ConwayLife.com

Why not a drifter loop? One of these was packaged with Golly. Drifter loops cost a lot more cells than phase-shifting reflector loops, but since the latter don't hit some of the periods below 1000 I've gone ahead and implemented them . Here's the p25 loop: x = 640, y = 64, rule = B2-a/S12 38bo4b2o1...

c/17 wickstretcher in Just Friends: x = 15, y = 17, rule = B2-a/S12 11b2o$13b2o2$13b2o$13o2$b2o2$3b3o$6bo$2b4o2$b12o$13b2o2$13b2o$11b2o! c/4 line mover (needs another line for support): x = 44, y = 19, rule = B2-a/S12 37bo3bo$28b2o3bo2b5o$30bo$28bo2bobo2b3obo$2o26b2obo6bo4bo$o37bobo$24b 16o$b2obo$23...

All periods that are multiples of 26 now have non-trivial population-minimised oscillator implementations in Skopje: x = 154, y = 115, rule = B2-a/S12 8b5o26b5o30b5o2$9bobo63bobo$9bobo63bobo$10bo29bobo33bo$41bo$41bo2$11bo 65bo$12b2o8bo2bo52b2o$22bo2bo14bo27bo$20b2obobo13bo14bo13bo2b3o19bo$o 2bo18bo2...

An edgy 7-glider mold insertion extracted from Catagolue soups yields successful syntheses of mold on 34P14 shuttle/Jason's p33/41P7.2: x = 203, y = 58, rule = LifeHistory .A72.A85.A$2.A72.A85.A$3A70.3A83.3A8$39.A72.A85.A$38.A72.A85.A$38.3A 70.3A83.3A3$22.A72.A85.A$23.2A71.2A84.2A$22.2A71.2A84.2A2$3...

There's a bunch of SKOPs to be had with the new p32 and a sufficiently high-clearance domino sparker. (The fumarole doesn't work.) x = 175, y = 33, rule = LifeHistory 26.AB46.AB39.AB52.AB$25.B2A45.B2A38.B2A51.B2A$25.2ABAB43.2ABAB36.2ABA B23.2A7.2A15.2ABAB$.2A21.A2B3A42.A2B3A35.A2B3A22.B2A2B3.2B2AB13...

With a different spark interaction site, a 62-cell variant exists using a caterer: x = 32, y = 45, rule = B3/S23 b2o5b2o$b2o5b2o16$2obo3bob2o$o2bo3bo2bo$b3o3b3o6$13b3o14b2o$13bo3bo12b 2o$13bo4bo$14bo3bo2$14bo3bo$13bo4bo$13bo3bo12b2o$13b3o14b2o4$2b3o$6bo$ bo4bo$5bo$2b2o$2bo$2bo$2bo! Which immediatel...

While toying around with banana sparks I found a new SKOP276, unix on 50P92.1 at 66 cells:

p136 glider shuttle (the smallest multiple-of-8 period for which this construction is the current SKOP) in 40 gliders: x = 499, y = 53, rule = B3/S23 50bo$48bobo$49b2o$obo50bobo$b2o39b3o8b2o270bo24bo2bo$bo4bo37bo9bo 57b2o39b2o48b2o67b2o49b2o24bo141b4o$5bo37bo68bo40bo49bo68bo51b2o 21b2o64bob2ob2o67b2...

The new SKOP32 is almost too easy to build, in 14 gliders: x = 254, y = 22, rule = B3/S23 50bo$51b2o69bo61bo60bo$50b2o68b3o59b3o58b3o$5bobo52b2o57bo8bo52bo 8b2o50bo$bo3b2o93bo18b2o7b2o51b2o59b2o7bo$2bo3bo41b2o9bo3bo34bobo 26b2obo58bo3bo$3o44bobo8bo4bo35b2o25bo2b3o56bo4bo55b3obo$49bo7bobo bo63bobobo5...

Bit pattern generator: x = 79, y = 24, rule = Trine CA2.B.3B3.2B3.B2.3B.3B.CAB.2B2.B2.B4.CA.CAB.B2.B.CAB.2B3.B2.3B2.B2.2B $B.C.A2.A4.A.A.A.A2.A3.A2.B3.A.A.2A.A3.B3.B3.2A.A.B3.A.A.A.A2.A2.A.A. A.A$AB2.C2.C4.2C2.3C2.C3.C2.AB2.2C2.C.2C3.A.C.AB2.C.2C.AB2.2C2.3C2.C 2.C.C.2C$B.C.B2.B4.B3.B.B2.B3.B2.B3.B.B...

p128 wing hassler in 48 gliders: x = 716, y = 60, rule = B3/S23 417bobo$418b2o$418bo2$137bobo286bobo$138b2o19bo266b2o$95bobo40bo 10bo7b2o268bo$96b2o50bo9b2o$96bo51b3o385b2o$201b4o65b4o57b4o66b2o 67b2o63b3o83b2o66b2o$104bobo93bo7b2o59bo7b2o51bo7b2o57b4o4b2o59b2o bo2bob2o58bobo2bo2b2o73b2obo2bob2o58b2...

This script allows "typing" text in Golly, suitable for annotation or other purposes: import golly as g letters = g.parse("""2bobobo2bobo4bo3b2o5bo4bo2bobo4bo20bo$2bobobo2bobo3b4ob2o2bobobo3bobo 3bobobobo3bo14bo$2bo5b5obo8bo2bobo5bo3bo2b3o4bo13bo$2bo6bobo3b3o4bo4bo 6bo3bobobobob5o4b3o4bo$2bo5b5o5bo2...

A new p8 180° reflector by A for awesome and Jormugant, paired with Elkies' reflector, yields 109 cells for 104+16n and 120 cells for 112+16n. The old records were 120 and 121 respectively: x = 33, y = 35, rule = B3/S23 2b2o2$o3bo$o4bo9bo2bobo$2bobobo6b2obob3o$3bobobo6bo6bo$4bo4bo3b2o5bo$ 5bo3bo$14b...

A galactic reduction: x = 29, y = 28, rule = LifeHistory 21.A2BA.2B$19.BA4B.2B$17.B2A6B.B$15.A6B2ABA$15.3BA.2BA.BAB$15.3B3ABA3B A$15.A5B.5BA$9.C6.A3BAB3A3B$10.2C4.BAB.A2B.A3B$9.2C6.AB2A6BA$17.6B2AB $18.4BAB$17.BA2BA$17.B.B$18.B2.3B$21.BA2B$17.B.2B3A$17.3B3ABA$20.BA2B .AB$11.2A9.A3BA2B$12.A10.AB3AB$...

At Redstoneboi's request I've annotated the "basics" file, including an explanation of the triangle rule: x = 163, y = 225, rule = Trine B3.2B2.2B2.CAB.CA2.CAB2.CA.CAB2.CA2.CA2.B2.2B$A3.A.A.A.A.B3.B.C.B3.B 3.B3.B3.B3.A.A.A.A$C3.2C2.2C2.AB2.A.B.AB2.A3.AB3.A3.A2.C.C.2C$B3.B3.B .B.B3.B.C.B3.C3.B5.C3.C....

While I was busy playing with my fewer-state, richer and crazily scrunchier rule Trine , I noticed that the crossover can lose a column: x = 27, y = 6, rule = Wireworld5 .C$G.D5.B$.F.FBCDABCDABCDABCDABCDABCD$7.AB$2.DFBCDABCDABCDABCDABCDABC D$2.C5.B! However the rule proposed in this thread does not ...

I've joined the 32 bits of memory to an index decoder and three demultiplexers to produce a full register unit without any additional crosses (i.e. all of them are welded to gates in the memory cells): x = 46, y = 28, rule = Trine 16.CBA7.A$15.A3.C6.B$15.B3.B6.C$15.C3.A6.A$16.ABC7.B3$15.A8.A$16.DEB ...

Fully readable and writeable 32-bit register together with its repeating unit (remember that all three colours are equal in Trine): x = 384, y = 30, rule = Trine 85.CABEFC$91.A$89.BC$85.ABCABC$85.C3.A.A$87.2C$86.B2.C$85.AB2.B$87.CA 2.B$87.CA2.A$86.B2.BC$85.A$86.BCFDAB5$85.CABEFC.BCADEB.ABCFDA.CABEFC...

The three-predecessor and crossover transitions of Trine suffice to make a neatly rectangular pixel display: x = 56, y = 40, rule = Trine 41.15A$41.15C$41.15B$41.15A$41.15C$41.15B$41.15A$41.15C$41.15B$41.15A $41.15C$41.15B$41.15A$41.15C$41.15B$41.15A$41.15C$41.15B$41.15A$41. 15C$41.15B$41.15A$41.15C...

Stator-minimised p53 from my all-symmetry apgsearch set a while back, that I forgot to post: x = 31, y = 31, rule = B2-a/S12 15bo$11b3obob3o$15bo$12bo5bo$12bobobobo$12bo5bo$9b2obo5bob2o2$12bo5bo $6bo17bo$6bo17bo$bo27bo$bob4obo13bob4obo$bo27bo$4bo21bo$3o25b3o$4bo21b o$bo27bo$bob4obo13bob4obo$bo27bo$6...

I've added the "triangle rule" which essentially says that in certain situations involving two predecessor cells, the presence of a successor cell disables the effect of one of them. This makes the orthogonal 180° collision usable; I've assigned XOR (and XNOR for the colour-advancing variant, which ...

In the Wireworld5 thread I saw this: I love the concept! I wonder how you arrived on five as the requisite number of states; is four not enough? After a sleepless night I managed to show that three is enough if you drop the symmetry from "outer-totalistic" to "isotropic", apply the Wireworld Modern ...

7-cell infinite growth in the form of a p296 gun: I think this is the smallest it can go.

76 is divisible by 4. Mold on p76 shouldn't be a period multiplier. Yes, that was a slip – but what was already in Skopje is this: x = 30, y = 41, rule = B3/S23 27bo2$25b3obo$24bob2o$24b2obo$3b2o17bob3o$3bobo$2obobo18bo$obobo$2bo$ b2o$3o$3o8bobo$3o8bobo$3o8b3o$b2o$2bo$obobo$2obobo$3bobo$3b2o5$20b2o...

Are these the smallest LCMs of 76×2, 3, 4, 5, 7, 11, and 15? (94, 98, 103, 100, 116, 120, 94) All these oscillators have been committed (no distinction between LCMs and other oscillators now in Skopje) except 76×2 (because mold on p76 also has 94 cells and was already there) and 76×15 because of th...