Thread For Your Accidental Discoveries

For discussion of specific patterns or specific families of patterns, both newly-discovered and well-known.
User avatar
Extrementhusiast
Posts: 1966
Joined: June 16th, 2009, 11:24 pm
Location: USA

Re: Thread For Your Accidental Discoveries

Post by Extrementhusiast » August 28th, 2015, 3:45 pm

An unexpected result from trying to build an R-to-X:

Code: Select all

x = 30, y = 18, rule = LifeHistory
4.2A$5.A$5.A.A7.2A$6.2A6.A.A$2A13.A$.A$.A.A$2.2A2$23.2C$22.2C$23.C3$
26.2A$26.A$27.3A$29.A!
I Like My Heisenburps! (and others)

User avatar
gmc_nxtman
Posts: 1150
Joined: May 26th, 2015, 7:20 pm

Re: Thread For Your Accidental Discoveries

Post by gmc_nxtman » August 28th, 2015, 8:52 pm

Very cool! Here is a p30 oscillator I found based on an old hassling reaction found a long time ago.

Code: Select all

x = 46, y = 28, rule = B3/S23
2o$bo$bobo8b2o$2b2o7b3o$8bob2o9b2o$8bo2bo4bo5bo$8bob2o5bo$11b3o3bo3bo$
12b2o5bo$22b2o$21bo2bo2b2o$21bo2bo2b2o$22b2o3$22b2o$17b2o2bo2bo$17b2o
2bo2bo$22b2o$26bo5b2o$24bo3bo3b3o$28bo5b2obo$23bo5bo4bo2bo$23b2o9b2obo
$32b3o7b2o$32b2o8bobo$44bo$44b2o!
Also, I really think that a spaceship is possible out of this. (5,14x)c/102 Herschel reaction, (it is also a [H5,6xB]c/82 reaction) triplicating the input glider into an extra block and two gliders, one forward and backward:

Code: Select all

x = 310, y = 261, rule = B3/S23
2bo$obo$b2o37$308bo$307bo$307b3o38$53bo$51bobo$52b2o37$257bo$256bo$
256b3o38$104bo$102bobo$103b2o65b2o$170bo$168b3o4$168bobobo3bobo5bo3bo$
163bo$172bo5bo5bo3bo$161bobobo$172bo5bo5bobobo$163bo$172bo3bobobo7bo
25$206bo$205bo$205b3o38$155bo$153bobo$154b2o19$164b3o$165bo$165b3o!
It may not be oblique, but I would love to work on it. It has probably been discovered before. I might try to find a helix (is there such a search program?)

EDIT: I found this, but all It can be used for is an oscillator. (H17,2Lx73H, or H9,2Lx53B)

Code: Select all

x = 29, y = 25, rule = B3/S23
26bo$26bobo$26b2o16$2bo$2bo$3o$o$16b3o$17bo$15b3o!
Last edited by gmc_nxtman on August 28th, 2015, 11:58 pm, edited 1 time in total.

User avatar
biggiemac
Posts: 515
Joined: September 17th, 2014, 12:21 am
Location: California, USA

Re: Thread For Your Accidental Discoveries

Post by biggiemac » August 28th, 2015, 11:31 pm

The helix program written by codeholic for caterpillars is here. The climb is slow enough that the helix should be plenty short at N=1.

One considerable difficulty with building a caterpillar with that is that the streams need to be synchronized from left and right. That means that each climber puts a large demand on the structure. Most important in design would probably be finding a single reburnable element made of a collection of these climbers.
Physics: sophistication from simplicity.

User avatar
Extrementhusiast
Posts: 1966
Joined: June 16th, 2009, 11:24 pm
Location: USA

Re: Thread For Your Accidental Discoveries

Post by Extrementhusiast » September 2nd, 2015, 11:41 am

An unexpected block catalyst for an R-pentomino:

Code: Select all

x = 9, y = 3, rule = LifeHistory
7.C$2A4.3C$2A4.C!
I Like My Heisenburps! (and others)

Jackk
Posts: 116
Joined: March 13th, 2012, 3:49 pm

Re: Thread For Your Accidental Discoveries

Post by Jackk » September 2nd, 2015, 12:08 pm

An R -> Pi using EE's post, with a sacrificial block that might be able to be restored:

Code: Select all

x = 15, y = 22, rule = LifeHistory
2.2CB$2.2CB.2B$2.7B$2.3D5B$.D3BD4B$.2DB2D7B$2.11B$3.11B$3.11B$3.10B$
3.10B$2.11B$3.9B$3.9B$3.10B$2.12B$3.11B$2.11B2A$.6BC5B2A$2A4B3C3B.B$
2AB.2BC5B$.B3.6B!
Unfortunately the R looks a bit cramped.

User avatar
gmc_nxtman
Posts: 1150
Joined: May 26th, 2015, 7:20 pm

Re: Thread For Your Accidental Discoveries

Post by gmc_nxtman » September 19th, 2015, 11:09 pm

Wow, found this by hand! I noticed the two herschels make a unix + sparks that mangle the unix, and then found that gliders could suppress the sparks.

Code: Select all

x = 14, y = 14, rule = B3/S23
7bo$7b3o$7bobo$9bo4$3o9bo$bo9b2o$b3o7bobo2$8b2o$7b2o$9bo!

M. I. Wright
Posts: 372
Joined: June 13th, 2015, 12:04 pm

Re: Thread For Your Accidental Discoveries

Post by M. I. Wright » September 19th, 2015, 11:32 pm

Nice! (although it unfortunately takes 8 gliders to synthesize :()

Code: Select all

x = 30, y = 30, rule = B3/S23
10b2o7b2o$9bobo6b2o$11bo8bo7$bo$o26b3o$3o24bo$17bo10bo$16b2o$16bobo2$
13b2o$12b2o$bo12bo$2o$obo7$10b2o$10bobo$10bo!

BobShemyakin
Posts: 214
Joined: June 15th, 2014, 6:24 am

Re: Thread For Your Accidental Discoveries

Post by BobShemyakin » September 20th, 2015, 3:08 am

M. I. Wright wrote:Nice! (although it unfortunately takes 8 gliders to synthesize :()

Code: Select all

x = 30, y = 30, rule = B3/S23
10b2o7b2o$9bobo6b2o$11bo8bo7$bo$o26b3o$3o24bo$17bo10bo$16b2o$16bobo2$
13b2o$12b2o$bo12bo$2o$obo7$10b2o$10bobo$10bo!
This is a error! 2 glider (left-upper and upper-left) cannot pass each other.
This problem is resolved by adding 1 glider:

Code: Select all

x = 49, y = 41, rule = B3/S23
11bo$10bo35b2o$10b3o33b2o$46b2o$47bo$46bobo$41b3obobo$41b4obo$45bo$bo
5bo$bobo2b2o12bo$b2o3bobo10b2o$19bobo8$28b2o$28bobo$28bo$17b2o$16b2o$
18bo$14bo$13b2o$13bobo$b2o$2o$2bo7$10b3o$10bo$11bo!
But the best solution for unix consists of 6 gliders:

Code: Select all

x = 32, y = 18, rule = B3/S23
12bobo$12b2o$13bo2$29b2o$16bo12b2o$15bo$15b3o12bo$11b3o15bobo$11bo12b
2o2bo2bo$12bo11b2obo$8b2o18b2o$2o6bobo$b2o5bo$o$6b2o$5bobo$7bo!
Bob Shemyakin

mniemiec
Posts: 1590
Joined: June 1st, 2013, 12:00 am

Re: Thread For Your Accidental Discoveries

Post by mniemiec » September 20th, 2015, 5:37 am

BobShemyakin wrote:This is a error! 2 glider (left-upper and upper-left) cannot pass each other.
This problem is resolved by adding 1 glider:
This can actually be done without the extra glider, by using a different 2-glider B-heptomino generator. Also, this general mechanism was already known, although the two essential cleanup gliders hit slightly differently, and resulted in forming the same Unix, but in a slightly different way, and a different phase:

Code: Select all

x = 50, y = 86, rule = B3/S23
47boo$47boo$$48bo$47bobo$15bo10boo14boobbobbo$15boo9bobo13boobo$14bobo
9bo19boo8$35boo$34boo$36bo$23b3o$23bo$bo4boo16bo$bbo3bobo11boo$3o3bo
13bobo$20bo9$18bo$17boo$17bobo14$24bo$25bo$23b3o21boo$47boo$$22boo24bo
$22bobo22bobo$22bo19boobbobbo$42boobo$46boo9$36boo$35boo$37bo$24b3o$5b
3o16bo$bbobbo19bo$obo3bo14boo$boo18bobo$21bo9$19bo$18boo$18bobo!
While the six-glider synthesis is usually optimal, this mechanism does have one advantage, in that it forms the Unix in the corner, making it more conducive to syntheses where a Unix needs to form in close quarters with something else being formed at the same time (e.g. Unix on complex still-life) or two siamese Unices, e.g. from a pre-existing long barge.

User avatar
praosylen
Posts: 2443
Joined: September 13th, 2014, 5:36 pm
Location: Pembina University, Home of the Gliders
Contact:

Re: Thread For Your Accidental Discoveries

Post by praosylen » September 20th, 2015, 1:57 pm

A p22 LOM hassler:

Code: Select all

x = 49, y = 31, rule = B3/S23
22b2o7bo$23bo8bo2bo$23bobo4bo4bo4b2o$24b2o13bo2bo$29bo8b2ob2o$28b2ob2o
8bo$28bo2bo13b2o$29b2o4bo4bo4bobo$35bo2bo8bo$39bo7b2o4$23bo8b2o$16bo5b
2o7bobo$15bobo3bo10bo$15b2o5bobo$22bobo2$19b2o$2o16bo2bo$bo16b2obo$bob
o4b3o5bo2b2o$2b2o4bobo4bo$7bo3bo3bo3bo$7bo3bo3bo3bo$11bo4bobo4b2o$6b2o
2bo5b3o4bobo$5bob2o16bo$5bo2bo16b2o$6b2o!
former username: A for Awesome
praosylen#5847 (Discord)

The only decision I made was made
of flowers, to jump universes to one of springtime in
a land of former winter, where no invisible walls stood,
or could stand for more than a few hours at most...

User avatar
codeholic
Moderator
Posts: 1147
Joined: September 13th, 2011, 8:23 am
Location: Hamburg, Germany

Re: Thread For Your Accidental Discoveries

Post by codeholic » September 20th, 2015, 2:57 pm

Very nice! How did you find it?
Ivan Fomichev

User avatar
praosylen
Posts: 2443
Joined: September 13th, 2014, 5:36 pm
Location: Pembina University, Home of the Gliders
Contact:

Re: Thread For Your Accidental Discoveries

Post by praosylen » September 20th, 2015, 3:10 pm

codeholic wrote:Very nice! How did you find it?
It was quite accidental; I was looking for transparent block reactions when I noticed that a LOM, when perturbed by a spark, would reappear along with some junk at gen. 11.
former username: A for Awesome
praosylen#5847 (Discord)

The only decision I made was made
of flowers, to jump universes to one of springtime in
a land of former winter, where no invisible walls stood,
or could stand for more than a few hours at most...

User avatar
Kazyan
Posts: 1247
Joined: February 6th, 2014, 11:02 pm

Re: Thread For Your Accidental Discoveries

Post by Kazyan » September 24th, 2015, 1:29 am

Here's a strange little reaction:

Code: Select all

x = 21, y = 22, rule = LifeHistory
3D2B$D4BC$.D4BC$2.2B3CB$3.6B$4.6B.B$5.8B$6.8B$7.10B$8.10B$8.10B$10.8B
$9.3B2C4B$7.2A3B2C4B$7.2AB.9B$11.10B$12.8B$14.4B$14.2B$13.BC2B$13.C.C
$14.C!
The sacrificial block would be replaceable with an ordinary fishhook or eater2, except that the reaction sticks around for a bit. A catalyst with a large (9 or more generations) recovery time could make this into a mini-rectifier, if it has a low enough profile to not be destroyed by nearby sparks. I haven't found any results with Bellman yet, though (max-active = 9, recovery = 30)
Tanner Jacobi
Coldlander, a novel, available in paperback and as an ebook. Now on Amazon.

Sphenocorona
Posts: 549
Joined: April 9th, 2013, 11:03 pm

Re: Thread For Your Accidental Discoveries

Post by Sphenocorona » September 24th, 2015, 7:50 pm

Kazyan wrote:Here's a strange little reaction
A sacrificial mango (pointing southwest/northeast) also works for the reaction, and produces a phi spark. Restoring the mango might not be a very promising search, but it drags the catalysis action away from the main reaction.

It's too bad a hive or loaf can't be used, and that boats, ships, and blocks produce no workable side reaction.

User avatar
gmc_nxtman
Posts: 1150
Joined: May 26th, 2015, 7:20 pm

Re: Thread For Your Accidental Discoveries

Post by gmc_nxtman » September 25th, 2015, 8:21 pm

Just found this interesting reaction that makes a 22-cell SL:

Code: Select all

x = 17, y = 57, rule = B3/S23
14bobo$14b2o$15bo23$o$obo$bobo6bo$bo2bo5bo$bobo6bo$obo$o23$15bo$14b2o$
14bobo!
And an 11-glider synthesis:

Code: Select all

#C Someone could probably come up with better.
x = 61, y = 79, rule = B3/S23
53bo$52bo$52b3o$47bo$45b2o$46b2o10$58bo$58bobo$58b2o10$o$b2o$2o$30bo$
28bobo$29b2o6$39b3o$20bobo16bo$20b2o18bo$21bo$23b3o$23bo$24bo4$17b2o$
18b2o$17bo23$46b2o$45b2o$47bo$52b3o$52bo$53bo!
I found it by noticing the QB + Blinker that makes it with incomplete induction coils, then added gliders to turn the pre-centuries into the inducting blocks.

Obsolete 5-glider synth of pentadecathlon:

Code: Select all

x = 19, y = 18, rule = B3/S23
5bo$5bobo$3o2b2o$2bo$bo$13bo$14bo$12b3o2$9b2o$10b2o$9bo4$16b2o$16bobo$
16bo!
And finally, a 2c/3 signal termination which is probably known:

Code: Select all

x = 22, y = 19, rule = B3/S23
4b2o$3bobo$2bo5bo$2b7o7bo$2o14b3obo$bo2b7o8b2o$bobo7bo4b2o$2obobob5o4b
ob3o$3bobo15bo$3bobo2b6o6b2o$4b2obo5bo5bo$7bo2b3o5bob2o$4b2obobo5bo2bo
bo$4b2obo2bo4b4obo$8b2o9bo$15b4o$16bo$14bo$14b2o!

Sphenocorona
Posts: 549
Joined: April 9th, 2013, 11:03 pm

Re: Thread For Your Accidental Discoveries

Post by Sphenocorona » October 2nd, 2015, 6:33 pm

I was looking for reactions that could possibly be made transparent in order to produce a new conduit, and... um, yeah... ...wasn't expecting a LWSS to appear so quickly...

Code: Select all

x = 17, y = 7, rule = LifeHistory
15.2C$15.2C2$10.C$.C7.C.C$3C6.C2.C$C9.2C!
The important intermediate is this very simple looking pattern (actually, a relative of this pattern):

Code: Select all

x = 5, y = 9, rule = B3/S23
2b2o$2b2o2$2bo$bobo$4bo$3bo$b2o$2o!
Is it possible to use this reaction in a conduit?

Bullet51
Posts: 663
Joined: July 21st, 2014, 4:35 am

Re: Thread For Your Accidental Discoveries

Post by Bullet51 » October 3rd, 2015, 6:59 pm

A useless p18 from Randagar:

Code: Select all

x = 54, y = 54, rule = B3/S23
23bo$23bo$22b3o2$22b3o2$18b2o2b3o2b2o$18bobo2bo2bobo$19bo7bo2$19bo7bo$
18bobo2bo2bobo$18b2o2b3o2b2o2$22b3o2$22b3o$23bo$23bo17b2o3b2o$41bobobo
bo$29bo12bo3bo$24bo3bobo$22b4ob2ob2o5bobobo5bobobo$21b2o3bobo2bo3b3obo
b2o3b2obob3o$20bo3b2o2b2ob2o4bobobo5bobobo$6b2o3b2o8b4o4bobo$6bobobobo
9bo7bo11bo3bo$7bo3bo11bo7bo9bobobobo$22bobo4b4o8b2o3b2o$2bobobo5bobobo
4b2ob2o2b2o3bo$3obob2o3b2obob3o3bo2bobo3b2o$2bobobo5bobobo5b2ob2ob4o$
23bobo3bo$7bo3bo12bo$6bobobobo$6b2o3b2o17bo$30bo$29b3o2$29b3o2$25b2o2b
3o2b2o$25bobo2bo2bobo$26bo7bo2$26bo7bo$25bobo2bo2bobo$25b2o2b3o2b2o2$
29b3o2$29b3o$30bo$30bo!
Still drifting.

Sokwe
Moderator
Posts: 2645
Joined: July 9th, 2009, 2:44 pm

Re: Thread For Your Accidental Discoveries

Post by Sokwe » October 3rd, 2015, 7:14 pm

Bullet51 wrote:A useless p18 from Randagar
This was first found by Gabriel Nivasch in October 2002. It's included in the jslife oscillator collection, supported by jams:

Code: Select all

x = 30, y = 30, rule = B3/S23
12b2o$11bo2bo$9bo2bobo$8bo4bo$8bo$10b3o2$27b2o$26bo2bo$16b3o8bobo$11b
3o2bobo4b4obo$9b2o2bo2bo2bo3b2ob2o$9bo2b2o3bobo6bo$9b3o5b3o2$b2o$o2bo
6b3o5b3o$obo2bo4bobo3b2o2bo$bo3bo4bo2bo2bo2b2o$5bo5bobo2b3o$2bo8b3o$3b
2o3$17b3o$21bo$16bo4bo$15bobo2bo$15bo2bo$16b2o!
-Matthias Merzenich

User avatar
dvgrn
Moderator
Posts: 10612
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Thread For Your Accidental Discoveries

Post by dvgrn » October 3rd, 2015, 9:41 pm

Sphenocorona wrote:I was looking for reactions that could possibly be made transparent in order to produce a new conduit, and... um, yeah... ...wasn't expecting a LWSS to appear so quickly...
Is it possible to use this reaction in a conduit?
Anything's possible, of course. But for similar reasons to the ones given here, I wouldn't say it's a likely line of investigation.

If you only had to use catalysts to tame the junk and restore the block, maybe that would be worth a try with a search program. But to restore the block, and simultaneously restore the loaf (or whatever object created a similar spark)... well, that would take a lot more beginner's luck than I probably have left. Your mileage may vary, of course.

-- I bet we're at the point where a distributed search using some kind of catgl / ptbsearch / CatForce / Bellman variant would eventually turn up a direct Herschel-to-spaceship converter. The main thing is to set up the program to recognize *WSS outputs reliably, instead of just adding more catalysts to kill them off.

Beyond that, it's just a matter of generating a huge variety of configurations that might possibly be viable catalysts, and testing them automatically. As Catagolue has shown, the weight of probability will do the rest sooner or later -- there are a lot of spaceships out there.

Might even dig up a Herschel-to-swimmer conduit eventually, with the same program (sadly, Herschel-to-loafer is probably too much to hope for...)

Sphenocorona
Posts: 549
Joined: April 9th, 2013, 11:03 pm

Re: Thread For Your Accidental Discoveries

Post by Sphenocorona » October 3rd, 2015, 9:51 pm

I wasn't expecting a stable object to produce the junk. I was saying that it might be possible to rig up some other connection to a Herschel that produces the junk. Having to restore another still life is obviously not going to work.

The loaf reaction was only shown as the original "accidental discovery" and I do not think that what you thought I wanted is in any way a good idea at all.

User avatar
gmc_nxtman
Posts: 1150
Joined: May 26th, 2015, 7:20 pm

Re: Thread For Your Accidental Discoveries

Post by gmc_nxtman » October 3rd, 2015, 11:18 pm

A truly strange component:

Code: Select all

x = 18, y = 9, rule = B3/S23
2b2o$2b2o12bo$9bobo3bo$6o3b2o4b3o$o4bo4bo$b4o$3bo4b2o$5bob2o$4b2o3bo!
EDIT: Similar, yet somehow 10 orders of magnitude weirder:

Code: Select all

#C No, the Pi cannot be moved closer.
x = 21, y = 12, rule = B3/S23
15bo$14bo$14b3o$ob2o$2obo$10bo7b3o$6o3bo8bo$o4bo3b3o7bo$b4o$3bo5bo$5bo
2b2o$4b2o2bobo!

User avatar
Scorbie
Posts: 1692
Joined: December 7th, 2013, 1:05 am

Re: Thread For Your Accidental Discoveries

Post by Scorbie » October 3rd, 2015, 11:29 pm

Sokwe wrote:
Bullet51 wrote:A useless p18 from Randagar
This was first found by Gabriel Nivasch in October 2002. It's included in the jslife oscillator collection, supported by jams:

Code: Select all

x = 30, y = 30, rule = B3/S23
12b2o$11bo2bo$9bo2bobo$8bo4bo$8bo$10b3o2$27b2o$26bo2bo$16b3o8bobo$11b
3o2bobo4b4obo$9b2o2bo2bo2bo3b2ob2o$9bo2b2o3bobo6bo$9b3o5b3o2$b2o$o2bo
6b3o5b3o$obo2bo4bobo3b2o2bo$bo3bo4bo2bo2bo2b2o$5bo5bobo2b3o$2bo8b3o$3b
2o3$17b3o$21bo$16bo4bo$15bobo2bo$15bo2bo$16b2o!
By the way, the minimal form looks to be the one supported by caterers, both in population and bounding box:

Code: Select all

x = 28, y = 28, rule = B3/S23
9bo$4b4o3bo$7bo3bo$11bo$8bo17bo$9b2o15bo$26bo$25b2o$10bo5bo6bo$9bobo4b
obo3bo4bo$11bo4bo2bo2bo$8b3o6b2o5b3o5$b3o5b2o6b3o$5bo2bo2bo4bo$o4bo3bo
bo4bobo$4bo6bo5bo$b2o$bo$bo15b2o$bo17bo$16bo$16bo3bo$16bo3b4o$18bo!

User avatar
praosylen
Posts: 2443
Joined: September 13th, 2014, 5:36 pm
Location: Pembina University, Home of the Gliders
Contact:

Re: Thread For Your Accidental Discoveries

Post by praosylen » October 6th, 2015, 6:39 pm

A small MWSS-to-G converter:

Code: Select all

x = 13, y = 9, rule = B3/S23
5b2o$5bobob2obo$7bobob2o$7b2o$2bo$o3bo$5bo4b2o$o4bo4b2o$b5o!
former username: A for Awesome
praosylen#5847 (Discord)

The only decision I made was made
of flowers, to jump universes to one of springtime in
a land of former winter, where no invisible walls stood,
or could stand for more than a few hours at most...

User avatar
dvgrn
Moderator
Posts: 10612
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Thread For Your Accidental Discoveries

Post by dvgrn » October 6th, 2015, 6:48 pm

A for awesome wrote:A small MWSS-to-G converter...
Found by Sokwe a couple of years ago. Still a good find. It's too bad it doesn't work for LWSS or HWSS.

Has anyone done the equivalent Bellman search with a HWSS input, by the way?

User avatar
Kazyan
Posts: 1247
Joined: February 6th, 2014, 11:02 pm

Re: Thread For Your Accidental Discoveries

Post by Kazyan » October 10th, 2015, 3:40 am

Me: "Bellman, can you turn a B-heptomino's natural block into a beehive?"
Bellman: "Not with 6 active cells, boss. Here's a funky traffic light catalyst as an apology."

Code: Select all

x = 15, y = 12, rule = B3/S23
12bo$10b3o$4bo4bo$3bobo3b2o$b3obo$o4b2o$b2obo2bo$2bob2obo$2bo2bobobo4b
o$3b2o3bobo2bo$9bo2b2o$14bo!
Tanner Jacobi
Coldlander, a novel, available in paperback and as an ebook. Now on Amazon.

Post Reply