Soup search results

[Apple Bottom]

Wow! I had seen that infinite growth, but I didn't know that my apgnano run discovered it!

If you check your user page often, you would see the 'Limitless' badge and the yl384 in your list of discoveries:

https://catagolue.appspot.com/user/EricABQ

You can also figure it out from the links. The sample soup link includes the seed used, and the MD5 value of that seed allows you to access the haul.

Specifically, the sample soup's link is this:

Code: Select all

https://catagolue.appspot.com/hashsoup/C1/n_B5FXtgU55bKd7493993/b3s23

That means it's soup 7493993 generated from seed n_B5FXtgU55bKd. All you need to do is get the MD5 value of the seed:

Code: Select all

$ echo -n n_B5FXtgU55bKd | md5sum
692083554f90c886d1db8fd52718fa0b *-
$ 

The link for the haul then is:

Code: Select all

https://catagolue.appspot.com/haul/b3s23/C1/692083554f90c886d1db8fd52718fa0b

And that, as they say, is that. It's a useful little trick to know!

[Scorbie]

Thanks to chris's suggestions, I think I found a good seed for a BTS:

Code: Select all

x = 26, y = 19, rule = B3/S23
4b2o$3bo2bo18bo$3bobo18b2o$4bo18b2o$24bo$25bo$12bo$b2o8bobo$o2bo7bobo$
bobo8bo$2bo$7b2o7b2o$6bo2bo5bo2bo$7b2o7b2o2$12bo$11bobo$11bobo$12bo!

Here are some other seeds:

Code: Select all

x = 38, y = 171, rule = B3/S23
12b2o$11bo2bo$12b2o3$15b2o$15b2o5$3o5bobo$11bo$11bo$8bo2bo$9b2o21b3o2$
36b2o$36b2o$23b3o33$16bo13bo$15bobo12b2o$15bobo11bo2bo$10b2o4bo4bo8b2o
3bo$10b2o9bo8bo3bobo$21bo12bobo$8bo5bo20bo$6b2obo3bob2o$5bo3bo3bo3bo$
5bo4bobo4bo$5b2o4bo4b2o16b2o$7bo7bo18b2o$9bo3bo$9bo3bo2$10b3o3$16bo$
15bobo$16bo22$13bo$12b3o2$21bob2o$21bo2bo$21b3o$18bo$18bo$18bo2$14b3o
3b3o2$14bobo7b2o$10b2o3bo8b2o$13bo$8bo4b4o$8bo2bo5bo$8bo2bo6bo$10b2o2b
o$12b2o2b2o$12bobobo2$13b3o31$15bo$15b2o$15b2o$16bo$21bob2o$21bo2bo$
21b3o$18bo$18bo$18bo2$14b3o3b3o2$14bobo7b2o$10b2o3bo8b2o$13bo$8bo4b4o$
8bo2bo5bo$8bo2bo6bo$10b2o2bo$12b2o2b2o$12bobobo2$13b3o!

All except the first needs synchronization. The mess in the last two is from the natural evolution of the R-pentomino.

And here's a glance at the first 30 soups or so of the Bookend with long pedestal catalyst:

Code: Select all

x = 271, y = 27, rule = B3/S23
b2o$bobo254bo10bo$2bobo253bo9bobo$3bo158bo95bo9bobo$6b2o153bobo105bo$
5bo2bo77b3o72bo2bo$2o4b2o154b2o98b2o$2o64bo189bo5bobo$64b2ob2o186b2o5b
o$64b2ob2o188bo2$159bo18b2o78bo$13b2o133bo9b3o17b2o78b2o$13b2o132bobo
8bo2bo97bo$81b2o64bobo10b2o$80bo2bo64bo$7b2o72b2o96b2o$7b2o169bo2bo$
179b2o2$264b2o$264b2o3$7b3o$7bo2bo$7bob2o!

[calcyman]

Brett Berger has found a yl4608:

Code: Select all

x = 16, y = 16, rule = B3/S23
2obobo3b3o2b2o$4bob2obobo2b2o$b3o2bo2bob4o$bobobob2obob4o$8b2o$b2obo4b
ob4o$2bob4obobobo$bo12bo$bo2bo3b3o2b3o$2obo2bobob3ob2o$ob2o4b2o4bo$obo
b2obo2bob2o$ob3ob2ob5obo$7bo2b2obo$o2b2ob6o2b2o$2bo2bo2bob2o2b2o!

It's currently in quarantine because apgsearch classified it as 'PATHOLOGICAL' due to its high period. I'll implement a way of editing the contents of the haul to make it say something more appropriate such as yl4608_2015_11_28 before letting it into the census.

[danieldb]

Brett Berger has found a yl4608:

Code: Select all

x = 16, y = 16, rule = B3/S23
2obobo3b3o2b2o$4bob2obobo2b2o$b3o2bo2bob4o$bobobob2obob4o$8b2o$b2obo4b
ob4o$2bob4obobobo$bo12bo$bo2bo3b3o2b3o$2obo2bobob3ob2o$ob2o4b2o4bo$obo
b2obo2bob2o$ob3ob2ob5obo$7bo2b2obo$o2b2ob6o2b2o$2bo2bo2bob2o2b2o!

It's currently in quarantine because apgsearch classified it as 'PATHOLOGICAL' due to its high period. I'll implement a way of editing the contents of the haul to make it say something more appropriate such as yl4608_2015_11_28 before letting it into the census.

o_0 that's huge! The whole hex notation for linear growth is confsing, maybe you should use timestamps instead?

[gameoflifeboy]

Why did it get classified as pathological? What is the largest period that can be recognized by linearlyse()?

The whole hex notation for linear growth is confsing, maybe you should use timestamps instead?

The current notation has the advantage that a linear growth pattern always gets the same apgcode, regardless of when it's discovered. According to the documentation of apgsearch, the three numbers are the period of the population, the period of the increase/decrease in population, and the first "moment" of the pattern, followed by the md5 hash of the second and third moments, separated by a #. I haven't really worked out what all this means though.

[danieldb]

Why did it get classified as pathological? What is the largest period that can be recognized by linearlyse()?

The whole hex notation for linear growth is confsing, maybe you should use timestamps instead?

The current notation has the advantage that a linear growth pattern always gets the same apgcode, regardless of when it's discovered. According to the documentation of apgsearch, the three numbers are the period of the population, the period of the increase/decrease in population, and the first "moment" of the pattern, followed by the md5 hash of the second and third moments, separated by a #. I haven't really worked out what all this means though.

True, I thought about that.

[danieldb]

-

[danieldb]

-

[Apple Bottom]

But why isn't it on my profile?

Not all still lifes are credited as discoveries; there's simply too many. There is significant overlap between those population counts where most still lifes haven't been found yet (higher) and those where new discoveries are commonplace (generally lower).

Put another way, a new xs26 isn't all that remarkable. There's, oh, perhaps 25 million possible ones, and a four-digit number of them have already turned up naturally.

I don't know what the rules are for "remarkable" still lifes; Calcyman may be able to tell you.

[danieldb]

But why isn't it on my profile?

Not all still lifes are credited as discoveries; there's simply too many. There is significant overlap between those population counts where most still lifes haven't been found yet (higher) and those where new discoveries are commonplace (generally lower).

Put another way, a new xs26 isn't all that remarkable. There's, oh, perhaps 25 million possible ones, and a four-digit number of them have already turned up naturally.

I don't know what the rules are for "remarkable" still lifes; Calcyman may be able to tell you.

I... Never mind. That wasn't my question. It was my first natural discovery and I was wondering why it hadn't appeared on my user page for catagolue. Anyways, now it has, and I hope this doesn't start any arguments

[danieldb]

Also, why are some soup types (8x32, 25%, 75%, 25pct, 75pct, etc.) not appear in the symmetry select?

[gameoflifeboy]

This soup produces two non-overlapping switch engines which were counted as one object:

Code: Select all

x = 16, y = 16, rule = B3/S23
bo5bo2b2obo$o3bo3b3ob2o$b3ob2ob4o2bo$2obob4o3bob2o$3bob3ob2ob2o$bobobo
3b2o2bobo$2o3bo3b4o2bo$obo5bo2b2o$bo6bobo$2o3b2o2bo3b3o$2bo2bob6ob2o$b
3o2bob8o$2b2o4b3obob2o$bobob5o2b2o$4obo2bobo$bo3bo3b2ob2obo!

Also, why are some soup types (8x32, 25%, 75%, 25pct, 75pct, etc.) not appear in the symmetry select?

25% and 75% never worked, because of the percent sign in the URL. 25pct and 75pct never took off, and apgsearch v1.1 doesn't even support them. However, I see no reason 8x32 shouldn't be added to the list of available "symmetries".

[danieldb]

25% and 75% never worked, because of the percent sign in the URL.

If you change it from "25%/75%" to "25%25/75%25", it works.

[towerator]

I'm surprized no one commented this:

Code: Select all

x = 32, y = 32, rule = B3/S23
4b2obo3b5ob4o5b5o$3bo2b4ob2ob4ob4obob2o3bo$3b2ob3ob4o6b5obobo2bo$b2ob
2obo2b2o2bo2b4ob4o3bobo$ob2o3bobo2b3obob2obo2b2o4b2o$o2bobobo2bobo2b2o
bob3o2b2o2b3o$b2o7b4obobo8b3o$6o6b2ob3obobo5b4o$b2o5b5ob2ob2o2bobo4b2o
$bo2bo3b3o5bob2ob2o3bob3o$2b2ob2ob2obo3bo3b2ob3ob2ob2o$4o2bobob2ob2o3b
ob2o4bob4o$3ob5o4b5o3b2obo2b2ob2o$obobob2o3b8obob2o2b3o2bo$2ob2o3bo2b
4ob4o3b3o2bob2o$2o3b4obob2o3b3o2bobob2o2bo$bo2b2obobo2b3o3b2obob4o3b2o
$2obo2b3o3b4ob4o2bo3b2ob2o$o2b3o2b2obob8o3b2obobobo$2ob2o2bob2o3b5o4b
5ob3o$4obo4b2obo3b2ob2obobo2b4o$b2ob2ob3ob2o3bo3bob2ob2ob2o$b3obo3b2ob
2obo5b3o3bo2bo$2b2o4bobo2b2ob2ob5o5b2o$b4o5bobob3ob2o6b6o$3b3o8bobob4o
7b2o$3o2b2o2b3obob2o2bobo2bobobo2bo$2o4b2o2bob2obob3o2bobo3b2obo$obo3b
4ob4o2bo2b2o2bob2ob2o$o2bobob5o6b4ob3ob2o$o3b2obob4ob4ob2ob4o2bo$b5o5b
4ob5o3bob2o!

Natural gun! Double barreled AND glide-symmmetrical!
Maybe we should get a new category for guns.

[Saka]

I'm surprized no one commented this:

Code: Select all

x = 32, y = 32, rule = B3/S23
4b2obo3b5ob4o5b5o$3bo2b4ob2ob4ob4obob2o3bo$3b2ob3ob4o6b5obobo2bo$b2ob
2obo2b2o2bo2b4ob4o3bobo$ob2o3bobo2b3obob2obo2b2o4b2o$o2bobobo2bobo2b2o
bob3o2b2o2b3o$b2o7b4obobo8b3o$6o6b2ob3obobo5b4o$b2o5b5ob2ob2o2bobo4b2o
$bo2bo3b3o5bob2ob2o3bob3o$2b2ob2ob2obo3bo3b2ob3ob2ob2o$4o2bobob2ob2o3b
ob2o4bob4o$3ob5o4b5o3b2obo2b2ob2o$obobob2o3b8obob2o2b3o2bo$2ob2o3bo2b
4ob4o3b3o2bob2o$2o3b4obob2o3b3o2bobob2o2bo$bo2b2obobo2b3o3b2obob4o3b2o
$2obo2b3o3b4ob4o2bo3b2ob2o$o2b3o2b2obob8o3b2obobobo$2ob2o2bob2o3b5o4b
5ob3o$4obo4b2obo3b2ob2obobo2b4o$b2ob2ob3ob2o3bo3bob2ob2ob2o$b3obo3b2ob
2obo5b3o3bo2bo$2b2o4bobo2b2ob2ob5o5b2o$b4o5bobob3ob2o6b6o$3b3o8bobob4o
7b2o$3o2b2o2b3obob2o2bobo2bobobo2bo$2o4b2o2bob2obob3o2bobo3b2obo$obo3b
4ob4o2bo2b2o2bob2ob2o$o2bobob5o6b4ob3ob2o$o3b2obob4ob4ob2ob4o2bo$b5o5b
4ob5o3bob2o!

Natural gun! Double barreled AND glide-symmmetrical!
Maybe we should get a new category for guns.

Who was it found by? I think it's the smallest 4 period gun! Maybe we should add it to the wiki

[Sphenocorona]

I'm surprized no one commented this

I'm pretty sure it was mentioned briefly months ago, but I think the consensus is that the GGG (or maybe even the 3 QB pseudo-GGG) is more likely than that one to occur in an asymmetric soup.

About adding it to the wiki: I think there was some mention of doing so the last time it was brought up, but I don't remember what the decision and reasons were. The gun in p156 oscillator form is mentioned in one line on the wiki, as an unnamed oscillator listed in a smallest oscillators per period list. The mention is a red link, so the wiki isn't very enlightening there...

[Scorbie]

Catagolue link:
https://catagolue.appspot.com/object/yl ... 9035/b3s23
Hmm. It was known before soup searching, (by randomagar) And it popped out from apgsearch once before Catagolue, but I'm also surprised that nobody noticed this till 76 occurrences, after Catagolue.
I'm not sure if this yields a new category. How would it be categorized? Natural? Maybe simkin's gun could be found in random ash.

EDIT: Saka: A related osc is already sort of in the wiki, although the page is deleted of spam. See 60P312. I'm not sure if this gun is the smallest.

[thunk]

I'm pretty sure this gun is already in the gun collection (for p156).

I'm pretty sure it was mentioned briefly months ago, but I think the consensus is that the GGG (or maybe even the 3 QB pseudo-GGG) is more likely than that one to occur in an asymmetric soup.

Or maybe the similarly-Blockic Simkin gun, for that matter.

[calcyman]

but I'm also surprised that nobody noticed this till 76 occurrences, after Catagolue.

What? I thought everyone knew about this gun having occurred semi-naturally. This is in one of my old e-mails from Dave Greene:

Code: Select all

#C soup producing a (known) p156 multi-glider shotgun,
#C found by Dongook Lee, 9 January 2015
#C http://www.conwaylife.com/forums/viewtopic.php?f=2&t=1452&start=200#p15913
x = 32, y = 32, rule = B3/S23
2o2bo2b3o7b3obobo2bo3b2o$2ob2o2b2o2b3o4b11ob2o$10bo2bo3b4o2b3o2bo$bob
4obobob3o2bobobob2ob5o$2ob3o4bo3bob2ob4o2b4obo$3b2ob2obo2b3ob3o2bo2bo
2b2ob2o$3bob2ob2obo8bo2b3obob2o$2o3bo2b2o3b3o9b2ob3o$2obo2b7o2b3ob5obo
2b4o$o4b4o2bob2obo2b2o2bo3bo2bo$2b3o3bobo3bobob2obobo2b3ob2o$bo4bob2ob
2o3bob3ob2obobob2o$bobobo2bo2bobo2bo3b4o3b5o$b3obobobo2b2o2bobob2o4bo
2b3o$3b3obob2o3b4o5bo2b4obo$7b2o5b10o2b2o$4b2o2b10o5b2o$ob4o2bo5b4o3b
2obob3o$3o2bo4b2obobo2b2o2bobobob3o$5o3b4o3bo2bobo2bo2bobobo$b2obobob
2ob3obo3b2ob2obo4bo$2ob3o2bobob2obobo3bobo3b3o$bo2bo3bo2b2o2bob2obo2b
4o4bo$4o2bob5ob3o2b7o2bob2o$b3ob2o9b3o3b2o2bo3b2o$b2obob3o2bo8bob2ob2o
bo$2ob2o2bo2bo2b3ob3o2bob2ob2o$bob4o2b4ob2obo3bo4b3ob2o$b5ob2obobobo2b
3obobob4obo$3bo2b3o2b4o3bo2bo$2ob11o4b3o2b2o2b2ob2o$2o3bo2bobob3o7b3o
2bo2b2o!

As for natural guns in b3s23/C1, I think both the Gosper glider gun and Simkin's glider gun could plausibly appear before the 10^15 object mark. Of course, we're only 3% of the way towards this target (at 30 * 10^12 objects), so we need either much higher engagement, GPU computing, or lots more time before we realise this goal.

[danieldb]

This 30-celler (http://catagolue.appspot.com/object/xs3 ... 079c/b3s23):

Code: Select all

x = 8, y = 9, rule = B3/S23
2o3b2o$obobo2bo$2bobob2o$b2obo$4bo$b2ob2o$2bobo$obobobo$2o3b2o!

Has 3 natural appearences (1 of which is from me), all of which use this predecessor:

Code: Select all

x = 13, y = 9, rule = B3/S23
5b2o3b2o$4bo2bobo2bo$4b2obobob2o$2o5bobo$b2o4bobo$o5b2ob2o$7bobo$5bobo
bobo$5b2o3b2o!

[Scorbie]

but I'm also surprised that nobody noticed this till 76 occurrences, after Catagolue.

What? I thought everyone knew about this gun having occurred semi-naturally. This is in one of my old e-mails from Dave Greene:

Code: Select all

#C soup producing a (known) p156 multi-glider shotgun,
#C found by Dongook Lee, 9 January 2015
#C http://www.conwaylife.com/forums/viewtopic.php?f=2&t=1452&start=200#p15913
#C Hehehehehehe that's me:)

Should have been clearer. Yep, that was found by apgsearch but prior to Catagolue. I just expected that on it's first occurrence on Catagolue, somebody would say "Hey now it's on Catagolue too!", and was slightly surprised that the gun was kinda forgotton for 76 soups.

[praosylen]

There is a new 42-cell still life:

Code: Select all

x = 9, y = 11, rule = B3/S23
b2o3b2o$bobobobo$3bobo$2b2ob2o$o7bo$4ob4o$3bobo$2obobob2o$o2bobo2bo$bo
bobobo$2bo3bo!

The soup suggests two possible synthesis mechanisms:

Code: Select all

x = 38, y = 9, rule = B3/S23
8b3o20b2ob2o$7bo3bo18b2o3b2o$7b2ob2o17b2o5b2o$8b3o2$31b5o$31bo3bo$bo
15bo13b5o$3o13b3o!

[mniemiec]

There is a new 42-cell still life:

Code: Select all

x = 9, y = 11, rule = B3/S23
b2o3b2o$bobobobo$3bobo$2b2ob2o$o7bo$4ob4o$3bobo$2obobob2o$o2bobo2bo$bo
bobobo$2bo3bo!

The soup suggests two possible synthesis mechanisms:

Code: Select all

x = 38, y = 9, rule = B3/S23
8b3o20b2ob2o$7bo3bo18b2o3b2o$7b2ob2o17b2o5b2o$8b3o2$31b5o$31bo3bo$bo
15bo13b5o$3o13b3o!

Here is a pretty gruesome 25-glider synthesis. The block and blinker insertions are particularly hideous. I am sure it can be done much more cheaply (probably around 9 gliders), by making two LOMs and a pulsar in close proximity, but I couldn't find a combination that worked.

Code: Select all

x = 235, y = 25, rule = B3/S23
107bo$106bo$106b3o$104bo90bo$9bo92bobo7bo81bo20bo$7bobo93boo6bo43bo35b
obb3o12bo5bobo$obo5boo101b3o41bobo31bobo15boo6boo10boo3boo$boo152boo
33boo16boobboo13bobobobo$bo6bo19booboo15booboo25booboo15booboo25booboo
15booboo7bo7booboo25booboo9bobo14bobo$8boo17boboboo14boboboo24boboboo
14boboboo24boboboo14boboboo3bobbo7bobobobo23bobobobo8bo15booboo$7bobo
17boo18boo28boo18boo28boo18boo6boobb3o5boo3boo23boo3boo22bo7bo$155bobo
68b4ob4o$229bobo$78bo19bo29bobboo15bobboo15bobboo25bobboo23booboboboo$
76b3o17b3o27b3obboo13b3obboo13b3obboo23b3obboo23bobbobobbo$75bo19bo29b
o19bo19bo29bo31bobobobo$75boo18boo28boo18boo18boo23bo4boo7bobo21bo3bo$
191boo11boo$55bo59bo74boo13bo$53boo53boo4boo79boo3b3o$50bo3boo52bobo3b
obo77bobo3bo$40boo9boo55bo79bo7bo4bo$39bobo8boo9boo125boo$41bo18boo
125bobo$62bo!

[Sphenocorona]

There is a new 42-cell still life:

Code: Select all

rle

The soup suggests two possible synthesis mechanisms:

Code: Select all

rle

Here is a pretty gruesome 25-glider synthesis. The block and blinker insertions are particularly hideous. I am sure it can be done much more cheaply (probably around 9 gliders), by making two LOMs and a pulsar in close proximity, but I couldn't find a combination that worked.

Code: Select all

rle

Here's a 15G synth based off the first of the synthesis mechanisms described above:

Code: Select all

x = 77, y = 58, rule = B3/S23
18bo$19b2o$18b2o6$26bo$24bobo$25b2o$32bo$30bobo$31b2o2$45bo$45bobo$45b
2o2$obo71bobo$b2o71b2o$bo73bo2$16bo43bo$17b2o39b2o$16b2o41b2o5$27bo$
25bobo$26b2o6$45b2o$45bobo$45bo2$31b2o$30bobo$32bo4$48bo$47b2o$47bobo
2$13b3o45b3o$15bo45bo$14bo47bo$49bo$48b2o$48bobo!

Because there's so much room with this method, there's probably a lot of room for improvement (reducing the 8 glider (!) setup of the dentral pattern to something more reasonable like 3 gliders).

[isaacg]

Check out this new type of 2 switch-engine interaction, with period 4608:

Code: Select all

x = 16, y = 16, rule = B3/S23
oobobobbbooobboo$
bbbboboobobobboo$
booobbobboboooob$
bobobobooboboooo$
bbbbbbbboobbbbbb$
boobobbbboboooob$
bboboooobobobobb$
bobbbbbbbbbbbbob$
bobbobbbooobbooo$
oobobboboboooboo$
oboobbbboobbbbob$
oboboobobboboobb$
oboooboobooooobo$
bbbbbbbobboobobb$
obbooboooooobboo$
bbobbobboboobboo!

https://catagolue.appspot.com/object/yl ... 1_28/b3s23