15 in 15: Efficient 15-bit Synthesis Project (DONE!)

For discussion of specific patterns or specific families of patterns, both newly-discovered and well-known.
mniemiec
Posts: 1590
Joined: June 1st, 2013, 12:00 am

Re: 15 in 15: Efficient 15-bit Synthesis Project (5 SLs remain)

Post by mniemiec » November 19th, 2016, 12:10 am

Extrementhusiast wrote:Reaction for 15.849, synthesizable in no more than thirteen gliders: ...
12, actually:

Code: Select all

x = 257, y = 58, rule = B3/S23
184bo$185boo$184boo50bo$236bobo$194bo41boo$189bobboo19boo18boo$187bobo
3boo18boo18boo$188boo$$145b3o47b3o$81bo24bo108bo19bo19bo$82bo23bobo34b
o5bo43bo5bo14bobo17bobo17bobo$48bo31b3o23boobboo31bo5bo43bo5bo12bobbo
16bobbo16bobbo$46bobo61bobo30bo5bo43bo5bo11boboo16boboo16boboo$47boo
19boo8boo18boo10bo27boo48boo20bobbo16bobbo16bobbo$68boo7bobo18boo24b3o
11boo5b3o26b3o11boo5b3o12boo3bo14boo3bo14boo3bo$bbobo21bo19bo17bo14bo
14bo39bo49bo29boo18boo18boo$3boo19b3o17b3o16bobo27bobo37bobo47bobo$3bo
19bo19bo18bobbo26bobbo36bobbo46bobbo$23boo18boo18boo28boo38boo48boo$3o
$bbo$bo33$152bo$152boo$151bobo!

User avatar
Goldtiger997
Posts: 762
Joined: June 21st, 2016, 8:00 am

Re: 15 in 15: Efficient 15-bit Synthesis Project (3 SLs remain)

Post by Goldtiger997 » November 19th, 2016, 9:00 am

Can anyone use this to make 14.558 in 10 gliders?

The complex mess at the top is just a wing.

Code: Select all

x = 22, y = 38, rule = B3/S23
6b2o$5bob2o$9bo$3b2o3bobo$2bo2bobo3bo$b3obobob2o$3b2o3bo$5b3o$5bo3$12b
3o$7bo$6bobo$5bo3bo$5bo3bo$5bo3bo$6bobo4b2o$7bo4bo2bo$13b2o4$9b3o$9bob
2obo$7bo3bobobo$7bo3bob3o$7bobobo2bo$8bo$9b2ob2o$8bobob2o$7b2o2bo$8bo
2b2o$9b3o$10bo!
If so, 15.1060 can be made in 14 gliders by converting the snake-like bit into a python-like bit.

User avatar
BlinkerSpawn
Posts: 1992
Joined: November 8th, 2014, 8:48 pm
Location: Getting a snacker from R-Bee's

Re: 15 in 15: Efficient 15-bit Synthesis Project (3 SLs remain)

Post by BlinkerSpawn » November 19th, 2016, 9:39 am

Goldtiger997 wrote:Can anyone use this to make 14.558 in 10 gliders?

The complex mess at the top is just a wing.

Code: Select all

rle
If so, 15.1060 can be made in 14 gliders by converting the snake-like bit into a python-like bit.
Sure:

Code: Select all

x = 74, y = 32, rule = B3/S23
70bo$65bo3bo$66b2ob3o$32bo32b2o$9bo22bobo$10bo8bo12b2o38b2o$8b3o8bobo
49b2o$19b2o52bo$28bo$26b3o28bo2bob2o$25bo31b5obo$25b2o$59bo$58bobo$59b
o$69b2o$69bobo$69bo$2b2o20b2o$3b2o19bobo$2bo21bo$20b3o$6b3o13bo$8bo12b
o$7bo5$2o$b2o$o!
LifeWiki: Like Wikipedia but with more spaceships. [citation needed]

Image

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

Re: 15 in 15: Efficient 15-bit Synthesis Project (2 SLs remain)

Post by BobShemyakin » November 19th, 2016, 12:12 pm

15.497:

Code: Select all

x = 44, y = 30, rule = B3/S23
5$13b2o$12bo2bo$13b2o2$9bo$8bobo4bo$8bobo4bo3b2o10b2o$9bo5bo4b2o9bobo$
20bo11b2o6$19bo$18b2o$18bob2o2$22bo$21b2o$21b2o!
Bob Shemyakin

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

Re: 15 in 15: Efficient 15-bit Synthesis Project (2 SLs remain)

Post by Extrementhusiast » November 19th, 2016, 3:59 pm

Possible predecessor for 15.1034:

Code: Select all

x = 10, y = 11, rule = B3/S23
3bo$3b2o$5b2o$o6b3o$o2b2o3bo$2bo2bo$3bobo2b2o$4bo3bo$9bo$6b3o$6bo!
The upper part would have to be made in four gliders or less in order to stay under budget.

EDIT: 15.1034 from twelve gliders, using a different predecessor:

Code: Select all

x = 42, y = 19, rule = B3/S23
8bobo$8b2o12bo$9bo5bo4b2o$13b2o6b2o$5bobo6b2o$6b2o$2bo3bo9b2o16b2o$3bo
12bobo15bo2b2o$b3o8b2o2bo18bobobo$12bo23bo3bo$13bo27bo$10b3o25b3o$10bo
27bo$2o$b2o$o$5b3o$5bo$6bo!
EDIT 2: 15.497 from fourteen gliders:

Code: Select all

x = 48, y = 38, rule = B3/S23
17bobo$17b2o$18bo$4bobo$5b2o$5bo$18bo$18bobo$18b2o2$obo$b2o39b2o$bo40b
o3b2o$20b3o21bo2bo$20bo22b2obo$21bo6bo16bo$8b2o18bobo14bobo$7bobo18b2o
16b2o$9bo2$5b2o$4bobo$6bo9b2o$10b2o3bobo$11b2o4bo$10bo4$8b3o$7bo2bo$
10bo$6bo3bo$10bo$7bobo$32b3o$32bo$33bo!
[/project]
I Like My Heisenburps! (and others)

User avatar
Goldtiger997
Posts: 762
Joined: June 21st, 2016, 8:00 am

Re: 15 in 15: Efficient 15-bit Synthesis Project (2 SLs remain)

Post by Goldtiger997 » November 19th, 2016, 7:11 pm

Extrementhusiast wrote:EDIT 2: 15.497 from fourteen gliders:

Code: Select all

x = 48, y = 38, rule = B3/S23
17bobo$17b2o$18bo$4bobo$5b2o$5bo$18bo$18bobo$18b2o2$obo$b2o39b2o$bo40b
o3b2o$20b3o21bo2bo$20bo22b2obo$21bo6bo16bo$8b2o18bobo14bobo$7bobo18b2o
16b2o$9bo2$5b2o$4bobo$6bo9b2o$10b2o3bobo$11b2o4bo$10bo4$8b3o$7bo2bo$
10bo$6bo3bo$10bo$7bobo$32b3o$32bo$33bo!
...
Great Job!

Thanks to everyone that contributed; Extrementhuiast, BobShemyakin, BlinkerSpawn, mniemiec, Kayzan, Sokwe, yootaa, AbhpzTa , gmc_nxtman, Alexey_Nigin, chris_c, ...

Did I miss anyone?

If we want to replicate this challenge with pseudo-still-lifes, all that needs to be synthesised in less than one glider per bit is this 14-bit pseudo-still-life:

Code: Select all

x = 11, y = 4, rule = B3/S23
2o$obobob2o$3b2obobobo$9b2o!

User avatar
BlinkerSpawn
Posts: 1992
Joined: November 8th, 2014, 8:48 pm
Location: Getting a snacker from R-Bee's

Re: 15 in 15: Efficient 15-bit Synthesis Project (2 SLs remain)

Post by BlinkerSpawn » November 19th, 2016, 8:34 pm

Goldtiger997 wrote:
Extrementhusiast wrote:[/project]
Great Job!
And in only a month and ten days! (judging from this post of yours)
I've updated the original post to reflect this tremendous event and now I think I can finally leave it alone. :lol:
LifeWiki: Like Wikipedia but with more spaceships. [citation needed]

Image

Gamedziner
Posts: 795
Joined: May 30th, 2016, 8:47 pm
Location: Milky Way Galaxy: Planet Earth

Re: 15 in 15: Efficient 15-bit Synthesis Project (DONE!)

Post by Gamedziner » November 19th, 2016, 10:21 pm

(I'm not sure how, but I ended up triple posting. Sorry!)
Last edited by Gamedziner on November 20th, 2016, 7:14 am, edited 1 time in total.

Code: Select all

x = 81, y = 96, rule = LifeHistory
58.2A$58.2A3$59.2A17.2A$59.2A17.2A3$79.2A$79.2A2$57.A$56.A$56.3A4$27.
A$27.A.A$27.2A21$3.2A$3.2A2.2A$7.2A18$7.2A$7.2A2.2A$11.2A11$2A$2A2.2A
$4.2A18$4.2A$4.2A2.2A$8.2A!

Gamedziner
Posts: 795
Joined: May 30th, 2016, 8:47 pm
Location: Milky Way Galaxy: Planet Earth

Re: 15 in 15: Efficient 15-bit Synthesis Project (DONE!)

Post by Gamedziner » November 19th, 2016, 10:21 pm

Congratulations!


...That being said, what's next?

Code: Select all

x = 81, y = 96, rule = LifeHistory
58.2A$58.2A3$59.2A17.2A$59.2A17.2A3$79.2A$79.2A2$57.A$56.A$56.3A4$27.
A$27.A.A$27.2A21$3.2A$3.2A2.2A$7.2A18$7.2A$7.2A2.2A$11.2A11$2A$2A2.2A
$4.2A18$4.2A$4.2A2.2A$8.2A!

Gamedziner
Posts: 795
Joined: May 30th, 2016, 8:47 pm
Location: Milky Way Galaxy: Planet Earth

Re: 15 in 15: Efficient 15-bit Synthesis Project (DONE!)

Post by Gamedziner » November 19th, 2016, 10:21 pm

(Sorry, I didn't intend to double post.)
Last edited by Gamedziner on November 20th, 2016, 7:13 am, edited 1 time in total.

Code: Select all

x = 81, y = 96, rule = LifeHistory
58.2A$58.2A3$59.2A17.2A$59.2A17.2A3$79.2A$79.2A2$57.A$56.A$56.3A4$27.
A$27.A.A$27.2A21$3.2A$3.2A2.2A$7.2A18$7.2A$7.2A2.2A$11.2A11$2A$2A2.2A
$4.2A18$4.2A$4.2A2.2A$8.2A!

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

Re: 15 in 15: Efficient 15-bit Synthesis Project (2 SLs remain)

Post by mniemiec » November 19th, 2016, 10:49 pm

Goldtiger997 wrote:If we want to replicate this challenge with pseudo-still-lifes, all that needs to be synthesised in less than one glider per bit is this 14-bit pseudo-still-life: ...
Actually, my site is has a typo on it; the cis version of the above also takes 14 gliders. The synthesis shows that, even though it is listed as 12 on the stamp collection (unless there was a 12-glider synthesis somewhere that I forgot about. Does anyone remember one?)

There are two 14-bit pseudo-still-lifes up to 16 gliders that cost 1 glider per bit (see above two), plus 8 16-bit ones that cost more than 1 glider per bit: 20, 18, 18, 27, 18, 23, 17, 25 respectively:

Code: Select all

x = 144, y = 7, rule = B3/S23
oo7boo4boo13boobo11boobo11boobo11boobo12booboo9bobooboo8boo13bo$obobob
obobo4boboboboo8bob3o9boboo11boboo11boboo11bobobobbo7boobobobo8bo13b3o
$3booboo10boobobobo4bo5bo29boo11boo9bo5boo14bo8boboobobo9boboo$24boo5b
3obo10booboboo8boobobbo8boobo11bo19bo10boboboobo7bo3bo$33boboo9booboob
o8booboo10boobbo11bobo13bobo18bo7boobo$79boo12boo13boo19boo10b3o$143bo
!
On the plus side, pseudo-still-lifes are usually easier to synthesize than still-lifes, since most small pieces can be added for around 1 glider/bit or less, and the synthesis of the (now smaller) larger pieces has already been solved. On the minus side, apgsearch decomposes all pseudo-still-lifes into their component pieces, so soup searches will be of no use in finding cheap predecessors, unless one searches all soups that produce one of the component objects in the remote hope that it's formed as part of a pseudo-object.

On a related note, all pseudo-still-lifes up to 19 bits have been solved. All 20-bit ones can be synthesized except for the following 57 (the last 3 of which can be synthesized based on some of the others plus a snake-to-carrier converter);

Code: Select all

x = 144, y = 83, rule = B3/S23
ooboobo8booboobo8booboobo8booboboo8booboobo8booboobo8booboobboo6booboo
boo8booboo10booboo$obooboo8bobooboo8booboboo8bobooboo8bobooboo8boboob
oo9bobobbobo7bobo3bo7bobobo10bobobobo$90bo3bobo8bo3bobo8bo4b3o7bo6bo$o
booboo8bobooboo8booboboo9booboobo8booboobo8booboobo7booboobbo7booboobo
bo7bo5bo8bo3b3o$ooboboo8booboobo8bobooboo9booboboo8bobooboo8booboboo
13boo14boo8bobobo10bobo$121booboo10booboo10$oo13booboo10booboo11booboo
9booboobo8boo3boo9boobo11booboo10booboo10booboo$obbooboo8bobobboo7bob
oobo9bobobobo9boboboo8bo5bo8bobboo12bobobo10bobobo8bobobo$bobo3bo7bo3b
obo14bo8bo6bo8bo15bobo10boo14bo5bo8bo4bo8bo4bo$oobbobo8booboobbo14bo8b
o6bo8bo13booboo15boo7bo4boo8bo5boo8bo4bo$bboboboo12bobo7boobobobo9bobo
bobo9boboboo7bo3bo9boo5bo7bobobbo9bobo4bo9bo4bo$bboo16boo8bobooboo11b
ooboo9booboobo8bobo10bobob3o9boo3bo9boobbo12bobobo$76booboo12boo15boo
13boo10booboo9$obooboo8boo13boo13boo13boobbo10boobboo9boobboo9boobo11b
oobobboo7booboo$oobobo10bo14bobbo10bo4boo8bo3b3o8bo3bo10bobobbo9boboo
11boboo3bo7boobo$7bo7bo5boo7bo3b3o10bobbobo8bo5bo8bo3bo11bobo16boo11b
3o11bobboo$6boo7b4obobo7boo5bo8boo4bo7boo4bo8boo4bo9bobb4o7boboo3bo8b
oobo13boobbo$bboobo12bobo10bo4bo8bo5bo8bo4bo9bo6bo8boo4bo7boobobbo10bo
16bobo$bobboo11bo3bo9bobobo10bobobo11bobo12bob3o14bo11bobo11bobo13bobb
oo$boo14booboo10booboo10booboo9booboo10boobo16boo10boo13boo13boo9$bboo
boo9boboo11boobbo10booboo10booboo9boboobo9boboobo9bobooboo8bobooboo8b
oobboo$3bobo10boobo10bobobb3o9bobobo8bobobobo8boobob3o7boobob3o7boobob
o9boobobobo7bobbobo$bbo3bo15boo6bo7bo7bo5bo7bo6bo15bo14bo12bo16bo8boo$
bo5bo7boboo4bo7bo5bo7bo7bo7bo6bo13boo10booboo13bo13boo10bobo$o7bo6boo
bbobbo9bo3bo8bobo3bobo8bo3bobo9boobbo11bo12boo4bo8boobbo10bobb3o$b3ob
3o10bobbo11bobo10boobbobo10bobboo10bobobo9bobo12bobb3o9bobobo10boo4bo$
3bobo13boo11booboo14bo10boo17bo10boo15boo14bo16boo9$ooboo10booboo10boo
boo12booboo9booboo10booboo10booboo9boo13boo13boo$bobo12boboo10bo3bo13b
obobo9bobo12bobo12bobobbo8bo5bo8bo4boo7bobo3boo$o3bo10bo6boo7bobo3boo
8bo5bo7bo3bo9bo4bo10bo4boo7bo4b3o8boboobbo9bo3bo$ooboobbo7booboo3bo6b
ooboo3bo7bo7bo5bo5bo8boo4bo8bo14boobbo12boboo11boo3bo$4bobobo10bobbo
11bobbo7bo5b3o6bobo4bo8bo5bo8b3o16bo30boo$4bobbo11bobo12bobo9b3obbo9bo
bobboo8bobobboo10b3o9boobboo11boboo11boobbo$3boo15bo14bo12bo14bo13bobo
16bo8bobbo13boobbo10bobobo$63boo13bo16boo10boo16boo13bo8$ooboo11booboo
9bo14boo15boobo11boobo10boo$o3bo12bobo10b3o4boo6bobo12bobbob3o7bobbob
3o9bo4bo$bobo3bo8bo3b3o10bo4bo8bo3boo7boo6bo6boo6bo6bo4b3o$oob5o7bo7bo
8bobb3o9boobbo15boo10booboo6boobbo$15boo5bo9boobo17bo9boobbo11bo15bo$
5bo15bo11bo13boo3boo9bobobo9bobo10boobboo$4bobo11bobo12bobo11bobobo14b
o10boo11bobbo$5bo12boo14boo14boo40boo!

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

Re: 15 in 15: Efficient 15-bit Synthesis Project (DONE!)

Post by Kazyan » November 19th, 2016, 11:18 pm

One of those unsynthesized pseudo still lifes was solved during the dart synthesis.
Tanner Jacobi
Coldlander, a novel, available in paperback and as an ebook. Now on Amazon.

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

Re: 15 in 15: Efficient 15-bit Synthesis Project (DONE!)

Post by mniemiec » November 20th, 2016, 12:23 am

Kayzan wrote:One of those unsynthesized pseudo still lifes was solved during the dart synthesis.
Thanks. I wasn't aware of that. I was in the hospital at that time, and their wifi login script was broken, so I didn't have internet access. I didn't look at the dart synthesis until much later, when it had a much simpler synthesis that no longer needed that pseudo-still-life.

EDIT: The initial pseudo-still-life (two blocks on two snakes) can be reduced by 4 from 17 to 13, by using the improved 5-glider up-snake-below-snake synthesis (I'm not sure where this came from), plus a new 4-glider block-hoisting converter. The final pseudo-still-life (#2) can be reduced by 6 more from 45 to 35 by using a pre-block rather than a house and a cheaper fleet. Luckily, the cleanup glider can be moved with no extra cost. (Sadly, these optimizations can only be done on one side, although there might be other ways to do one and/or the other on the other side as well).

Code: Select all

x = 195, y = 132, rule = B3/S23
130bo$128bobo49bobo$129boo49boo$181bo$$122bo$123bo$121b3o22$94bo$93bo$
93b3o$$83bo17boobo46boobo33booboobo$81boo18boboo46boboo33booboboo$82b
oo$101boboo46boboo33booboboo$84boo15boobo46boobo33booboobo$70boo12bobo
$71boo11bo$70bo3$88boo$87boo$89bo68bo$157bo$157b3o$155bo$126b3o25boo$
128bo25bobo$127bo7$166boo$165boo$167bo3$121b3o$123bo$122bo19$126bo$93b
oo20bo9bo$92bobobboo15bo10b3o$94bobbobo14b3o$97bo6$180bo$179bo$119bo
59b3o$118bo13boo22boo18boo$118b3o3bo6b4o20bobbo16bobbo$123bo6booboo21b
oo18boo$123b3o5boo3$8booboobo13booboobo13booboobo13booboobo23booboobo
43booboobo13booboobo13booboobo$8booboboo13booboboo13booboboo13booboboo
23booboboo43bobooboo13bobooboo13bobooboo$$8booboboo13booboboo13boobob
oo13booboboo12bo10booboboo43bobooboo13bobooboo13bobooboo$8booboobo13b
ooboobo13booboobo13booboobo12boo9booboobo43booboobo13booboobo13booboob
o$86bobo$$27boo18boo18boo28boo$26bobo10bo6bobo17bobo27bobo$9boo16bo12b
o6bo19bo29bo$9bobo26b3o23bo29bo$9bo53bobo27bobo$5boo34boobboo17bo29bo$
4boo34boboboo$boo3bo35bo3bo$obo117bo$bbo116boo$119bobo3$114b3o$114bo$
115bo$$104bo$103boo4bo$103bobobboo6b3o$108bobo5bo$117bo!
EDIT: The same method also gives #6 from 21 gliders:

Code: Select all

x = 160, y = 49, rule=B3/S23
140bo$138boo$139boo5$107boo20bo$106bobobboo15bo$108bobbobo14b3o$111bo
$$133bobo$133boo$134bo5$10bobo$10boo$bo9bo$bbobbo$3o3boobbo$5boobbo46b
o3bo$9b3o13boobo25boboboo5boobo13booboobo23booboobo33booboobo$25boboo
26boobboo4boboo13booboboo23booboboo33bobooboo$$26boobo36boobo16boobo
11bo14boobo33booboobo$3b3o20boboo36boboo16boboo11boo13boboo33booboboo$
5bobboo90bobo$4bobboo3b3o$9bobbo124b3o$3bo9bo123bo$3boo127b3o3bo$bbobo
127bo$133bo4$38b3o$40bo$39bo59boo$100boo$99bo$$127boo$126boo$128bo!

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

Re: 15 in 15: Efficient 15-bit Synthesis Project (DONE!)

Post by BobShemyakin » November 20th, 2016, 9:35 am

Congratulations!

Quote database of 15 in 15:
still15.rar
15 in 15
(254.25 KiB) Downloaded 1511 times
Gamedziner wrote: ...That being said, what's next?
How about 16 in 16?

Bob Shemyakin

User avatar
BlinkerSpawn
Posts: 1992
Joined: November 8th, 2014, 8:48 pm
Location: Getting a snacker from R-Bee's

Re: 15 in 15: Efficient 15-bit Synthesis Project (DONE!)

Post by BlinkerSpawn » November 20th, 2016, 9:56 am

BobShemyakin wrote:
Gamedziner wrote: ...That being said, what's next?
How about 16 in 16?

Bob Shemyakin
*sigh* Am I going to have to make another thread?
LifeWiki: Like Wikipedia but with more spaceships. [citation needed]

Image

User avatar
dvgrn
Moderator
Posts: 10565
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: 15 in 15: Efficient 15-bit Synthesis Project (DONE!)

Post by dvgrn » November 20th, 2016, 10:20 am

BlinkerSpawn wrote:
BobShemyakin wrote:How about 16 in 16?
*sigh* Am I going to have to make another thread?
Well, it's traditional to leave a decent interval between these projects for rest and recuperation -- or at least to allow Mark to get his database caught up with all the new discoveries.

That way there's some hope of getting a latest and greatest version of the database hosted on conwaylife.com, instead of Tom Rokicki's temporary "codercontest" website...!

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

Re: 15 in 15: Efficient 15-bit Synthesis Project (DONE!)

Post by mniemiec » November 20th, 2016, 2:29 pm

BobShemyakin wrote:How about 16 in 16?
The main reason I didn't complete the 15-bit still-lifes myself in the early 00s wasn't the difficulty of the task (I got about half of them done, and identified the problematic ones), but because of the sheer tedium of manually instantiating 1353 separate syntheses. There are 3286 16-bit still-lifes, more than any other collection of objects yet attempted for synthesis. I have syntheses for 352 of them (which would probably be good candidates for reduction), but that leaves an additional 2934 trivial syntheses that would have to be manually instantiated and vetted for cost. That is a task probably better suited to an automatic process.

User avatar
BlinkerSpawn
Posts: 1992
Joined: November 8th, 2014, 8:48 pm
Location: Getting a snacker from R-Bee's

Re: 15 in 15: Efficient 15-bit Synthesis Project (DONE!)

Post by BlinkerSpawn » November 20th, 2016, 3:58 pm

mniemiec wrote:...That is a task probably better suited to an automatic process.
Maybe have some place where fully-exploded (i.e. gliders only, no intermediate steps) syntheses can be uploaded, then some web-based script could verify the glider count (simply read the total on-cells and divide by 5), run the synthesis to stabilization, then look up the resulting object and check whether it's a record?
That looks like a lot but there are far messier options available. :?
LifeWiki: Like Wikipedia but with more spaceships. [citation needed]

Image

chris_c
Posts: 966
Joined: June 28th, 2014, 7:15 am

Re: 15 in 15: Efficient 15-bit Synthesis Project (DONE!)

Post by chris_c » November 20th, 2016, 8:23 pm

I have been working on my glider synthesis analyser code that I have alluded to a couple of times on the forum. As of a few days ago I had computer-based verification that all 14-bit still lifes could be made in less than 14 gliders.

This evening I turned my code loose on Bob's still15.rar from a few posts back. Sadly there are some errors in the collection and I have not been able to fix them by looking at alternative sources (mniemiec's website or patterns posted on these forums).

In short, I do not know how to synthesise the following still lifes:

Code: Select all

15.594    15.784    15.1021    15.1150    15.1158    15.1181    15.1223
First, the synthesis of 15.594 is actually a synthesis of 15.954 (this problem occurs both in Bob's collection and on mniemiec's site).

Second, the synthesis of 15.784 in Bob's collection is actually of 15.794. I couldn't find a valid synthesis of 15.784 anywhere.

The last 5 still lifes all contain an invalid synthesis in the last step in Bob's collection. I could not find working versions of these on mniemiec's website or in the forums.

Apart from these I do not have verified syntheses for roughly 10 other still lifes. I believe that I can fix all of these myself (for example, my code does not recognise *WSS).

Can anyone help with the ones I am struggling with?

Hopefully I can get all my code tidied up and then publish something useful soon.

User avatar
BlinkerSpawn
Posts: 1992
Joined: November 8th, 2014, 8:48 pm
Location: Getting a snacker from R-Bee's

Re: 15 in 15: Efficient 15-bit Synthesis Project (DONE!)

Post by BlinkerSpawn » November 20th, 2016, 9:48 pm

15.784:

Code: Select all

x = 100, y = 19, rule = B3/S23
83bobo$84b2o$84bo$92bo$11bo80bobo$bo8bo70bo10b2o$2b2o6b3o16bo19bo19bo
6bo2bobo17bo$b2o24b3o17b3o17b3o6b2o2b2o15b3o$26bo11bo7bo19bo8bobo18bo$
6b2o3b2o14bo8bobo8bo19bo29bo$6bobob2o12b3o10b2o5b3o14b2ob3o24b2ob3o$6b
o5bo11bo15b2o2bo16b2obo19b2o5b2obo$bobo35bobo43b2o$2b2o37bo42bo$2bo2$
3o$2bo$bo!
LifeWiki: Like Wikipedia but with more spaceships. [citation needed]

Image

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

Re: 15 in 15: Efficient 15-bit Synthesis Project (DONE!)

Post by Extrementhusiast » November 20th, 2016, 10:32 pm

Corrected final steps for the final five:

Code: Select all

x = 36, y = 169, rule = B3/S23
4bobo$5b2o$5bo3$10bo$9bobo$8bobo21bob2o$8b2o22b2obo$6b2o10bobo9b2o$7bo
4bobo3b2o11bo$6bo5b2o5bo10bo$5bo7bo15bo$5bobo21bobo$6b2o5b2o15b2o$14b
2o$13bo21$9bo$8bo$3bo4b3o$bobo$2b2o3$10b2o18b2o$10bobo16bo2bo$13bo15b
2o2bo$obo11bo19bo$b2o3bobo4b2o18b2o$bo5b2o4bo19bo$7bo6bo19bo$15bo19bo$
6b2o6b2o18b2o$5b2o$7bo25$5b2ob2o15b2ob2o$5bob2obo14bob2obo$11bo19bo$
12bo19bo$3bo7b2o20bo$bobo26b3o$2b2o26bo$5b2o$6b2o$5bo7b3o$13bo$14bo2$
11b2o$12b2o$11bo4$4b2o$5b2o$4bo21$11bo$9b2o$6bo3b2o$4bobo$b2o2b2o$obo
9bo$2bo9b3o18b2o$15bo16bo2bo$9b2o3b2o15bo2b2o$10bo19bo$5b2o2bo15b2o2bo
$5bo2bo16bo2bo$6b2o18b2o27$4b2o18b2o$4bobo17bobo$6bo6bo12bo$6b2o4bo13b
2o$7bo4b3o12bo2bo$7bobo17bobobo$8bobo4b2o11bobo$9bo5bobo11bo$15bo!
Although I had to use an extra glider for the first three of those, everything still comes in under budget.

Also, 15.594 is trivial to make from 14.471:

Code: Select all

x = 30, y = 13, rule = B3/S23
7bo$5bobo$6b2o$9bo$9bobo$9b2o$2o20b2o$bo21bo4b2o$bobo2b2o15bobo2bo$2b
2o3bo16b2o3bo$4b3o6b2o11b3o$4bo7b2o12bo$14bo!
I Like My Heisenburps! (and others)

chris_c
Posts: 966
Joined: June 28th, 2014, 7:15 am

Re: 15 in 15: Efficient 15-bit Synthesis Project (DONE!)

Post by chris_c » November 21st, 2016, 6:19 am

Thanks guys. I should be able to finish off the 15 bit still lifes without too much more fiddling around. As a bit of a sneak preview here is a 3 glider tail adder that I found within the last week. It sets quite a few new records, for example I think 11.3 (a.k.a. down long boat with tail) goes from 7 gliders to 6:

Code: Select all

x = 22, y = 22, rule = B3/S23
19bo$19bobo$19b2o10$bo$o$3o2$14b2o$b3o9bobo$bo10bobo$2bo10bo!

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

Re: 15 in 15: Efficient 15-bit Synthesis Project (DONE!)

Post by mniemiec » November 21st, 2016, 1:14 pm

chris_c wrote:Thanks guys. I should be able to finish off the 15 bit still lifes without too much more fiddling around. As a bit of a sneak preview here is a 3 glider tail adder that I found within the last week. It sets quite a few new records, for example I think 11.3 (a.k.a. down long boat with tail) goes from 7 gliders to 6: ...
Very nice! While it's a little bit bulkier than some of the previous ones, it's also more general than the previous 3-glider tail-adder (that only works on boats, barges, etc. facing away from the tail). This should improve a mind-bogglingly large number of syntheses. (Oh well, I had been planning on re-examining all 11000+ of them for recent improvement anyway...)

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

Re: 15 in 15: Efficient 15-bit Synthesis Project (DONE!)

Post by BobShemyakin » November 21st, 2016, 1:42 pm

chris_c wrote:...
This evening I turned my code loose on Bob's still15.rar from a few posts back. Sadly there are some errors in the collection...
Thanks! When processing such a large amount of information manually by one person mistakes are inevitable. So good that her check and others.
I corrected these mistakes here:
still15.rar
15 in 15
(254.45 KiB) Downloaded 1451 times
Bob Shemyakin

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

Re: 15 in 15: Efficient 15-bit Synthesis Project (DONE!)

Post by Extrementhusiast » November 21st, 2016, 11:43 pm

mniemiec wrote:(Oh well, I had been planning on re-examining all 11000+ of them for recent improvement anyway...)
Is this a good time to drop another large improvement?

Code: Select all

x = 17, y = 16, rule = B3/S23
o7bo$b2o5bobo$2o6b2o2$7bo$5bobo$6b2o3$13bo2bo$13b4o2$13b2o$14bo$13bo$
13b2o!
I Like My Heisenburps! (and others)

Post Reply