38bo!
Currently writing a utility in Lua that may be helpful for faster manual pattern manipulation.
User avatar
PK22
Posts: 618
Joined: January 25th, 2025, 11:38 am
Location: United Kingdom
Contact:

Re: Thread for your unsure discoveries

Post by PK22 »

This might have been discovered before due to its simplicity, but a glider colliding with a beehive and honey farm produces a pseudo still life made up of 2 long snakes:

Code: Select all

x = 26, y = 23, rule = B3/S23
6bo$7bo$5b3o8$6bo$5bobo$5bobo$6bo2$b2o7b2o12bo$o2bo5bo2bo10bobo$b2o7b
2o11bobo$24bo$6bo$5bobo$5bobo$6bo!
Maybe this could lead to a 1G reduction if you could clean up the debris in 3G.
User:PK22
Learning miscellaneous programming languages.
User avatar
KtT
Posts: 609
Joined: December 17th, 2022, 9:29 am

Re: Thread for your unsure discoveries

Post by KtT »

PK22 wrote: September 25th, 2025, 11:24 am This might have been discovered before due to its simplicity, but a glider colliding with a beehive and honey farm produces a pseudo still life made up of 2 long snakes:

Code: Select all

x = 26, y = 23, rule = B3/S23
6bo$7bo$5b3o8$6bo$5bobo$5bobo$6bo2$b2o7b2o12bo$o2bo5bo2bo10bobo$b2o7b
2o11bobo$24bo$6bo$5bobo$5bobo$6bo!
Maybe this could lead to a 1G reduction if you could clean up the debris in 3G.
I found the way to clean up the debris with 3 gliders:

Code: Select all

x = 73, y = 53, rule = B3/S23
26bobo$27b2o$27bo2$obo$b2o$bo7$18bobo$19b2o$19bo15$53bo$54bo$52b3o8$53b
o$52bobo$52bobo$53bo2$48b2o7b2o12bo$47bo2bo5bo2bo10bobo$48b2o7b2o11bo
bo$71bo$53bo$52bobo$52bobo$53bo!
And I found if 3 blocks attached to the pattern, then it will be an 8-objects LWSS seed:

Code: Select all

x = 31, y = 30, rule = B3/S23
11bo$12bo$10b3o8$11bo16b2o$10bobo15b2o$10bobo$11bo2$6b2o7b2o12bo$5bo2b
o5bo2bo10bobo$6b2o7b2o11bobo$29bo$2o9bo$2o8bobo$10bobo$11bo6$19b2o$19b
2o!
User avatar
qqd
Posts: 622
Joined: September 10th, 2022, 4:24 pm
Location: In a superposition of multiple different locations.

Re: Thread for your unsure discoveries

Post by qqd »

Not sure if this periodic pi to G is going to be useful, since it has low input clearance (based on a reaction in a p260 oscillator):

Code: Select all

x = 15, y = 24, rule = B3/S23
10bobo$9bo$10bo2bo$5b2o3bobobo$5b2o4bo2bo$12b2o3$3b3o4b2o$5bo4b2o$3b3o
3$10b2o$10b2o3$13b2o$9b2o2b2o$9b2o$2b2o$3bo$3o$o!
Currently writing a utility in Lua that may be helpful for faster manual pattern manipulation.
User avatar
ThePlayzr
Posts: 824
Joined: April 19th, 2025, 1:33 am
Location: Australia
Contact:

Re: Thread for your unsure discoveries

Post by ThePlayzr »

I think this still life cannot be destroyed with 2 gliders, but I am not quite sure.

Code: Select all

x = 91, y = 91, rule = B3/S23
bo$obo$bobo$2bobo$3bobo$4bobo$5bobo$6bobo$7bobo$8bobo$9bobo$10bobo$11b
obo$12bobo$13bobo$14bobo$15bobo$16bobo$17bobo$18bobo$19bobo$20bobo$21b
obo$22bobo$23bobo$24bobo$25bobo$26bobo$27bobo$28bobo$29bobo$30bobo$31b
obo$32bobo$33bobo$34bobo$35bobo$36bobo$37bobo$38bobo$39bobo$40bobo$41b
obo$42bobo$43bobo$44bobo$45bobo$46bobo$47bobo$48bobo$49bobo$50bobo$51b
obo$52bobo$53bobo$54bobo$55bobo$56bobo$57bobo$58bobo$59bobo$60bobo$61b
obo$62bobo$63bobo$64bobo$65bobo$66bobo$67bobo$68bobo$69bobo$70bobo$71b
obo$72bobo$73bobo$74bobo$75bobo$76bobo$77bobo$78bobo$79bobo$80bobo$81b
obo$82bobo$83bobo$84bobo$85bobo$86bobo$87bobo$88bobo$89bo!
If it can't, then we need to update the wiki, because it says somewhere that there are no known still lives that can't be destroyed with at most 2 gliders (and reduce the still life).
Please visit my rules (found on my wiki page) and contribute!
User:ThePlayzr
I have LLS and qfind.
Working on a jslife-like Travelling Ts pattern collection.
User avatar
qqd
Posts: 622
Joined: September 10th, 2022, 4:24 pm
Location: In a superposition of multiple different locations.

Re: Thread for your unsure discoveries

Post by qqd »

Attempting a smaller (and potentially faster) HWSS to G:

Code: Select all

x = 32, y = 27, rule = B3/S23
9bo$8bobo$6b3obo$5bo4bob2o$5bobobobobo$2o4b2obobo$bo8bobo14bo$bobo7bo
9b2o3bobo$2b2o17b2o3bobo$25b2ob3o$31bo$25b2ob3o$25b2obo4$9b6o6b2o$8bo
5bo6b2o$14bo$8bo4bo$10b2o5$17b2o$17b2o!
Currently writing a utility in Lua that may be helpful for faster manual pattern manipulation.
User avatar
PK22
Posts: 618
Joined: January 25th, 2025, 11:38 am
Location: United Kingdom
Contact:

Re: Thread for your unsure discoveries

Post by PK22 »

Small p240 puffer based on a combination of a Coe ship and Schick engine:

Code: Select all

x = 15, y = 10, rule = B3/S23
2bo6b3o$b3o5b3o$10bo$bo$b2o4bobobobo$b3o2bo7bo$b3o2bo7bo$ob2o2bo2bobo
2bo$3o3b3o3b3o$bo!
User:PK22
Learning miscellaneous programming languages.
User avatar
qqd
Posts: 622
Joined: September 10th, 2022, 4:24 pm
Location: In a superposition of multiple different locations.

Re: Thread for your unsure discoveries

Post by qqd »

4.66920 16091 wrote: September 30th, 2025, 2:18 am Is this HFx217H known?

Code: Select all

x = 56, y = 58, rule = LifeSuper
38.B$37.3D$37.BDB$36.2B3D$36.5B$36.6B$36.6B$36.5B$35.6B$36.6B$35.7B$
35.6B$35.6B$35.6B$34.8B$35.8B$34.9B$34.9B$34.10B$34.5B2A3B$34.5B2A4B$
34.11B$21.A12.12BA.2A$19.3A8.A4.9B2.2A.A$18.A11.3A3.7B$18.2A13.A2.6B$
16.4B12.2A2.6B$15.2AB14.11B$15.A.AB15.8B$16.2AB14.11B$16.3B12.14B$16.
4B9.15B$17.4B3.3B.16B$5.2A8.29B$4.BA.A35B$4.2B2A34B$2.2AB.37B5.2A$.A.
AB4.34B5.A.A.2A.A$.A7.34B7.A.A.2A$2A9.33B5.B2A$9.3B.32B5.2B$7.A.2A2.
2B.30B3.2B$5.3AB2A5.31B.B2AB$4.A4.B7.32B2A$5.3A.2A6.20B2.9B.B$7.A.A7.
16B7.7B$7.A.A8.12B.B6.3B.B2A2B7.A$8.A7.4B.4B3CB9.2A2.B2AB3.2BA.2BA$
16.2A4.4BC2B10.A2.4B3.BA4BA$17.A4.2B3C2B7.3A4.2B2.3BA2B.A$14.3A6.6B7.
A7.2B.2BAB2A$14.A8.7B13.B2A2B3A$25.B.4B13.2AB.B2A$28.4B13.B$29.4B$30.
4B$31.4B$32.4B!
H to block, one of the longest known without using gliders for cleanup or catalysis, using two impossibly close eaters and a beehive keeper:

Code: Select all

x = 30, y = 58, rule = LifeHistory
12.A$12.3A$15.A$14.2A$8.A$7.A.A7.2A$7.A.A6.A.A$5.3A.2A5.2A$4.A$5.3A.
2A$7.A.2A4$2A$.A21.2A$.A.A18.A.A$2.2A18.2A$26.2A$26.A.A$28.A$28.2A3$
2.C$2.C.C$2.3C$4.C2$25.2A$25.A$23.A.A$23.2A$27.2A$5.2E20.A$4.E2.E17.A
.A$5.2E18.2A13$2A15.2D2.2A$2A15.2D2.2A4$3.2A$4.A$.3A$.A!
There is not much room to perturb the honey farm + beehive combination that emerges, and this is the most useful thing I could find, for some definition of 'useful'. If you remove the bottom most eater, you get an H to traffic light instead, where the rightmost block catalyses in two directions.
Currently writing a utility in Lua that may be helpful for faster manual pattern manipulation.
Post Reply