Soup search results

[gameoflifeboy]

17-cell predecessor to a 40-cell still life:

Code: Select all

x = 8, y = 6, rule = B3/S23
b3o$o2bob2o$4bobo$o2bobo$b3o$6b2o!

16-cell predecessor:

Code: Select all

x = 8, y = 7, rule = B3/S23
3b2o$bob3o$2o$2b2ob2o$6bo$2bo3bo$7bo!
[[ PAUSE 1 GPS 5 STOP 20 ]]

[mniemiec]

If you use the "candlefrobra" terminology, this should be (at least in my opinion) "candlefrobra with trans-block", not "candlefrobra on trans-block". On usually means the parts are attached, but sometimes it just means one is being stabilized by the other, such as block on table, although usually the word and is used for such purposes. However, because candlefrobra is the name of the actual oscillator stabilized with a cis-block, rather than just the inductor (the long hook eating tub), with would be used to describe a slightly different version of the candlefrobra, but with the important part (the rotor) still the same. So I would call this either "candlefrobra with trans-block", or (my preferred terminology) "trans-block and long hook eating tub".
There. That's my nomenclature system for naming variations of oscillators.

As far as I am aware, "on" was used to indicate induction (e.g. "block on table") long before "and" was. When I created the old version of my web site (in the '90s), I was not aware of any standard name for the P3 inductor, so I called it "long bookend eating tub". In the new incarnation, I have been calling it "candelfrobra" (although this is technically the name of one particular complete oscillator, rather than just the active inductor itself). "With" has always seemed to have a much less specific and ambiguous meaning, as it does in English (i.e. eating strawberries with whipped cream, or eating strawberries with a spoon). There is certainly a lot of work that needs to be done to formalize Life object nomenclature, because the current state is frequently contested, ambiguous, and/or inconsistent.

[dvgrn]

The Twitterbot has reported another eater2 variant that looks slow-salvo compatible. It's not quite perfect, of course, but the block that does the eating is actually on the edge of the reaction envelope:

Code: Select all

x = 46, y = 23, rule = LifeHistory
21.B$20.4B$14.B2.9B$13.4B2DBD5B$12.2BA2B2DB3D5B$11.3BA8BD5B7.2B$12.2B
A2B2DB2D2BD4B5.7BA2B$11.5BD2B2DB2D6B2.9B2A2B$10.3A2BD3A20BA2B.2AB$4.B
3.7B2D21B2A2B.2A$3.4B.6BA22B2A3B$.13BA10BA9BD2B2A2B$11B2DBA9BABA7BDBD
6B$.9BD2BD3B3A5BA3B4.BDBD6B$4.6BDBD15B5.2BD6B$4.3B.3BD15B7.8B$10.16B
9.7B$10.16B12.3B$10.17B12.B$11.2B.13B$15.10B$18.7B$18.4B!
[[ VIEWONLY THUMBNAIL ZOOM 12 HEIGHT 300 ]]

So... what's the cheapest way to slow-salvo-construct that sabotaged B-heptomino? Or there's at least one later stage that might be a good line of attack, with interactions that need replacements at T=20 and T=35.

[chris_c]

So... what's the cheapest way to slow-salvo-construct that sabotaged B-heptomino?

I searched G colliding with 2 common still lifes and this mess is the only thing that showed up:

Code: Select all

x = 49, y = 11, rule = B3/S23
o10b2o$b2o7bo2bo30bo$2o9bobo30bo$12bo31bo$8bo$7bobo30b3o3b3o$7bobo$8bo
35bo$33bo10bo$32bobo9bo$33bo5b3o!

[Scorbie]

I searched G colliding with 2 common still lifes and this mess is the only thing that showed up:

I think that's good enough, isn't it? I don't think cleanup is necessary, but just in case: the worst is 2 gliders:

Code: Select all

x = 91, y = 31, rule = B3/S23
bo$2bo$3o7$38bo$39bo$37b3o7$80bob2o$78b3ob2o$77bo$37bo16b2o20bo2b2ob2o
$37bo15bo2bo20b2ob2o2bo$37bo6b2o7bo2bo26bobo$43bo2bo7b2o28b2o$44bobo$
45bo19bo$64bobo21b2o$64bobo20bo2bo$65bo22bobo$89bo!

[dvgrn]

I searched G colliding with 2 common still lifes and this mess is the only thing that showed up...

Looks pretty good to me!

I also had a quick look at the previous eater2 variant, which has three soups available now:

Code: Select all

x = 355, y = 46, rule = LifeHistory
333.A.5A2.A$335.2A7.A.A$332.5A2.3A4.2A$189.A.A.2A.3A2.2A129.2A3.A.A.
2A.A$191.A.A2.A3.4A129.A3.6A.4A$51.A.A5.A5.A125.A.7A.A.A129.2A7.A2.A$
51.A.2A.4A2.2A.A122.5A.A3.2A.2A130.A.A.A2.A.A2.A.A$52.3A2.3A.2A2.2A
122.A.2A2.2A5.2A130.2A5.2A.4A$51.2A.2A3.2A.A.A.A121.A.7A.A.2A.A129.2A
.2A3.2A3.A$51.2A.A3.A3.A.2A123.A2.2A2.3A.A2.A128.4A.2A2.A.A3.A$56.A2.
A2.2A124.A.A4.A3.2A2.A128.A.4A.A$52.6A.A.A.3A122.A.A2.A2.3A.2A130.2A
3.3A.A.4A$51.3A.A.A2.2A.A.A124.6A.6A131.2A.A.A.2A2.A$51.2A.3A4.A.A.2A
122.A.A2.A2.A4.2A128.5A2.A3.A.A$51.2A.A.2A.2A2.A.2A122.3A2.A.3A.A.2A
128.3A2.A.A.A2.A.A$53.3A2.2A.4A.A121.A2.A.A2.3A3.A129.4A4.A.A2.3A$52.
A3.2A.2A.5A122.A2.A2.A.A5.A$52.4A.3A3.A2.A122.A.5A2.2ADA$51.3A3.A5.4A
122.A.A3.6AD$51.2A3.2A3.3A2.A128.3D.2D$53.6A2.2A131.D$195.3D.2D$197.D
.2D8$.2D.D$.2D.3D$7.D$.2D.3D$D2.2D$2D3$352.D$351.D.D$351.D.D$350.2D.
2D$350.D$351.D.2D$349.D.D.2D$349.2D!

EDIT: Now four soups, and this last one looks somewhat promising:

Code: Select all

x = 31, y = 21, rule = LifeHistory
4.A.2A.A.A.3A$.2A3.2A2.A3.A$2.A.A2.2A.2A.A$A.A.8A.2A$A.5A2.A.A3.A$A.
3A.A.A2.A2.A$A2.2A2.2A.A3.2A$A3.A.6A$5.2A2.2A2.2A$2.A.A2.A.A.2A.2A$2.
A2.4A.A4.A$4A6.A2.2A$5.7A.A$4.2A2.2A3.A.A$5.A.A.A2.2A$2.2A.A.A.A2.3A
11.D.2D$24.3D.2D$23.D$24.3D.2D$26.2D2.D$29.2D!

There's nothing nearly as good here as the new version, but it looked as if possibly one of them could be adapted into a workable backward eater2 recipe -- the back of this eater2 variant is sort of vaguely close to the edge of the construction envelope, and maybe the reaction in the southwest can be suppressed a little better:

Code: Select all

x = 40, y = 37, rule = LifeHistory
25.2A$24.A2.A$25.2A8$29.2A$28.A2.A$28.A2.A$29.2A5$28.A$28.A$14.2A12.A
$14.2A3$6.2A$6.2A30.2A$38.2A$38.2A$36.A$36.A$24.2D.D8.A$24.2D.3D$30.D
$24.2D.3D$23.D2.2D$2A21.2D$2A!

So the question here is, what's the simplest constellation that can build a pre-beehive on the edge of its envelope, or otherwise make the resulting active reaction? I don't think there are any glider+1sL beehive conversions that quite allow this ignition to work.

[gameoflifeboy]

A 42 cell still life just appeared!

Code: Select all

x = 16, y = 16, rule = B3/S23
oobbbobobbbooobb$
bbobbbobbooobbob$
bobbobobbooobboo$
bbboooobobobbbbb$
obbooooobobbboob$
bbboooobbobboboo$
boobbbbbbooobboo$
oooboobooobbooob$
bbbooobooobbbbob$
bboboboooobbooob$
oooooboobooooooo$
booboboboooobobo$
bobboobbobbbbooo$
bbooobooobobbooo$
ooobbooboobboobo$
ooobbbbobboobboo!
[[ THEME 7 AUTOSTART ZOOM 2 ]]

[biggiemac]

The 13th natural C1 spaceship: MWSS on HWSS 4, formed in gen 15 of this soup:

Code: Select all

#C [[ ZOOM 7 ]]
x = 16, y = 16, rule = B3/S23
ob2ob2o2b2obo$obobo2b2ob3o2bo$2bob2obob3o3bo$5bo3bob2o$3bob4o2b2ob2o$
2o2b2o2b2o5bo$8bo5b2o$2bo3bo2bo2bo$obobobo2bobob3o$bobo2bo2bob2o$obobo
6bob2o$bob3o2b2o$b8o2bo2b2o$ob4o3b7o$2obob6ob2o$2obobobo2b3obo!

[Freywa]

I think this thread has veered too far from its original purpose of providing new syntheses or other insights into objects that apgsearch has produced (it's more of a news reporting centre where ordinary objects are posted and then forgotten). Can we have a catalogue of novel syntheses and objects in one continuous string of posts here?

[chris_c]

So the question here is, what's the simplest constellation that can build a pre-beehive on the edge of its envelope, or otherwise make the resulting active reaction? I don't think there are any glider+1sL beehive conversions that quite allow this ignition to work.

I found some options to make the beehive + blinker in a single collision. Unfortunately I found no options with the trigger glider coming from the NE:

Code: Select all

x = 255, y = 100, rule = B3/S23
bo$2b2o$b2o3$70b2o$69bo2bo$64b2o4b2o$64bobo$65b2o75b3o2$221b2o$8bo129b
2o81bobo6bo$7bobo127bo2bo81bo6bobo$7bobo128b2o89bo2bo$8bo221b2o5$5b2o
222b2o$5b2o222bobo$229bo$155b2o$155bobo$155bo$77b3o$77bo$78bo49$77bo$
76bobo$75bo2bo$76b2o2$129b2o$bo8b2o65b2o50bobo6bo100b3o$obo6bo2bo63bob
o51bo6bobo$o2bo5bobo65bo59bo2bo94bo$b2o7bo127b2o94bobo$234bo2bo$235b2o
3$137b2o$137bobo$137bo$63b3o$65bo$19b2o43bo$19bobo230b2o$19bo232bobo$
252bo!

[dvgrn]

I found some options to make the beehive + blinker in a single collision. Unfortunately I found no options with the trigger glider coming from the NE...

Looks good. Given the relative messiness of this reaction, it probably makes sense to just add one more object to do a reflection, declare victory and move on -- at least until some specific construction problem shows up where this orientation of eater2 turns out to be useful.

Code: Select all

x = 52, y = 37, rule = LifeHistory
25.2C$24.C2.C$25.2C22.C.C$49.2C$50.C6$29.2C$28.C2.C$28.C2.C$29.2C$35.
2C$37.C$34.C$35.2C$28.C$28.C$14.2C12.C9.C$14.2C21.C.C$37.C.C$38.C$6.
2C$6.2C3$35.2C$35.2C$24.2D.D$24.2D.3D$30.D$24.2D.3D$23.D2.2D$2C21.2D$
2C!

[praosylen]

I just can't believe that we don't have a 13-snake yet, given that we have a ship-tie-13-snake:

Code: Select all

x = 16, y = 16, rule = B3/S23
3o2b2obo2b2obo$o2b2o2b2o3bo2bo$ob4o2b2ob3obo$4b4o2b5o$ob2o2bobo3b4o$2b
ob9obo$3ob2obob3o3bo$2bob3obo6bo$ob4o6bobo$3b2o4b2o2b2o$3bo2bobo3b3o$
4b2ob2ob4obo$o2bob2o4b3o$bob3o3bobo3bo$bo2bobobo3b3o$2o2bobob3o4bo!

Edit: As pointed out below, this isn't actually a ship-tie-13-snake. Oops.

[biggiemac]

That's a ship-tie-13-canoe, and the 13-canoe has been observed 19 times.

[gameoflifeboy]

I think this thread has veered too far from its original purpose of providing new syntheses or other insights into objects that apgsearch has produced (it's more of a news reporting centre where ordinary objects are posted and then forgotten). Can we have a catalogue of novel syntheses and objects in one continuous string of posts here?

I disagree. I am not just running the soup search to search for new glider syntheses, but also to make a list of the most common patterns in Life and other rules; to build a catalogue of natural objects, like LifeWiki, that one can browse when one finds an object and doesn't know what it is; and in general, to just make a better list of what the most common objects are in Life and other rules, and I hope other people are running apgsearch for the same reason.
I think there should be a general place for talking about what new objects have occurred naturally, and the thread titled "Soup search results" seems like the most obvious choice.

[calcyman]

7 block slow-salvo seed for a p30 reflector (which incidentally hasn't been seen before in the Catagolue):

Code: Select all

x = 49, y = 34, rule = B3/S23
37b2o$21b2o14b2o$21b2o4$32b2o3b2o$32b2o3b2o4$35b2o$35b2o12$bo41b2o$b2o
40b2o$obo6$47b2o$47b2o!

It came (practically verbatim) from this soup:

Code: Select all

x = 16, y = 16, rule = B3/S23
b3ob5o2b4o$2o2b3o2b2o2bobo$3b2o3b3ob3o$3o2b2o3bob2obo$2ob3obob2ob2obo$
4obo5bobo$3b2obobobob2obo$3b2ob2o5b3o$obob2o3bob3o$bo2b3o2bobo$2bo2b4o
2bobo$b2o2bo2b2obob3o$obo2bob2ob4o$2o2b2o2b2o3bo$ob3obobob2o3bo$2b2ob
2ob3obobo!

It's from this haul, which is one of four huge hauls uploaded by Brett Berger, which together contributed 31 billion objects. (I didn't even think Catagolue could cope with such large hauls...)

http://catagolue.appspot.com/haul/b3s23 ... efcc7bf939

Anyway, why 362 250 000 soups per haul?

[biggiemac]

I just saw this and am at a loss to how that even came about. I didn't instruct any of my instances to run that long between hauls. 362M soups at 300/second is about two weeks; I'll try to see if instances I began two weeks ago are all accounted for. I've been keeping a tab on the hauls committed to see which from the now 76-104 * 10^5 range are active and a few have yet to appear in the last 24 hours.

I'm glad catagolue didn't choke, it wasn't my intention to throw that much in at once.

[calcyman]

I just saw this and am at a loss to how that even came about. I didn't instruct any of my instances to run that long between hauls.

That implies that machine didn't have internet access for two weeks, and then internet access was restored. (There were only four superhauls, so I guess they're on the same quad-core computer.)

Considering the fact that there were three hauls of size 362 250 000 and one haul of size 363 300 000, it was almost certainly programmed with a period of 1 050 000 (the GCD of these two numbers).

I've checked http://catagolue.appspot.com/haul/b3s23/C1?committed=2 (when logged in as an administrator, it displays the last 100 hauls instead of 20), and can confirm that there have since been submissions from your payosha256 account of size 1 050 000. So it looks like it now has access to the internet.

[HartmutHolzwart]

Did anyone already notice this cute p156 multiple glider gun (pi+block):

Code: Select all

x = 42, y = 42, rule = B3/S23
20b2o$20b2o13bo$35b2o$34bobo2$38b2o$18b3o18b2o$20bo17bo$18b3o10$6bobo$
6bobo$2o4b3o31b2o$2o31b3o4b2o$33bobo$33bobo10$21b3o$3bo17bo$b2o18b3o$
2b2o2$5bobo$5b2o$6bo13b2o$20b2o!

coming form this soup:

Code: Select all

x = 32, y = 32, rule = B3/S23
bbobbobbbobooboboobooboooobobbbb$
bobbbbboobbbbbobbbbbboboboobooob$
obbobbooboobboobbbbbobooboobbbob$
bboboooooooobobbbbbbbooobobobbob$
bbboooobbobbboooooboboooobooobbo$
obboooobbbooboobbobooobobbooboob$
bboooobooboobobbbbbobbobobbboooo$
booobboooobooobbobbboobboobobbbo$
bobobbooooboboobbobooobobboooooo$
obooobboobboboooboooobbbbobooobo$
bbooboobbbbbobboobboobboobooobob$
obbobooooobobobbobobooooobobbobo$
obbbbbbobbobobobbbbobooobooobbbo$
bbooooooooboboobbbobobobbbbbbbbb$
oooboobboobbooobbobbbboobboobbbo$
bbbbobbbboobbbbbbbbboobbobbobbbo$
obbbobbobboobbbbbbbbboobbbbobbbb$
obbboobboobbbbobbooobboobboobooo$
bbbbbbbbbobobobbbooboboooooooobb$
obbbooobooobobbbbobobobbobbbbbbo$
obobbobooooobobobbobobooooobobbo$
boboooboobboobboobbobbbbbooboobb$
obooobobbbboooobooobobboobbooobo$
oooooobbobooobobbooboboooobbobob$
obbboboobboobbbobboooboooobbooob$
oooobbbobobbobbbbboboobooboooobb$
booboobbobooobobbooboobbboooobbo$
obboooboooobobooooobbbobboooobbb$
bobbobobooobbbbbbboboooooooobobb$
bobbbooboobobbbbboobbooboobbobbo$
boooboobobobbbbbbobbbbboobbbbbob$
bbbbobooooboobooboboobobbbobbobb!

[Scorbie]

Did anyone already notice this cute p156 multiple glider gun (pi+block):

Yeah, I was excited to see the rather unusual yl78 and felt nice to see that gun again!

[dvgrn]

Did anyone already notice this cute p156 multiple glider gun (pi+block):

Yeah, I was excited to see the rather unusual yl78 and felt nice to see that gun again!

Just for the record, "again" means that Scorbie turned up a p156 gun soup in a pre-Catagolue run of apgsearch, back in January. Jason Summers originally reported it at the end of October 2004, from a RandomAgar search.

Looks like the p156 is probably going to appear fairly regularly in D4_x4 symmetry searches. Given how easy it seems to be to rediscover the single beehive that can block a p156 glider pair, I wouldn't be surprised if a natural 60P312 variant shows up at some point, too.

[calcyman]

Looks like the p156 is probably going to appear fairly regularly in D4_x4 symmetry searches. Given how easy it seems to be to rediscover the single beehive that can block a p156 glider pair, I wouldn't be surprised if a natural 60P312 variant shows up at some point, too.

How? The beehive can't appear on any axis of symmetry.

[dvgrn]

Looks like the p156 is probably going to appear fairly regularly in D4_x4 symmetry searches. Given how easy it seems to be to rediscover the single beehive that can block a p156 glider pair, I wouldn't be surprised if a natural 60P312 variant shows up at some point, too.

How? The beehive can't appear on any axis of symmetry.

Darn. Well, now I would be very surprised. Will have to wait until the p156 gun shows up in an asymmetric soup, I suppose, and that will be quite a while!

[Scorbie]

Would be interesting if this is the smallest way to build this.

Code: Select all

x = 83, y = 83, rule = B3/S23
25b2o$25b2o5$27bo$26bobo$25bo2bo$26b2o4$27b2o$26bo2bo$15b3o9b2o38bo$
67bo$38bo28bo$37b3o$36bob2o$36b2o10b2o$48b2o4$74bo6b2o$48b2o18bo4bobo
5b2o$48b2o17bobo3bo2bo$67bobo4b2o$68bo4$20b2o4b2o$20b2o4b2o2$62b2o$62b
obo$63b3o$63b2o$41bo$40bobo$41bo$18b2o$17b3o$18bobo$19b2o2$55b2o4b2o$
55b2o4b2o4$14bo$7b2o4bobo$6bo2bo3bobo17b2o$2o5bobo4bo18b2o$2o6bo4$33b
2o$33b2o10b2o$43b2obo$43b3o$15bo28bo$15bo$15bo38b2o9b3o$53bo2bo$54b2o
4$55b2o$54bo2bo$54bobo$55bo5$56b2o$56b2o!

[simsim314]

New eater2 variant that can be reduced to this:

Code: Select all

x = 35, y = 49, rule = B3/S23
32b2o$31bo2bo$32bobo$33bo$29b2o$24b2o3b2o$25b2o$24bo3$b2o$o2bo$b2o8$6b
o$6bo$6bo2$10b2o$10bobo$11bo4$29b2o$29b2o$19bo$20bo$18b3o2$25b2o$25b2o
4$8bo$7bobo$6bo2bo$7b2o3$2b2o$2b2o!

[dvgrn]

New eater2 variant that can be reduced to this...

It's a nice synthesis, from this soup, right on the edge of its reaction envelope:

Code: Select all

x = 31, y = 22, rule = LifeHistory
30B$30B$28B$28B$24B.2B$23B$21B$9B2ABAB2AB$9B2AB2A2BA$15BA$9B2AB3A$10B
ABA2B$8B.A2BA$8B2.2A$7B$5B$3B2$B$B$B!
#C [[ VIEWONLY HEIGHT 240 ZOOM 11 ]]

-- but unless the same reaction can be started from a later stage, it doesn't seem too likely that it will be worth placing nine objects and synchronizing two gliders to produce that eater2.

There are ways to reduce the leftover junk, as usual:

Code: Select all

x = 64, y = 68, rule = LifeHistory
23.2A$23.2A16$47.2A$46.A2.A$47.A.A$48.A13.A$2A42.2A15.A.A$2A37.2A3.2A
16.A$40.2A$39.A3$16.2A$15.A2.A$16.2A8$21.A$21.A$21.A2$25.2A$25.A.A$
26.A4$44.2A$44.2A$34.A$35.A$33.3A2$40.2A$40.2A4$23.A$22.A.A$21.A2.A$
22.2A3$17.2A$17.2A$24.2A$24.2A!

But it's hard to know if adding more still lifes at the beginning will really make the overall recipe any shorter or less unwieldy.

There are some later stages that might be worth looking at -- e.g., at T=247 in the above reaction there's a B-heptomino, pre-honeyfarm, and a miscellaneous spark converging on a pi explosion, which produces the eater2 immediately. If those three objects could be replaced by small Spartan-ish objects (unlikely but vaguely possible, at least until we look closer...) then the resulting recipe might be pretty darn good.