Pattern of the Year 2016 (Entries)

For general discussion about Conway's Game of Life.
User avatar
Alexey_Nigin
Posts: 323
Joined: August 4th, 2014, 12:33 pm
Location: Ann Arbor, MI
Contact:

Pattern of the Year 2016 (Entries)

Post by Alexey_Nigin » January 8th, 2017, 3:43 pm

Even though most people agree that 2016 was a terrible year overall, it was certainly quite a good year for the Game of Life. Many discoveries were made, and it is only logical to select the coolest one of them by organizing yet another Pattern of the Year competition.

As usual, we start by compiling a list of entries. You are welcome to enter any reasonable number of patterns or pattern collections that you consider notable enough for this competition. It doesn't matter whether the patterns were discovered by you or not. There are only two strict requirements: every entry should work under B3/S23 and be discovered in 2016. In addition, if everyone thinks that the pattern you entered should not be in this competition, I will remove it. You do not need to provide any attributions, links or descriptions at this point.

Here are my entries:

Copperhead and all related stuff
Caterloopillars
New gun periods
New reflector mechanism

Here are combined entries from everybody else:

8-bit computer
Spaghetti Monster
15-bit SL syntheses
Rich's p16
Pattern with parents but no grandparents and relatives
Small GoEs
Compact stable herschel splitter
Statorless p5
Aperiodic tiles
p7 sparker

Rejected entries:

Meta-meta-blinker (Sure, it's an impressive demonstration of Golly and .mc format, but it seems quite trivial from CGoL point of view, right?)

When there are no new entries for 48 hours, the next part of the competition will start.

P. S. You are probably not wondering why I haven't discovered a single notable pattern this year, but in case you do, I can assure you that I haven't given up cellular automata. On the contrary, I am now working on a big project, the details of which will be revealed in a few days.
Last edited by Alexey_Nigin on January 22nd, 2017, 3:13 pm, edited 3 times in total.
There are 10 types of people in the world: those who understand binary and those who don't.

Naszvadi
Posts: 388
Joined: May 7th, 2016, 8:53 am
Contact:

Re: Pattern of the Year 2016 (Entries)

Post by Naszvadi » January 8th, 2017, 3:56 pm

Enter this 8bit computer: viewtopic.php?f=2&t=2561

User avatar
gmc_nxtman
Posts: 1148
Joined: May 26th, 2015, 7:20 pm

Re: Pattern of the Year 2016 (Entries)

Post by gmc_nxtman » January 8th, 2017, 3:57 pm


muzik
Posts: 3522
Joined: January 28th, 2016, 2:47 pm
Location: Scotland

Re: Pattern of the Year 2016 (Entries)

Post by muzik » January 8th, 2017, 4:36 pm

How about the fireship and the related c/10 technology?
Bored of using the Moore neighbourhood for everything? Introducing the Range-2 von Neumann isotropic non-totalistic rulespace!

User avatar
Alexey_Nigin
Posts: 323
Joined: August 4th, 2014, 12:33 pm
Location: Ann Arbor, MI
Contact:

Re: Pattern of the Year 2016 (Entries)

Post by Alexey_Nigin » January 8th, 2017, 6:11 pm

muzik wrote:How about the fireship and the related c/10 technology?
It is included with the copperhead.
There are 10 types of people in the world: those who understand binary and those who don't.

User avatar
gmc_nxtman
Posts: 1148
Joined: May 26th, 2015, 7:20 pm

Re: Pattern of the Year 2016 (Entries)

Post by gmc_nxtman » January 8th, 2017, 6:27 pm

Alexey_Nigin wrote:Questionable entries (tell me what you think about them):

Compact stable herschel splitter (Is it useful? Is it better than alternatives?)
It's by far the most compact and applicable splitter discovered so far.
Alexey_Nigin wrote: Statorless p5 (First of its kind?)
p7 sparker (What's notable about it?)
I guess these might not be entirely suited to the place of "pattern of the year", but I think they are definitely interesting in their own right. And the p7 sparker may have many applications in hassler oscillators.

User avatar
gameoflifeboy
Posts: 474
Joined: January 15th, 2015, 2:08 am

Re: Pattern of the Year 2016 (Entries)

Post by gameoflifeboy » January 9th, 2017, 12:46 am

I'm surprised no one's mentioned the pattern with parents but no grandparents yet. It definitely deserves to be an entry because it was the first solution to a 44-year old problem. I'm nominating it right now.

User avatar
dvgrn
Moderator
Posts: 6046
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Pattern of the Year 2016 (Entries)

Post by dvgrn » January 9th, 2017, 1:53 am

gameoflifeboy wrote:I'm surprised no one's mentioned the pattern with parents but no grandparents yet. It definitely deserves to be an entry because it was the first solution to a 44-year old problem. I'm nominating it right now.
I don't know if they should be separate entries or not -- seems like maybe they'd be better as a family group -- but also found in 2016 were a grandparents-but-no-great-grandparents pattern, and a great-grandparents-but-no-great-great-grandparents pattern.

I'd like to nominate Steven Eker's record-breaking Garden of Eden patterns -- in particular, an 89-cell orphan, and another new orphan pattern that's only 5 cells high.

Also worth a nomination in my book: NotLiving's tiles that create an infinite aperiodic P1 agar, but no infinite periodic agars.

User avatar
simsim314
Posts: 1731
Joined: February 10th, 2014, 1:27 pm

Re: Pattern of the Year 2016 (Entries)

Post by simsim314 » January 9th, 2017, 8:32 am

Here are my thoughts for nomination:

1. In large pattern construction arena, I think caterloopillar beats 8 bit computer, although I do admit it's a cool idea I was thinking of myself, and it deserves attention. Yet high speed constructible ships with totally new design pattern (strange loop) is more significant in my opinion. For a while now we had only Gemini-like design ships, or caterpillar designs, caterloopillar is a whole new concept, as well as solving so many relatively high speeds.

2. In small and significant patterns category, copperhead is definitely the first place. Just because it's so small and elegant, unlike spaghetti monster. Spaghetti monster is second in this category, as it was speed that was searched already for couple of years without success. Statorless p5 is third, and new p16 is fourth.

3. In theoretical challenges, like GOE and grandparent problem, I think the first pattern that was proven to have parents but no grandparents is definitely a new breakthrough. Aperiodic p1 is also very interesting.

My final list:

1. Caterloopillar
2. GOE without grandparent
3. Copperhead
4. Spaghetti monster
5. Herschel splitter
6. Statorless p5
7. 8 bit computer
8. Aperiodic p1

HartmutHolzwart
Posts: 423
Joined: June 27th, 2009, 10:58 am
Location: Germany

Re: Pattern of the Year 2016 (Entries)

Post by HartmutHolzwart » January 9th, 2017, 9:02 am

I second the list of simsim314, i.e. same votes.

Hartmut

User avatar
Alexey_Nigin
Posts: 323
Joined: August 4th, 2014, 12:33 pm
Location: Ann Arbor, MI
Contact:

Re: Pattern of the Year 2016 (Entries)

Post by Alexey_Nigin » January 9th, 2017, 10:42 am

Please do not vote in this thread. The votes you cast here will not be counted.
There are 10 types of people in the world: those who understand binary and those who don't.

User avatar
dvgrn
Moderator
Posts: 6046
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Pattern of the Year 2016 (Entries)

Post by dvgrn » January 9th, 2017, 1:08 pm

Alexey_Nigin wrote:Please do not vote in this thread. The votes you cast here will not be counted.
Yes, it hasn't even been 48 hours without a new nomination yet. Hold those thoughts!

What about the synthesis of 30P4H2V0.4? It was mentioned in the 2015 pattern-of-the-year list, but technically was only completed in 2016. It's an impressive-looking example of Kazyan-style synthesis, and unlike last year we don't have a lot of other new spaceship recipes to compete with it (do we?)

User avatar
Alexey_Nigin
Posts: 323
Joined: August 4th, 2014, 12:33 pm
Location: Ann Arbor, MI
Contact:

Re: Pattern of the Year 2016 (Entries)

Post by Alexey_Nigin » January 9th, 2017, 2:44 pm

dvgrn wrote:What about the synthesis of 30P4H2V0.4? It was mentioned in the 2015 pattern-of-the-year list, but technically was only completed in 2016. It's an impressive-looking example of Kazyan-style synthesis, and unlike last year we don't have a lot of other new spaceship recipes to compete with it (do we?)
I don't think that any pattern should be able to enter two Pattern of the Year competitions.
There are 10 types of people in the world: those who understand binary and those who don't.

User avatar
dvgrn
Moderator
Posts: 6046
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Pattern of the Year 2016 (Entries)

Post by dvgrn » January 9th, 2017, 3:59 pm

Alexey_Nigin wrote:I don't think that any pattern should be able to enter two Pattern of the Year competitions.
Okay -- you're the boss of this thread. I was definitely trying to sneak that glider synthesis in on a technicality: that 85-glider pattern definitely didn't exist in 2015. In the Pattern of the Year 2015 thread, 30P4H2V0.4 was mentioned as one of a long list of spaceships synthesized in 2015 -- probably shouldn't have been there, since the synthesis wasn't quite complete yet.

However, now that it's been mentioned again, there's no point in having it be an official contender -- it wouldn't win anything except maybe the New Small Spaceship Synthesis category, anyway...!

muzik
Posts: 3522
Joined: January 28th, 2016, 2:47 pm
Location: Scotland

Re: Pattern of the Year 2016 (Entries)

Post by muzik » January 12th, 2017, 6:55 am

that's been about 48 hours
Bored of using the Moore neighbourhood for everything? Introducing the Range-2 von Neumann isotropic non-totalistic rulespace!

User avatar
Alexey_Nigin
Posts: 323
Joined: August 4th, 2014, 12:33 pm
Location: Ann Arbor, MI
Contact:

Re: Pattern of the Year 2016 (Entries)

Post by Alexey_Nigin » January 22nd, 2017, 3:49 pm

The second phase of the competition is documenting entries. Anyone can submit description of an entry. It is recommended that you include a full list of authors, although you may list only the most important contributors if there are too many of them. Also, don't hesitate to suggest edits to existing descriptions. I take this part of the competition very seriously (I guess documenting entries is even more important than which one wins in the end), so this phase will last until I am satisfied with the descriptions.

This is just a sandbox, and the descriptions provided here may not be accurate. For accurate descriptions, see the voting thread.

#10 - Copperhead: In addition to being the first of its speed, this c/10 orthogonal spaceship is so small that it is surprising nobody discovered it before. Its small size allowed the creation of a 22-glider synthesis, and a gun was made soon afterwards. A pseudo-tagalong found a bit later made it possible to create c/10 puffers and rakes. Much of copperhead's fame is due to the mystery surrounding its discovery: it was posted by a newly-joined anonymous user who got inactive afterwards. By zdr, Simon Ekström, et al.

#6 - Caterloopillars: Previously, all engineered spaceships had speeds that were either fixed (Caterpillar) or adjustable but always very low (Gemini). Caterloopillars have the best of both worlds: their speed is variable, and the fastest ones can travel as quickly as c/4. So far all caterloopillars have been orthogonal, but a project to build one with adjustable slope is underway. By Michael Simkin, David Bell, et. al.

#11 - New gun periods: In three days, three new true gun periods were discovered: p61, p58, and p57. By Luka Okanishi, thunk, Matthias Merzenich, and Chris Cain.

#8 - Bumper: This reflector mechanism can be attached to a variety of sparky oscillators to make reflectors of the corresponding periods. Since the resulting reflectors are color-preserving, they can be used in many places where the color-changing snark does not work. As such, they solve various "my glider is in the wrong phase" wiring problems. A small wave of record-breaking guns followed soon after the bumper's debut. By Tanner Jacobi.

#1 - 8-bit Computer: It is the first 8-bit computer in any cellular automaton, and easily the most realistic computer ever constructed in Life. It has a ROM which contains the program, consisting of 32 instructions which are stored in 21 bits each. For RAM, it has a register bank containing 8 8-bit registers. For output, it has a "printer" with 8 pixels. By Coban.

#13 - Spaghetti Monster: This is the first spaceship with a speed of 3c/7, which has been one of the most sought-after speeds in the last 15 years. The ship was discovered with knightt, a relatively new search program. The search took about two months, even though the width was continiously tweaked to make it faster. By Tim Coe.

#2 - 15-bit SL syntheses: A project to find syntheses of all 15-bit still lifes in under 15 gliders turned out to be a very prolific collaboration. Using Catagolue soups along with known and new converter mechanisms, several hundred new syntheses were created. The goal was achieved in just 41 days. By BlinkerSpawn, Goldtiger997, Bob Shemyakin, Mark Niemiec, Martin Grant, Chris Cain, et al.

#4 - Rich's p16: This is a small p16 oscillator that appeared naturally. It is compact for its period and has easily accessible sparks. Most notably, it can filter glider streams similarly to the blocker, which led to reductions in many guns that previously utilized other filter mechanisms. By Rich Holmes.

#3 - Grandfather-less pattern: "Is there a configuration which has a father but no grandfather?" This is the statement of the grandfather problem, one of the longest-standing open problems in Life, having been unsolved for decades in spite of a $50 prize offer from John Conway Himself in September 1972. It was only this year that the first grandfather-less pattern was discovered using a SAT solver. By mtve.

#5 - Small GoEs: A few record-breaking Gardens of Eden and orphans were found including an orphan with only 89 defined cells. By Steven Eker.

#7 - Herschel splitter: This is an elegant solution to an old circuitry problem: a direct Herschel signal splitter that recovers in less than a hundred ticks. People had been looking for something exactly like this for very nearly twenty years, ever since the original 1996-97 universal set of Herschel conduits. It was made by extending an Fx119 conduit with a partial copy of itself to extract the second Herschel output. By Luka Okanishi.

#9 - Statorless p5: This is the first statorless oscillator of its period. Since the period is prime, the lack of stator implies that the oscillator is strictly volatile, which is a rare property. By Josh Ball.

#12 - Aperiodic tiles: This is an implementation of a particular set of Wang tiles in the Game of Life. The Life tiles form a stable pattern if and only if the corresponding arrangement of Wang tiles is valid. Since this set of Wang tiles can only form aperiodic arrangements, so can the Life tiles. By NotLiving.

#14 - p7 sparker: This pattern adds a new relatively strong HW emulator to the p7 collection, which previously only had the pipsquirter. It was used to complete a p28 wick with no prior stabilization. By Dongook Lee.
Last edited by Alexey_Nigin on March 24th, 2017, 3:03 am, edited 37 times in total.
There are 10 types of people in the world: those who understand binary and those who don't.

User avatar
BlinkerSpawn
Posts: 1929
Joined: November 8th, 2014, 8:48 pm
Location: Getting a snacker from R-Bee's

Re: Pattern of the Year 2016 (Entries)

Post by BlinkerSpawn » January 22nd, 2017, 4:02 pm

RIch's p16: A small p16 oscillator that appeared naturally and was found by Rich Holmes. Notable in that it is compact for its period and has easily accessible sparks, so it can filter glider streams similarly to the blocker and similar oscillators, leading to reductions in many guns that previously utilized other filter mechanisms. Rich's p16 also has a relatively straightforward 12 (right?) glider synthesis based upon a synthesis by BlinkerSpawn, later refined by Mark Niemiec and subsequently by Extrementhusiast.
LifeWiki: Like Wikipedia but with more spaceships. [citation needed]

Image

User avatar
blah
Posts: 244
Joined: April 9th, 2016, 7:22 pm

Re: Pattern of the Year 2016 (Entries)

Post by blah » January 23rd, 2017, 2:41 am

8-bit Computer: The first 8-bit computer in a cellular automaton (I'm pretty sure), and easily the most realistic computer ever constructed in life. Which isn't hard when your only competition is a turing machine and an MRM. It has a ROM which contains a program, consisting of 32 instructions which are stored in 21 bits each. For RAM, it has a register bank containing 8 8-bit registers. For output, it has a "printer" with 8 pixels. Like all the greatest discoveries and inventions it was posted by a member who hasn't been very active before or since, with terrible English. By Coban.
succ

User avatar
simsim314
Posts: 1731
Joined: February 10th, 2014, 1:27 pm

Re: Pattern of the Year 2016 (Entries)

Post by simsim314 » January 23rd, 2017, 7:54 am

Cateloopillars: Before the discovery of Cateloopillars, we had only two design pattern for constructible spaceships, namely Geminioids and Caterpilloids.

Geminoids were slow, very slow - but below the threshold speed could move in any direction and with any slow speed as we chose. The fastest Geminoid i.e. Demonoid is slower than c/5,000.

Caterpilloids are much faster than Geminoids, they can move faster than 0.37c, and before the discovery of spaghetti monster (~0.43c) were the fastest ships below c/2. Their limitation is the basic interaction they're based on. The engine so to speak of the Caterpillar, which limits it's ability to a specific speed. Thus Caterpillars were adding specific speeds to the list of spaceships, but couldn't ever cover a wide range of speeds.

The engineless ship, was an old idea, but the detailed design was elusive. Caterloopillar solved many technical problems in single elegant solution. It uses two sides, the head and the tail as universal constructor. It's based on *WSS, and uses SLs as instruction tape, converting them into slow salvo construction, that in turn build *WSS stream, of which the reading heads are built. Special search was written to find edgy slow salvo *WSS construction. Unlike other designs, it's speed is variable, and quite high (but limited by c/4). It has both the advantages of universal constructor based ships (and can vary the speed), and the engine based ships (which can move pretty fast).

Caterloopillars are not only new and unique constructible ships design pattern, they also solve quite old problem in GOL, finding spaceships with various speeds. Until Caterloopillars were discovered, the number of relatively high speed spaceships was very limited. Caterloopillars could compete with other search techniques, and were the first examples of long searched speeds like: c/8, c/9, c/11, c/12.

Caterloopillars prove that in GOL there exist a ship of any speed < c/4. The fastest caterloopillar explicitly constructed is 5c/21, and there exist a script to construct caterloopillar of many speeds.

LifeWiki has a page for Caterloopillar for more details.

User avatar
Alexey_Nigin
Posts: 323
Joined: August 4th, 2014, 12:33 pm
Location: Ann Arbor, MI
Contact:

Re: Pattern of the Year 2016 (Entries)

Post by Alexey_Nigin » January 23rd, 2017, 9:21 am

simsim314: Since there are 14 entries, and people will be supposed to read the descriptions of them all before voting, I am inclined to limit descriptions to one paragraph each. Anyway, thank you for what you provided: I guess compiling the gist of your post will be easier than writing something from scratch.
There are 10 types of people in the world: those who understand binary and those who don't.

User avatar
dvgrn
Moderator
Posts: 6046
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Pattern of the Year 2016 (Entries)

Post by dvgrn » January 23rd, 2017, 11:36 am

Compact stable Herschel splitter:

AbhpzTa joined the forums on April 13th, immediately posted an innovative p61 gun, and within another hour posted an elegant solution to an old circuitry problem: a direct Herschel signal splitter that recovers in less than a hundred ticks. People had been looking for something exactly like this for very nearly twenty years, ever since the original 1996-97 universal set of Herschel conduits. Periodic, oscillator-assisted Herschel splitters have been known almost from the beginning, and big slow stable splitters were cobbled together pretty quickly also, using multi-stage stable glider reflectors -- and a *lot* of time has been spent over the years, trying to patch up dozens of promising near misses. Somehow everyone had missed this method of extending an Fx119 conduit with a partial copy of itself, to extract a second Herschel output. We had I think a six-stage stable Herschel duplicator courtesy of Guam, and of course Kazyan’s syringe counts as a two-stage Herschel splitter, or three-stage if it has to be Spartan like AbhpzTa’s contribution. But this is the first-ever single-stage stable Spartan signal splitter, if you only count outputs of the same type as the input.

[Yes, it's a long paragraph. Please feel free to chop it shorter as necessary...]

User avatar
simsim314
Posts: 1731
Joined: February 10th, 2014, 1:27 pm

Re: Pattern of the Year 2016 (Entries)

Post by simsim314 » January 25th, 2017, 6:22 am

Alexey_Nigin wrote: I guess compiling the gist of your post will be easier than writing something from scratch.
Just take the paragraph starting with "The engineless ship". And add link to LifeWiki.

Aperiodic tiles: Wang tiles is a relatively old example of "dominoes" which on the one hand can cover the whole plane, and on the other the tiling with those dominoes must be aperiodic. Lately there were found pretty small Wang tiles (11 tiles with 4 colors). NotLiving used this finding to transform them into set of "quasi Still Life" that in order to connect and be valid Still life, have to obey the same rules as the published Wang tiles. This is the first example where aperoidic infinite SL was shown. Although the problem has no history in GOL community, the connection between tiling and SLs is obvious, and finding connection between GOL and other areas of mathematics is always intriguing and fun, especially when it's done in such elegant way.

User avatar
Scorbie
Posts: 1439
Joined: December 7th, 2013, 1:05 am

Re: Pattern of the Year 2016 (Entries)

Post by Scorbie » January 25th, 2017, 7:43 am

Statorless p5: A very notable thing of this discovery is that the oscillator is Strictly volatile. Strict volatility is a very tight restriction, with the only periods with strictly volatile oscillators being 1, 2, 3, 5, 6, 8, 13, 15, 22, 30, 33, and 177. The latest strictly volatile oscillator "seems to be" (confirmation needed) derived from 34P13 on 2009. This means that this discovery adds a new period in 7 years. This is also the second statorless oscillator of a prime period greater than 2. (The first being the 34P13 derivative.)

p7 sparker: (Actually this discovery was more or less a personal goal and is special to me :) Thanks anyways x) )
This adds a new relatively strong HW emulator to the p7 collection, which only had a pipsquirter. This can overclock the biblock hassling reaction (which only works at p15+) at p14. With the overclocked p14 I was able to complete a p28 wick by Jason Summers, which had no prior stabilization.
Best wishes to you, Scorbie

User avatar
blah
Posts: 244
Joined: April 9th, 2016, 7:22 pm

Re: Pattern of the Year 2016 (Entries)

Post by blah » January 25th, 2017, 7:55 am

In my description of the 8-bit computer I said "your only competition is a turing machine and an MRM". Which isn't true, I forgot about the spartan universal computer-constructor.
succ

Sokwe
Moderator
Posts: 1504
Joined: July 9th, 2009, 2:44 pm

Re: Pattern of the Year 2016 (Entries)

Post by Sokwe » January 25th, 2017, 4:36 pm

Spaghetti Monster: This is the first 3c/7 spaceship to be discovered. 3c/7 has been one of the most sought-after speeds in the last 15 years. Several years ago Paul Tooke ran a gfind search for a bilaterally-symmetric width-27 3c/7 spaceship. This search took nearly a year to complete and proved that no such spaceships exist. In 2016 Tim Coe used his own search program, knightt, to search for 3c/7 ships. By carefully changing the symmetry type and width mid-search, he was able to find this width-29 ship in about two months.
-Matthias Merzenich

Post Reply