Searching for tiny oblique ships

For discussion of specific patterns or specific families of patterns, both newly-discovered and well-known.
Post Reply
Gamedziner
Posts: 795
Joined: May 30th, 2016, 8:47 pm
Location: Milky Way Galaxy: Planet Earth

Searching for tiny oblique ships

Post by Gamedziner » September 15th, 2016, 7:02 am

Enough spaceships in other rules are oblique that I feel it is now time for us to make a combined effort to search for a tiny oblique ship in Life. We already know that Almost-Knightship exists, so perhaps one is right around the corner.

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!

User avatar
muzik
Posts: 5614
Joined: January 28th, 2016, 2:47 pm
Location: Scotland

Re: Searching for tiny oblique ships

Post by muzik » September 15th, 2016, 8:40 am

Slap the almost knightship in a search and hope a microscopic "tagalong"-type thing stabilises it.

There's also a slope 3 partial somewhere

EDIT:
wildmyron wrote:
muzik wrote:Looking though the forum, I haven't seen any partials for slope 3 knightships...
Here's my best attempt at a (3,1)c/8 knightship (from a width 11 search in gfind):

Code: Select all

x = 9, y = 13, rule = B3/S23
7b2o$4b2ob2o$2o4b2o$3o2bo$2b6o$2bo$b4ob2o$bo5bo$5bo$3b2o3bo$4bo2b2o$4b
3o$5bo!

What I would like to know: upon discovery of the first elementary oblique spaceship, would the forums blow up just like how the Copperhead made it? Gemini was the first ever oblique spaceship and even it didn't bring as much popularity

User avatar
gmc_nxtman
Posts: 1150
Joined: May 26th, 2015, 7:20 pm

Re: Searching for tiny oblique ships

Post by gmc_nxtman » September 15th, 2016, 9:25 am

muzik wrote:What I would like to know: upon discovery of the first elementary oblique spaceship, would the forums blow up just like how the Copperhead made it? Gemini was the first ever oblique spaceship and even it didn't bring as much popularity
One of the main reasons that the copperhead blew up so much was as a result of The Register tweeting about it. I doubt if that will happen if a small oblique spaceship is discovered.

User avatar
muzik
Posts: 5614
Joined: January 28th, 2016, 2:47 pm
Location: Scotland

Re: Searching for tiny oblique ships

Post by muzik » September 15th, 2016, 3:49 pm

gmc_nxtman wrote:
muzik wrote:What I would like to know: upon discovery of the first elementary oblique spaceship, would the forums blow up just like how the Copperhead made it? Gemini was the first ever oblique spaceship and even it didn't bring as much popularity
One of the main reasons that the copperhead blew up so much was as a result of The Register tweeting about it. I doubt if that will happen if a small oblique spaceship is discovered.
I'd be very surprised if a microscopic knight didn't blow this place up even more.

User avatar
biggiemac
Posts: 515
Joined: September 17th, 2014, 12:21 am
Location: California, USA

Re: Searching for tiny oblique ships

Post by biggiemac » September 15th, 2016, 4:06 pm

Gemini brought the previous popularity record, then copperhead the new popularity record. The issue with Gemini is that it wasn't really digestible to outside readers. Copperhead was visibly "simple" and so it was probably really interesting to outside readers that it had gone undiscovered for 45 years. A new tiny knight would have to be equally simple to attract as much attention, I think. The almost-knightship is already large enough that I doubt any completion to it could generate the same buzz as the copperhead, even if many community members daydream about that sort of thing.

I don't think the hit count should be the metric of awesomeness, I guess, since virality is fickle. Let's just celebrate when such a thing is finally uncovered :)
Physics: sophistication from simplicity.

User avatar
muzik
Posts: 5614
Joined: January 28th, 2016, 2:47 pm
Location: Scotland

Re: Searching for tiny oblique ships

Post by muzik » September 15th, 2016, 5:06 pm

I'd say that a smallish oblique ship should be easy to understand, even if clarified with a simple sentence such as "The term knightship means that this spaceship moves similarly to a knight in chess".


Does anyone have any partials for any slopes other than 2 and 3?

Sokwe
Moderator
Posts: 2645
Joined: July 9th, 2009, 2:44 pm

Re: Searching for tiny oblique ships

Post by Sokwe » September 15th, 2016, 7:16 pm

We should first consider what is currently known. Most of what we know is in the table of known spaceships by width. Note that this table is for "long" spaceships. For oblique ships, the long direction is the same as the fastest direction of travel. As an example, Tim Coe ran the width-15 search for a p6 knightship, and found this partial:

Code: Select all

x = 15, y = 37, rule = B3/S23
7bo$5b5o$5bo4b2o$3bo3b3obo$3bo3b2obo$2bo3bo$2b2obo$2b4o2bo$4bo$2b4o3b
2o$2b3o2bob2o$6b2obobo$6b3o2b2o$b3o6b3o$2b2o5b2o2$7b3o2b2o$9bob3o$4b3o
3b3o$4bo6bo$3bo3b3ob2o$b2obo5b2obo$4bob3obo$o5bo2bo$4bob2o4bo$2bobo5b
2o$2bo5b2o$2bo$2bo3bobo$o3bo2b2o$o2b2o2b2o$3b3o$2ob2ob2o$2obo4b4o$bobo
b2o3b2o$2bo4bo5b2o$2b2o4bob2o!
Keep in mind that there are still a lot of ways for a single computer to potentially find an oblique ship. For example, I think it would be possible to complete width-12 searches for (2,1)c/7 and (3,1)c/8 ships in a reasonable amount of time (a few weeks if no ships exist and possibly longer if ships do exist) using knight2 (note that knight2 is incapable of properly searching for (2,1)c/8 ships).

For the most part, oblique ship searches have been limited to these "long" searches. This means that there is a lot of untapped ground in terms of search shapes for oblique spaceships. I did a bit of searching for short p6 kinghtships and for diagonally-shaped p6 knightships, but I never got very far. These could be potentially fruitful directions of research.

The p6 knightship is the most searched-for oblique ship, which means that very little searching has been done for higher-period ships. This may mean that a small p7 or p8 oblique ship exists and just hasn't been searched for yet.

Another question that we need to consider is, how do we make a "combined effort"? There isn't much in the way of distributed searches for spaceships. The closest thing to a distributed search program is probably the modification to knight2 described here. We could potentially distribute a knight2 search in the following manner:
  1. Run a search using the -t command.
  2. Kill the search after a while, and upload the output to the forums in a new thread.
  3. Let people claim portions of the search to run, much like in the hunt for a tiny diagonal spaceship.
  4. Run portions of the search according to Tim Coe's instructions.
Before starting such a search, it would probably be a good idea to get some advice from Tim Coe on how to choose the parameters.

Another option for distributing the search would be for different people to run different starting rows with WLS or JLS. This has the advantage of allowing searches with different shapes.
muzik wrote:Does anyone have any partials for any slopes other than 2 and 3?
Such a ship would have a period of at least 10. This is a very high period and is probably beyond our current search capabilities.
-Matthias Merzenich

User avatar
muzik
Posts: 5614
Joined: January 28th, 2016, 2:47 pm
Location: Scotland

Re: Searching for tiny oblique ships

Post by muzik » September 16th, 2016, 3:36 am

I see the front end of almost knightship in the partial above...

User avatar
Rhombic
Posts: 1072
Joined: June 1st, 2013, 5:41 pm

Re: Searching for tiny oblique ships

Post by Rhombic » September 16th, 2016, 12:00 pm

Can zfind be modified to allow for oblique ships?
SoL : FreeElectronics : DeadlyEnemies : 6a-ite : Rule X3VI
what is “sesame oil”?

Sokwe
Moderator
Posts: 2645
Joined: July 9th, 2009, 2:44 pm

Re: Searching for tiny oblique ships

Post by Sokwe » September 16th, 2016, 9:47 pm

Rhombic wrote:Can zfind be modified to allow for oblique ships?
It potentially could be, but zfind is limited to searches of width 10 or less. This is already insufficient to find a p6 or p7 knightship, and is probably insufficient for p8 ships as well.
-Matthias Merzenich

User avatar
praosylen
Posts: 2443
Joined: September 13th, 2014, 5:36 pm
Location: Pembina University, Home of the Gliders
Contact:

Re: Searching for tiny oblique ships

Post by praosylen » September 16th, 2016, 10:24 pm

Sokwe wrote:zfind is limited to searches of width 10 or less
Is it possible to raise this (albeit with lower performance) by splitting rows into two or more parts, or are the technical difficulties or performance losses too much to justify this?
former username: A for Awesome
praosylen#5847 (Discord)

The only decision I made was made
of flowers, to jump universes to one of springtime in
a land of former winter, where no invisible walls stood,
or could stand for more than a few hours at most...

Sokwe
Moderator
Posts: 2645
Joined: July 9th, 2009, 2:44 pm

Re: Searching for tiny oblique ships

Post by Sokwe » September 17th, 2016, 6:17 am

A for awesome wrote:
Sokwe wrote:zfind is limited to searches of width 10 or less
Is it possible to raise this (albeit with lower performance) by splitting rows into two or more parts, or are the technical difficulties or performance losses too much to justify this?
I have considered this idea, but it's quite a bit more complicated to implement than the simple things I've been doing. It would be hard for me to judge whether it would be faster than gfind without actually writing it. I've never profiled gfind or zfind to see what takes the most time.
-Matthias Merzenich

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

Re: Searching for tiny oblique ships

Post by Gamedziner » September 17th, 2016, 7:16 am

So, what reactions are known that move objects obliquely?

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!

Sokwe
Moderator
Posts: 2645
Joined: July 9th, 2009, 2:44 pm

Re: Searching for tiny oblique ships

Post by Sokwe » September 17th, 2016, 8:02 am

Gamedziner wrote:So, what reactions are known that move objects obliquely?
The half-bakery reaction is by far the best such reaction, but the ships constructed from it are large. I think reactions that "move objects obliquely" are unlikely to help with the construction of "tiny" ships. Of course, I would welcome evidence to the contrary.
-Matthias Merzenich

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

Re: Searching for tiny oblique ships

Post by Gamedziner » September 17th, 2016, 8:15 am

Is there a way to search for stabilizations of things like Knightwave?

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!

Sokwe
Moderator
Posts: 2645
Joined: July 9th, 2009, 2:44 pm

Re: Searching for tiny oblique ships

Post by Sokwe » September 17th, 2016, 9:10 am

Gamedziner wrote:Is there a way to search for stabilizations of things like Knightwave?
Sure there is: just use WLS or JLS. Of course, actually finding something is much more difficult. With that particular wave, you would need to find stabilizations for both ends, but there's no reason to think that stabilizing even one end will take less effort than finding a ship directly. Waveguides are often difficult to find. Consider the following p5 guides for simple waves:

Code: Select all

x = 333, y = 59, rule = B3/S23
203bo102bo$203b2o99b2ob2o5bo$201bo2bo96b2obo2b2o4b3o$200bo98b2o5b2ob2o
5bo$197b2o2bo94b2ob2o6b2obo3b3o$196bo2bo94b2o5b2o7bo2b2o$197bo93b2ob2o
11b3o13bo$193bo3b2o90b2o5b2o8b2o5b5o4b3o$20b2o170bob2o90b2ob2o14bo4bo
3b4ob2o3bo$20b2o169b2o2bo88b2o5b2o12bo4bo7bo4b5o$19bo2bo168b2ob2o85b2o
b2o19b3o8b2o2bo2b4ob2o$16b2obo2bo168bo2bo2b2o80b2o5b2o18b2o13b3o2bo$
22bo167b2o6b2o76b2ob2o39bo4bo$14b2o3bo2bo168bob2obob2o74b2o5b2o22b2o
11b6o$14b2o5bo168bob3obo74b2ob2o41bo6b3obo$15bob5o8b2o4bo153bobobobo
72b2o5b2o38b2o3bo$16bo7b2o3bo3bo3bo5b2ob3o3b2o215b2o46bo3bo4bobo$24b2o
2bo3b3o3bo4bo5bo2b2o3bo3bo126bo3bo78b2o44bo4bo4b3o$24b5obobo5bob2obo4b
2o2bob3obo2b2o3b2o119b2o128b6o6bo$13b3o10bo4b3o3b4o12b3obo5bobo3b3o3b
2o108b3o2bo126bo4bo$13bo16b4o3bo20bo3b2o2bo5bo2b2o3bo3bo97b2ob2o132b2o
bo$11b2o13bo3b2o2b2o31bo3b2o2bob3obo2b2o3b2o92bo136b2o$5b2o4bo14b2ob2o
37bo7b3obo5bobo4b2o89bobo132b3o6bo$5b3o3bo16bobo50bo3b2o2bo3bob2o86bo
138b3o2b3o$3bo4bo20bo60bo2bo3b3o83bo2bo130b2o6b2o2b2o$3bo3bo83bo6b2o
82bo133b3ob3o3bo2bo$7bo177bo130b3o2bo7bo$2b2obobo84b3o87b4o131b2o2b2o
6bo$2o5bo84b3o85bob2obo141b2o$2o4b2o85bo91bo27bo3b3o5b3o3bo89b2o$2b4o
174bo29b2ob5ob2o3b2ob5ob2o85b3o$133b2o46bo2bo23bob2obo5bobobobo5bob2ob
o82b2o3b2o$91b3o41b2ob2o43bo23bo3bobo3b5ob5o3bobo3bo80bo4b2obo$91bo43b
2o3b2ob2o28bo4b3o20b3o7b3o5b2o3b2o5b3o83bo2b2ob4o$94bo38b2o5b2o3b2ob2o
22bob3o4bo22bo3bo2bob3o13b3obo2bo80bo4b3o$94bo43b2o5b2o3b2ob2o17bo2bo
4bo20bo3bo4bo23bo81b2ob4obobo$92bo50b2o5b2o3b2ob2o15bo2bo16b2obo3bob4o
108b2o8bo$92b3o53b2o5b2o3b2ob2o9b2ob3o3b2o2b2ob3obo4bo5bo2bo108b3ob2ob
obo$153b2o5b2o3b2ob2o6bob2o2bobo2b3ob2obobo3bo6bo111bo2bo3bo$158b2o5b
2o3b2obo3bo4bobobo2b2o2bo2bo3b4o117b2obob2o$163b2o5b2o2bobob2o2bo2bo
136bo3b2o$90b3o75b2o11bo4bo10b2o3b2o118b2obo$90bo96bo10bo126bo$88b2o
94bo$82b2o4bo95b2o140b2o$82b3o3bo238b2o$80bo4bo239bo$80bo3bo236b3ob6o$
84bo235bo4bo5b2o$79b2obobo235bobob2o4b3o$77b2o5bo237bob3obo2b2o$77b2o
4b2o241b3obo$79b4o239bo2b3ob2o$323bobo2b2obo2$325b2o$323b5o$323bo2b2o$
324bo!
They are all large and have complicated shapes, and the c/5 diagonal and c/5 orthogonal guides already need known spaceships just to work.

But don't let this discourage you from trying. It might be that small stabilizations exist for that particular wave. I just wouldn't bet on this as the path to the first small/low-period obliquely travelling ship.
-Matthias Merzenich

User avatar
muzik
Posts: 5614
Joined: January 28th, 2016, 2:47 pm
Location: Scotland

Re: Searching for tiny oblique ships

Post by muzik » September 17th, 2016, 2:47 pm

I've tried meshing tigether almost knightship with knightwave, and haven't found any combinations.


Why not try to find a backend for it?

Sokwe
Moderator
Posts: 2645
Joined: July 9th, 2009, 2:44 pm

Re: Searching for tiny oblique ships

Post by Sokwe » September 18th, 2016, 6:59 am

Rhombic wrote:Can zfind be modified to allow for oblique ships?
I have modified zfind to search for spaceships with a horizontal offset of 1 (see here). Unfortunately, one phase of the search pattern must have width 1 less than the total width.

Edit: Here is the longest partial from "zfind p9 k2 w8 a x1 n0":

Code: Select all

..o.....
.o.o....
.o.o....
.oo...o.
......o.
..o.....
..oo....
..ooo...
...ooo..
....o...
......o.
...o..o.
..o..oo.
..o.....
..o..o..
.......o
...oo.oo
..o.....
..o.....
.oo.....
-Matthias Merzenich

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

Re: Searching for tiny oblique ships

Post by Gamedziner » September 18th, 2016, 12:44 pm

Searched for a P6 knightship using WinLifeSearch and came up with no solutions within a 7 by 7 bounding box.
(I got permission to download WinLifeSearch, yay!)

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!

User avatar
muzik
Posts: 5614
Joined: January 28th, 2016, 2:47 pm
Location: Scotland

Re: Searching for tiny oblique ships

Post by muzik » December 12th, 2016, 8:58 am

Note how this:
Sokwe wrote:

Code: Select all

x = 15, y = 37, rule = B3/S23
7bo$5b5o$5bo4b2o$3bo3b3obo$3bo3b2obo$2bo3bo$2b2obo$2b4o2bo$4bo$2b4o3b
2o$2b3o2bob2o$6b2obobo$6b3o2b2o$b3o6b3o$2b2o5b2o2$7b3o2b2o$9bob3o$4b3o
3b3o$4bo6bo$3bo3b3ob2o$b2obo5b2obo$4bob3obo$o5bo2bo$4bob2o4bo$2bobo5b
2o$2bo5b2o$2bo$2bo3bobo$o3bo2b2o$o2b2o2b2o$3b3o$2ob2ob2o$2obo4b4o$bobo
b2o3b2o$2bo4bo5b2o$2b2o4bob2o!
and this:

Code: Select all

4$16bo$11b2o2bo$11bob2o$10bo5b2o$6bobo2bo$5bo$5bo3bo2b3ob2o$5bo3bo4bo$ 10bob2o$6bo3bob2o$5b2ob3o$9b2obo$5b2o2b3o$5b2obo$8b2o4b2o$6b3o$7bo2b3o b2o$8b2o2b2o$9b3o!

actually share a common front end.

User avatar
Mr. Missed Her
Posts: 90
Joined: December 7th, 2016, 12:27 pm
Location: Somewhere within [time in years since this was entered] light-years of you.

Re: Searching for tiny oblique ships

Post by Mr. Missed Her » February 3rd, 2017, 7:41 pm

I was messing around with the almost knightship a while back. I found a way to add cells to generation 5 so generation 6 is the same as generation 0, albeit a three-cell-long diagonal spark that doesn't do much.

Code: Select all

x = 15, y = 21, rule = LifeHistory
11.2C$13.C$5.C5.2A.C$11.2A.C$6.3A2.A$6.3A.2A$2.2A2.A.2A.A$.3A4.A2.2A$
3A4.2A2.2A$.4A3.A$2.3A2.2A$.A2.A$.4A3.A$.A4.A$2A.A2.A$3.A.A$.A.A5.2A$
2.A6.2A$2.A3.2A.A$3.2A.2A$5.A!
Hope that helps. That spark might be useful for a tagalong that applies those cells at generations 6n+5.

Edit: Also, if we find a stabilization for the almost knightship, I'd name it the almost not knightship.
Last edited by Mr. Missed Her on March 5th, 2017, 5:34 pm, edited 2 times in total.
There is life on Mars. We put it there with not-completely-sterilized rovers.
And, for that matter, the Moon, Jupiter, Titan, and 67P/Churyumov–Gerasimenko.

User avatar
muzik
Posts: 5614
Joined: January 28th, 2016, 2:47 pm
Location: Scotland

Re: Searching for tiny oblique ships

Post by muzik » March 3rd, 2017, 3:07 pm

A for awesome, in Spaceship Discussion Thread wrote:
EDIT 2: I also just ran (3,1)c/9 width 9 with knight2. Longest partial:

Code: Select all

..*.**
.**.*
.**..*..*
.**..*.**
..*...**
......*
...**.**
......*
.....**
...***
...**
...*
Unfortunately, as explained later in said thread, the search program is buggy with (3,x)c/9 speeds.


Come on, we need to find a baby knight before this game reaches the age of 50.

Post Reply