Synthesising Oscillators

For discussion of specific patterns or specific families of patterns, both newly-discovered and well-known.
User avatar
GUYTU6J
Posts: 1305
Joined: August 5th, 2016, 10:27 am
Location: 拆哪!I repeat, CHINA! (a.k.a. 种花家)
Contact:

Re: Synthesising Oscillators

Post by GUYTU6J » January 4th, 2021, 11:55 am

New page and 1101st post in this thread!
Final steps for a troublesome target: a p504 loop made of four glider pair reflectors.

Code: Select all

x = 1372, y = 198, rule = B3/S23
404bobo$405b2o$405bo663bobo$1070b2o$1070bo4$876bobo190bobo4bo$464b2o
32bo377b2o192b2o3bo$464b2o33bo377bo192bo4b3o9bo$440b2o32bo22b3o362bobo
221bo$441bo31bobo3bo146bo236b2o221b3o$438b3o32bobo2bobo143bobo236bo$
438bo35bo4b2o144b2o$439bobo840bo$438b2ob3o838b3o8b2o$444bo426bobo411bo
7b2o$443b2o421bobo2b2o411b2o$630bo23bo212b2o3bo$631b2o19b2o213bo218b2o
199bo$66bo563b2o21b2o26b2o403b2o198bobo$65bo583bo31b2o603bobo$65b3o
437bo144b2o5b2o32bo595bo$503bobo143b2o7bo31bobo3bo383bo$40bo361b3o99b
2o24bo124b3o32bobo2bobo381bobo$41bo362bo123b2o125bo35bo4b2o181b2o198bo
bo$39b3o361bo125b2o118bo6bobo220b2o199bo$31bobo614bobo4b2ob3o194b2o32b
o$32b2o607bobo4b2o11bo194bo31bobo3bo405b2o$32bo609b2o16b2o191b3o32bobo
2bobo404b2o$236bobo150b2o251bo210bo35bo4b2o380b2o32bo$87b2o148b2o34b2o
113bobo463bobo420bo31bobo3bo$62bo24b2o148bo35b2o115bo462b2ob3o234b2o
179b3o32bobo2bobo$62bobo32bo151b2o32bo575bo233b2o179bo35bo4b2o$62b2o
32bobo3bo138bo8bo31bobo3bo569b2o209b2o32bo171bobo$96bobo2bobo123bo12bo
bo4b3o32bobo2bobo132b2o646bo31bobo3bo165b2ob3o$35bo61bo4b2o124b2o10bob
o4bo35bo4b2o133bo643b3o32bobo2bobo170bo$36b2o27bo161b2o12bo6bobo172bob
o641bo35bo4b2o169b2o$35b2o27bobo180b2ob3o171b2o642bobo$64bobo164b2o20b
o205b2o285bo320b2ob3o$65bo165b2o19b2o204bobo284bo327bo$459bo285b3o324b
2o$59b2o186b2o245bo349b2o451bobo$58bo2bo184bobo243b3o350bo452b2o$58bob
o185b2o243bo353bobo450bo$59bo431b2o245bobo105b2o417b2o$738b2o526bo$
432b2o205b2o98bo526bobo$431bobo16b2o188bo626b2o$392b2o37bo18b2o188bobo
415b2o$391bobo36b2o47bo161b2o416bo$148bo243bo86b3o194b2o381bobo$146b2o
334bo192bobo159b2o221b2o$147b2o332b2o193bo161bo$391b2o318bo126bobo$
137bobo250bo2bo315b3o127b2o417b2o$137b2o252b2o315bo21bo143b2o383bo$
138bo569b2o18b2o143bobo383bobo$45b2o184b2o281bob2o211b2o143bo385b2o$
46bo102bo82bo281b2obo131b2o258bo141b2o242b2o$46bobo100bobo80bobo282bob
o128bobo16b2o238b3o142bo241bobo$47b2o100b2o82b2o89bobo187b2o2b2o89b2o
37bo18b2o237bo145bobo240bo$82b2o184b2o54b2o188bo93bobo36b2o47bo209b2o
145b2o275bo$81bobo183bobo55bo186bobo94bo86b3o31b2o356b2o238b3o$82bo
185bo119b2o122b2o185bo30bobo2bobo109b2o238bobo237bo$117bo185bo17b2o65b
2o308b2o31bo3b2o109bobo16b2o221bo238b2o$115b3o183b3o17b2o285b2o126bo
70b2o37bo18b2o52bo203bo$114bo185bo306bo2bo128bo66bobo36b2o47bo22b3o
201b3o144b2o$114b2o184b2o306b2o128b2o67bo86b3o19bo203bo146bobo16b2o$
419b2o317bobo156bo18b2o202b2o106b2o37bo18b2o52bo$55b2o184b2o88bo87bobo
58b2o249bob2o161b2o329bobo36b2o47bo22b3o$54bobo16b2o165bobo16b2o68b2o
89bo59b2o249b2obo71b2o253b2o165bo86b3o19bo$15b2o37bo18b2o126b2o37bo18b
2o69b2o402bobo68bo2bo251bobo16b2o237bo18b2o$14bobo36b2o47bo97bobo36b2o
47bo442b2o2b2o69b2o213b2o37bo18b2o52bo183b2o$15bo86b3o96bo86b3o34b2o
404bo288bobo36b2o47bo22b3o93b2o$105bo185bo32bo2bo401bobo197bob2o88bo
86b3o19bo95bo2bo$104b2o184b2o33b2o278b2o122b2o198b2obo178bo18b2o95b2o$
14b2o184b2o403b2o325bobo175b2o$13bo2bo114b2o66bo2bo225b2o59bo439b2o2b
2o85b2o328bob2o$14b2o114bo2bo66b2o226b2o58bobo438bo89bo2bo327b2obo$
131bobo183b2o170b2o436bobo90b2o331bobo$132bo184bobo3bob2o309b2o165b2o
122b2o421b2o2b2o$141b3o174b2o3b2obo309bobo58b2o104b2o338bob2o203bo$
141bo184bobo308bo59b2o86bo357b2obo201bobo$136b2o4bo180b2o2b2o191b2o
264b2o163bo194bobo26bo5bo42b2o122b2o$135bo2bo14b2o168bo72b2o122b2o263b
2o162b2o192b2o2b2o24b2o4b2o43b2o$136b2o15bobo165bobo71bobo436b2o114b2o
191bo30b2o4b2o56b2o130b2o$11b2o140bo43b2o122b2o72bo438bobo58b2o244bobo
95b2o129bo$11b2o184b2o191b2o2b2o398bo40bo59b2o47bo72b2o122b2o95bo129bo
bo$390bobo402b2o145b2o73b2o236b2o111b2o$392bob2o249b2o59bo87b2o147b2o
228b2o80bobo58b2o46b2o$392b2obo249b2o58bobo465bobo35b2o43bo59b2o45bo2b
o$42b2o184b2o476b2o465bo37b2o151b2o$42bobo58b2o123bobo58b2o226b2o276b
2o147b2o102b2o$43bo59b2o124bo59b2o225bo2bo276b2o145b2o51bo51bobo58b2o
46b2o$517b2o276bo47b2o59bo40bo48bobo7b2o43bo59b2o45bo2bo$427b2o308b2o
104b2o58bobo89b2o7b2o151b2o$427bo185b2o122b2o49b2o114b2o310b2o151b2o$
428b3o86bo94bobo174b2o162b2o260bo2bo45b2o59bo43b2o$430bo47b2o36bobo93b
o175bo163b2o262b2o46b2o58bobo$458b2o18bo37b2o89b2o2b2o341bo257b2o111b
2o$51b2o59bo124b2o59bo159b2o16bobo128bobo325b2o72b2o151b2o7b2o38bobo
129bo$51b2o58bobo123b2o58bobo176b2o131bob2o198b2o122b2o71bo2bo45b2o59b
o43b2o7bobo37bo129b2o$112b2o184b2o309b2obo197bobo196b2o46b2o58bobo51bo
38b2o130b2o$417b2o184bobo204bo307b2o236b2o$418bo185b2o128b2o69b2o2b2o
183bo237b2o122b2o$415b3o186bo128bo2bo68bobo184bobo236bobo$143b2o184b2o
84bo191bo126b2o71bob2o182b2o236bo$2bo140b2o60b2o122b2o119bo156b2o3bo
31b2o161b2obo338b2o75b2o2b2o$obo15b2o184bobo242bobo154bobo2bobo30bo
380b2o122b2o75bobo$b2o14bo2bo183bo244b2o161b2o31b3o86bo197b2o90bobo
201bob2o$13bo4b2o179b2o2b2o279b2o161bo47b2o36bobo195bo2bo51b2o4b2o30bo
203b2obo$14bo184bobo282bobo188b2o18bo37b2o197b2o53b2o4b2o24b2o2b2o$12b
3o186bob2o3b2o276bo188b2o16bobo146b2o142bo5bo26bobo331b2o$23bo177b2obo
3bobo275b2o205b2o127b2o18bo178bob2o327bo2bo$22bobo184b2o402b2o208bo19b
3o86bo88b2obo328b2o$22bo2bo114b2o184b2o286b2o18b2o184b3o22bo47b2o36bob
o329b2o$23b2o114bo2bo182bo2bo190bo93bo21bo184bo52b2o18bo37b2o213b2o95b
2o18bo$140b2o184b2o191bobo110b3o238b2o16bobo251bo2bo95bo19b3o86bo$50b
2o149b2o33b2o281b2o111bo258b2o253b2o93b3o22bo47b2o36bobo$50bo149bo2bo
32bo430bo388b2o183bo52b2o18bo37b2o$51b3o86bo60b2o34b3o86bo339bobo163b
2o202b2o18bo237b2o16bobo$53bo47b2o36bobo97bo47b2o36bobo338b2o165bo203b
o19b3o86bo165b2o$81b2o18bo37b2o55b2o69b2o18bo37b2o52b2o125bo194b2o127b
3o201b3o22bo47b2o36bobo$81b2o16bobo95b2o68b2o16bobo92b2o123bo195bobo
126bo203bo52b2o18bo37b2o106b2o$99b2o95bo88b2o92bo24b2o99b3o195bo161bo
221b2o16bobo146bo$404bobo197bo98b2o159bobo238b2o144b3o$40b2o184b2o176b
o199b2o258b2o385bo$41bo185bo375bobo293b2o145b2o238bo$38b3o164b2o17b3o
672bobo145bo237bobo$38bo166b2o17bo676bo142b3o238b2o$73bo185bo641b2o
141bo275b2o$72bobo127bo55bobo204b2o129b3o480bo240bobo$72b2o128b2o54b2o
205bo132bo479bobo241bo$5b2o100b2o92bobo89b2o171b3ob2o125bo480b2o242b2o
$4bobo100bobo183bobo172bobo642b2o$6bo102bo185bo133b2o4bo35bo641bobo$
109b2o184b2o132bobo2bobo32b3o643bo$17bo412bo3bobo31bo423b2o221b2o$17b
2o391b3o22bo32b2o422bobo$16bobo391bo33b2o448bo$411bo32b2o448b2o417b2o$
7b2o1304bobo$8b2o691bo613bo$7bo674b2o16b2o613b2o$682bo11b2o4bobo403b2o
175bo$683b3ob2o4bobo410bobo173b2o$685bobo6bo413bo173bobo$504bo141b2o4b
o35bo419b2o$503b2o141bobo2bobo32b3o191b2o$96bo406bobo141bo3bobo31bo7b
2o185bo$95bobo182b2o370bo32b2o5b2o187b3ob2o$94bo2bo181bobo379b2o31bo
188bobo415b2o$95b2o182b2o380b2o26b2o21b2o130b2o4bo35bo414bo$690b2o19b
2o131bobo2bobo32b3o415b3ob2o$90bo183b2o19b2o392bo23bo131bo3bobo31bo
420bobo$89bobo182bo20b2o553bo32b2o209b2o169b2o4bo35bo$89bobo27b2o154b
3ob2o578b2o233bo170bobo2bobo32b3o$90bo27b2o157bobo6bo12b2o558b2o234b3o
b2o165bo3bobo31bo$52b2o4bo61bo117b2o4bo35bo4bobo10b2o797bobo171bo32b2o
$52bobo2bobo178bobo2bobo32b3o4bobo12bo416b2o339b2o4bo35bo179b2o$53bo3b
obo32b2o145bo3bobo31bo8bo430bobo338bobo2bobo32b3o179b2o$58bo32bobo150b
o32b2o438bo341bo3bobo31bo$67b2o24bo159b2o35bo773bo32b2o$67b2o184b2o34b
2o581bo200b2o$289bobo575bo3b2o200b2o$123bo743b2o2bobo$122b2o742bobo
425bo$122bobo1168bobo$114b3o1176bobo$114bo1179bo$115bo760bo210bo$875b
2o209bobo207b2o$88b3o784bobo208bobo198b2o7bo$90bo771bo224bo199b2o8b3o$
89bo772b2o435bo$861bobo$1080b2o$1080b2o8$1079b3o$1081bo$1080bo9b3o4bo$
1092bo3b2o$1091bo4bobo4$1097bo$1096b2o$1096bobo!
I like the beautiful mechanism, but lots of catalysts within are placed inconveniently for a synthesis. How many gliders will the oscillator cost when completed?
Lifequote:
In the drama The Peony Pavilion, Tang Xianzu wrote: 原来姹紫嫣红开遍,似这般都付与断井颓垣。
(Here multiflorate splendour blooms forlorn
Midst broken fountains, mouldering walls.)

MathAndCode
Posts: 3631
Joined: August 31st, 2020, 5:58 pm

Re: Synthesising Oscillators

Post by MathAndCode » January 18th, 2021, 12:17 am

Did whoever synthesized Pentoad 1H2 not realize that the part in the upper right is just a block and a century? For reference, here is the activation stage of the current synthesis in Catagolue:

Code: Select all

x = 35, y = 31, rule = B3/S23
17bobo$18b2o$18bo$26bo$25bo$19bo5b3o$19bobo$19b2o2$2bo$obo5bo$b2o6b2o
12b2o$8b2o7bo4bobo$17bo3bobo$17bo2bobo6bo$21bo5b2o$4b2o22b2o$3bobo$5b
o$10b2o15bo$9bobo14b2o$9bo8b2o6bobo$8b2o8bobo12b2o$18bo13b2o$34bo$13b
o$13b2o$12bobo$21bo$20b2o$20bobo!
Here is a five-glider synthesis of the century and block:

Code: Select all

x = 16, y = 14, rule = B3/S23
11bobo$11b2o$6bo5bo$6bobo$6b2o5$7b2o$6b2o6b2o$3o5bo4b2o$2bo12bo$bo!
Here is a demonstration of compatibility:

Code: Select all

x = 29, y = 27, rule = B3/S23
14bo$15bo12bo$13b3o5bo4b2o$19b2o6b2o$20b2o$bo$2bo$3o4bobo$8b2o$8bo10b
2o$19bobo$19bo5bo$3b3o18b2o$5bo18bobo$4bo$9b2o$8bobo$8bo7b3o$7b2o7bo$
17bo2$12b2o$13b2o$12bo$19b2o$18b2o$20bo!
A four- or three-glider synthesis of the century and block may be possible.
I have reduced the cost of universal construction to seventeen gliders and probably to sixteen. All that remains is for the universal operations to be found.

mniemiec
Posts: 1241
Joined: June 1st, 2013, 12:00 am

Re: Synthesising Oscillators

Post by mniemiec » January 18th, 2021, 1:13 am

MathAndCode wrote:
January 18th, 2021, 12:17 am
Did whoever synthesized Pentoad 1H2 not realize that the part in the upper right is just a block and a century? For reference, here is the activation stage of the current synthesis in Catagolue: ...
That looks like mine from 2013-08-29. Yes, I knew about the bookend/century and the block; I just couldn't find any convenient way to make both of them close enough together to work here. I have (manually) kept track of a fair number of 3 glider collisions that make constellations of two stable objects, I've never tried to systematically keep track of ones that make unstable patterns, or combinations of pairs of such (e.g. honeyfarm interacting with block).

hkoenig
Posts: 163
Joined: June 20th, 2009, 11:40 am

Re: Synthesising Oscillators

Post by hkoenig » January 18th, 2021, 12:19 pm

I set up a searchable database from a posting of 3 Glider collisions that was published here three years ago. I've found it useful to find candidate 3 Gliders constructions, by searching for final census matches.

That Century and Block results in a single Boat. Which is unfortunate, as the database shows 1167 ways that one Boat can be constructed. Adding the criteria that the lifetime must be at least 25 gens narrows that to 681, but what's left is not a good way to look for candidates. Anyhow, attached is a tab delimited file of all the candidates.
Attachments
Century+Block.txt
(46.53 KiB) Downloaded 14 times

User avatar
dvgrn
Moderator
Posts: 7634
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Synthesising Oscillators

Post by dvgrn » January 18th, 2021, 1:25 pm

hkoenig wrote:
January 18th, 2021, 12:19 pm
I set up a searchable database from a posting of 3 Glider collisions that was published here three years ago. I've found it useful to find candidate 3 Gliders constructions, by searching for final census matches.

That Century and Block results in a single Boat. Which is unfortunate, as the database shows 1167 ways that one Boat can be constructed. Adding the criteria that the lifetime must be at least 25 gens narrows that to 681, but what's left is not a good way to look for candidates. Anyhow, attached is a tab delimited file of all the candidates.
I've been meaning to get back to dealing with the problem of applying octohashes to the 3G collisions (and probably also to wildmyron's additional subset of 4G collisions, while I'm at it.)

Once the octohash database is built, it will be very quick and easy to search directly for the hash of the century+block pattern itself, or whichever of its descendants is really needed. This has worked really well for the 2-object seed database. With nine-character hashes, I didn't see any hash collisions at all, and searches across the full database complete in just a second or three.

What we have currently is synthesise-patt.py, but that database unfortunately includes only the 3G collisions, with quite a few duplicates.

Duplicates were removed by wildmyron's synthesise-constellation project, because the database was moved to Shinjuku's .sjk format, which uses a unique representation of each collision. But synthesise-constellation only searches the final census, so it has the same problem with false positives as your database, for cases like this.

I did try running synthesise-patt.py against the century+block, but didn't get any results. Seems like there are likely to be results among the 4G collisions, but that will require rebuilding the hash database. The hashes for synthesise-patt.py are currently just the sequence of pattern populations, so that's another significant source of false positives -- which we can remove by switching to multibyte "octohash" hashes (or similar).

MathAndCode
Posts: 3631
Joined: August 31st, 2020, 5:58 pm

Re: Synthesising Oscillators

Post by MathAndCode » January 18th, 2021, 1:52 pm

mniemiec wrote:
January 18th, 2021, 1:13 am
That looks like mine from 2013-08-29. Yes, I knew about the bookend/century and the block; I just couldn't find any convenient way to make both of them close enough together to work here. I have (manually) kept track of a fair number of 3 glider collisions that make constellations of two stable objects, I've never tried to systematically keep track of ones that make unstable patterns, or combinations of pairs of such (e.g. honeyfarm interacting with block).
Using a spark from the block's formation to activate the century was clever. I understand the difficulty of getting the block and century in directly now that you've explained it; the method that I used required a fair bit of cleverness as well due to using a different parent of the century's grandchild from the one that appears in the oscillator and having to set up the block catalysis in a nonobvious way because of that.
By the way, I independently discovered that p5 partial when I was just starting to make conduits (and didn't know what I was doing).
I have reduced the cost of universal construction to seventeen gliders and probably to sixteen. All that remains is for the universal operations to be found.

User avatar
goldenratio
Posts: 264
Joined: July 26th, 2020, 10:39 pm

Re: Synthesising Oscillators

Post by goldenratio » February 16th, 2021, 2:30 pm

Cheaper, but lower clearance spark-first fumarole inserter:

Code: Select all

x = 30, y = 16, rule = B3/S23
5bo18bo$3bobo18bobo$4b2o18b2o6$11bo6bo$12bo4bo$10b3o4b3o2$2bo24bo$obo
6bo10bo6bobo$b2o7bo8bo7b2o$8b3o8b3o!
For instance, it improves the fumarole on Rich's p16 synthesis:

Code: Select all

x = 30, y = 35, rule = B3/S23
5bo18bo$3bobo18bobo$4b2o18b2o6$11bo6bo$12bo4bo$10b3o4b3o2$2bo24bo$obo
6bo10bo6bobo$b2o7bo8bo7b2o$8b3o8b3o10$12b3o3b3o$11bo3bobo3bo$11bo3bob
o3bo$10bob4ob4obo$10b2o9b2o3$14b2ob2o$13bobobobo$14bo3bo!
Temporarily leaving for an undetermined amount of time

MathAndCode
Posts: 3631
Joined: August 31st, 2020, 5:58 pm

Re: Synthesising Oscillators

Post by MathAndCode » February 22nd, 2021, 5:37 pm

mniemiec wrote:
June 30th, 2015, 5:44 pm
The essential predecessor is a pair of interacting pi heptominos on each side. The left ones must be formed unobtrusively, while the right ones require a very specific predecessor which occurs rarely. I have a partial solution that makes three of them, but good luck in squeezing the fourth one in!
I noticed that moving the bottom house parent two cells to the left makes the pattern closer to stable.

Code: Select all

x = 9, y = 8, rule = B3/S23
obo$obob2ob2o$3o2b3o$6bo2$4bo$3b3o$2b2ob2o!
This would be stable if the house were lengthened at the right time, but one of the gliders coming from the bottom-left threatened to interfere with getting the spark in, so I hooked up the highest-clearance spark inserter that I know, and it worked.

Code: Select all

x = 99, y = 71, rule = B3/S23
44bo$39bo5b2o$40bo3b2o$38b3o$96bobo$47bobo46b2o$48b2o47bo$48bo$2bo$ob
o$b2o5$24bo$25bo$23b3o2$60bo$58b2o$59b2o31$32bo$32b2o4b2o$31bobo5b2o$
38bo50b3o$89bo$90bo4$46bo$46b2o47bo$45bobo46b2o$94bobo2$42b2o$43b2o$26b
3o13bo$28bo$27bo!
Before hooking up the spark inserter, I checked the oscillator's Catagolue page, and that oscillator indeed has not been previously synthesized, so I shall submit my synthesis.
I have reduced the cost of universal construction to seventeen gliders and probably to sixteen. All that remains is for the universal operations to be found.

Post Reply