Thread for basic questions

For general discussion about Conway's Game of Life.
User avatar
dvgrn
Moderator
Posts: 8110
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Thread for basic questions

Post by dvgrn » June 19th, 2021, 10:44 pm

ColorfulGabrielsp138 wrote:
June 19th, 2021, 9:55 pm
Why did you prefer RLE instead of apgcode?
I can copy RLE straight out of Golly, and someone can copy and paste RLE straight back into Golly and see what it is -- or get a picture of it by decoding it manually, for small enough patterns.

I'm not as confident of my ability to get the picture exactly right for a small-to-middling apgcode -- just haven't practiced as much.

User avatar
bubblegum
Posts: 938
Joined: August 25th, 2019, 11:59 pm
Location: click here to do nothing

Re: Thread for basic questions

Post by bubblegum » June 20th, 2021, 12:01 am

ColorfulGabrielsp138 wrote:
June 19th, 2021, 9:55 pm
Why did you prefer RLE instead of apgcode?
Why did you prefer apgcode instead of RLE?
Each day is a hidden opportunity, a frozen waterfall that's waiting to be realised, and one that I'll probably be ignoring
sonata wrote:
July 2nd, 2020, 8:33 pm
conwaylife signatures are amazing[citation needed]
anything

User avatar
yujh
Posts: 2385
Joined: February 27th, 2020, 11:23 pm
Location: Toronto, On, Canada
Contact:

Re: Thread for basic questions

Post by yujh » June 20th, 2021, 12:06 am

bubblegum wrote:
June 20th, 2021, 12:01 am
ColorfulGabrielsp138 wrote:
June 19th, 2021, 9:55 pm
Why did you prefer RLE instead of apgcode?
Why did you prefer apgcode instead of RLE?
Yeah, same question

mniemiec
Posts: 1311
Joined: June 1st, 2013, 12:00 am

Re: Thread for basic questions

Post by mniemiec » June 20th, 2021, 12:23 am

ColorfulGabrielsp138 wrote:
June 19th, 2021, 9:55 pm
Why did you prefer RLE instead of apgcode?
One advantage RLE has over apgcode is that it encodes the rule with the pattern, making it easy to copy/paste a pattern into a progarm like Golly, instead of the rule having to be carried through a separate and independent channel. apgcode is also only suitable for encoding a small subset of patterns (i.e. still-lifes, oscillators, and spaceships); patterns that aren't one of these can't easily be encoded. In particular, linearly expanding patterns are encoded as yl_... which only encodes a histogram of the pattern's population, but not its actual bit image, making the mapping from apgcode to bit image not unique, and also impossible to reverse. xs0_... is a work-around for arbitrary patterns, but apgcode wasn't really designed for that. It also can't encode images in rules with more than two states.

apgcode is useful for encoding many small one-line pattern images together in one large database.

ColorfulGabrielsp138
Posts: 174
Joined: March 29th, 2021, 5:45 am

Re: Thread for basic questions

Post by ColorfulGabrielsp138 » June 20th, 2021, 12:45 am

dvgrn wrote:
June 19th, 2021, 10:44 pm
ColorfulGabrielsp138 wrote:
June 19th, 2021, 9:55 pm
Why did you prefer RLE instead of apgcode?
I can copy RLE straight out of Golly, and someone can copy and paste RLE straight back into Golly and see what it is -- or get a picture of it by decoding it manually, for small enough patterns.

I'm not as confident of my ability to get the picture exactly right for a small-to-middling apgcode -- just haven't practiced as much.
I had an apgcode to RLE converter program but I can't attach it to Golly.

EDIT: How common is the xp0_16o4 relative to the xp0_15o4 or the Herschel parent from the B-heptomino?
Have you ever seen xp0_16o4 naturally appear from random soup?

EDIT 2: Can this be considered a "turner"? Is it color-preserving or color-changing?

Code: Select all

x = 12, y = 12, rule = TripleB3S23
2F$.F$.2F$2.2F6$9.3F$9.F$10.F!

Code: Select all

x = 21, y = 21, rule = LifeColorful
11.E$10.3E$10.E.2E$13.E4$2.2B$.2B$2B$.2B15.2D$19.2D$18.2D$17.2D4$7.C$
7.2C.C$8.3C$9.C!
I'm free at last

hkoenig
Posts: 175
Joined: June 20th, 2009, 11:40 am

Re: Thread for basic questions

Post by hkoenig » June 20th, 2021, 1:11 am

Also, apgcode will lose any location and transform information which can be necessary when placing objects. If you wanted to use it as an RLE substitute, you need to include that info.

I've found that an object oriented Life editor can have some advantages over the standard infinite bitplane when it comes to manipulating patterns that can be broken down into component parts, like Glider constructions. The editor applies transforms to the canonical objects and then combines them all into the bitpattern that gets run. Make a modification to the transform-- rebuild the pattern and run again.

Also, the apgcode prefix gets in the way, as it is more for gross indexing and a convenience to human readers. (And incomplete, as I would prefer it contain both period and bits for oscillators, for example.) Without the prefix, it does function as a very useful key/index for databases, or in places where you want to compare objects for equality.

ColorfulGabrielsp138
Posts: 174
Joined: March 29th, 2021, 5:45 am

Re: Thread for basic questions

Post by ColorfulGabrielsp138 » June 20th, 2021, 3:04 am

hkoenig wrote:
June 20th, 2021, 1:11 am
(And incomplete, as I would prefer it contain both period and bits for oscillators, for example.)
You can try this: x3p2_7 (it just won't load sample soups and comments)

Code: Select all

x = 21, y = 21, rule = LifeColorful
11.E$10.3E$10.E.2E$13.E4$2.2B$.2B$2B$.2B15.2D$19.2D$18.2D$17.2D4$7.C$
7.2C.C$8.3C$9.C!
I'm free at last

User avatar
yujh
Posts: 2385
Joined: February 27th, 2020, 11:23 pm
Location: Toronto, On, Canada
Contact:

Re: Thread for basic questions

Post by yujh » June 20th, 2021, 3:08 am

ColorfulGabrielsp138 wrote:
June 20th, 2021, 3:04 am
hkoenig wrote:
June 20th, 2021, 1:11 am
(And incomplete, as I would prefer it contain both period and bits for oscillators, for example.)
You can try this: x3p2_7 (it just won't load sample soups and comments)
It is just a magical example that fits the apgcode. False claim.

User avatar
bibunsekibun
Posts: 176
Joined: April 17th, 2021, 7:58 pm
Location: TRAPPIST-1e

Re: Thread for basic questions

Post by bibunsekibun » June 20th, 2021, 5:40 am

Is there an Phi Spark Synthesis?
I'm sleepy
I'm not good at English, so please let me know if you have any grammatical mistakes
My Favorite Pattern is 60P312

User avatar
yujh
Posts: 2385
Joined: February 27th, 2020, 11:23 pm
Location: Toronto, On, Canada
Contact:

Re: Thread for basic questions

Post by yujh » June 20th, 2021, 5:45 am

bibunsekibun wrote:
June 20th, 2021, 5:40 am
Is there an Phi Spark Synthesis?
............^
............|
I don’t think there should be an n
Anyways, yes, but I don’t think there’s a clean 2g one

Selena Silverstep
Posts: 2
Joined: June 20th, 2021, 12:00 pm

Re: Thread for basic questions

Post by Selena Silverstep » June 20th, 2021, 12:05 pm

Is it possible to run apgsearch using custom seeds?

User avatar
JP21
Posts: 1495
Joined: October 26th, 2019, 5:39 am
Location: PH

Re: Thread for basic questions

Post by JP21 » June 20th, 2021, 1:22 pm

bibunsekibun wrote:
June 20th, 2021, 5:40 am
Is there a Phi Spark Synthesis?
I have these lying around:

Code: Select all

x = 61, y = 50, rule = B3/S23
13b2o$6b3o4bobo$8bo4bo$7bo4$16b2o$16bobo$16bo$6b2o$7b2o$6bo$50bo$37b3o
9b2o$39bo9bobo$38bo3$31b3o$33bo$32bo$7b2o$3o4bobo$2bo4bo$bo4$10b2o$10b
obo$10bo2$58bo$57b2o$57bobo2$49b3o$51bo$21bo28bo$20b2o17bo$20bobo15b2o
18b3o$38bobo17bo$12b3o44bo$14bo15b3o$13bo18bo$31bo$20b2o19b3o$20bobo
18bo$20bo21bo!

mniemiec
Posts: 1311
Joined: June 1st, 2013, 12:00 am

Re: Thread for basic questions

Post by mniemiec » June 20th, 2021, 2:48 pm

bibunsekibun wrote:
June 20th, 2021, 5:40 am
Is there an Phi Spark Synthesis?
The easiest way to make a phi is via a glider-blinker collision. There is also a 2-glider collision that edge-shoots a phi with some junk behind it; there are many ways of suppressing that junk with a third glider, if necessary:

Code: Select all

x = 26, y = 8, rule = B3/S23
24bo$23bo$o22b3o$o20bo$o19boo$bboo16bobo$bbobo$bbo!

MathAndCode
Posts: 4804
Joined: August 31st, 2020, 5:58 pm

Re: Thread for basic questions

Post by MathAndCode » June 20th, 2021, 10:21 pm

mniemiec wrote:
June 20th, 2021, 2:48 pm
The easiest way to make a phi is via a glider-blinker collision.
A boat also can serve as a one-glider seed for a phi spark.

Code: Select all

x = 5, y = 10, rule = B3/S23
3bo$2bobo$3b2o5$bo$b2o$obo!
I have historically worked on conduits, but recently, I've been working on glider syntheses and investigating SnakeLife.

User avatar
dvgrn
Moderator
Posts: 8110
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Thread for basic questions

Post by dvgrn » June 20th, 2021, 11:20 pm

Selena Silverstep wrote:
June 20th, 2021, 12:05 pm
Is it possible to run apgsearch using custom seeds?
Yes! A starting point is Catagolue as a generic pattern inventory.

It does seem as if there should be a quick LifeWiki tutorial written giving all the details of an example of how to run a set of RLEs through apgsearch and upload them to Catagolue. With enough encouragement I might try doing that for some sample database -- the current set of octohash collisions, let's say, just to see if any likely-looking expensive objects show up where a synthesis might be improved.

If you're familiar with piping the output of one executable to the input of another, then a thread like this one, plus a little experimentation, should be enough to get something working.

EDIT: Here's another thread that was helpful for figuring out some of the rules -- how to recompile apgsearch, pass in -t 0 to upload to Catagolue, etc.

MathAndCode
Posts: 4804
Joined: August 31st, 2020, 5:58 pm

Re: Thread for basic questions

Post by MathAndCode » June 20th, 2021, 11:44 pm

Are these the smallest known loafer predecessors?

Code: Select all

x = 109, y = 12, rule = B3/S23
5b2o18b2o18b2o18b2o18b2o18b2o$5bo19bo19bo19bo19bo19bo$o4b2o18b2o12bo5b
2o18b2o11bo6b2o18b2o$o6bo10bo8bo12bo6bo12bo6bo12bo6bo12bo6bo$o18b4o17b
2o17b3o18b2o17b3o$o6bo19bo12bo6bo14bo4bo12bo6bo19bo$bo6bo19bo19bo19bo
19bo13bo5bo$8bo19bo19bo19bo19bo19bo$8bo19bo19bo19bo19bo19bo$7bo19bo19b
o19bo19bo19bo$7bo19bo19bo19bo19bo19bo$5bo19bo19bo19bo19bo19bo!
I have historically worked on conduits, but recently, I've been working on glider syntheses and investigating SnakeLife.

Selena Silverstep
Posts: 2
Joined: June 20th, 2021, 12:00 pm

Re: Thread for basic questions

Post by Selena Silverstep » June 21st, 2021, 12:30 am

dvgrn wrote:
June 20th, 2021, 11:20 pm
Selena Silverstep wrote:
June 20th, 2021, 12:05 pm
Is it possible to run apgsearch using custom seeds?
Yes! A starting point is Catagolue as a generic pattern inventory.

It does seem as if there should be a quick LifeWiki tutorial written giving all the details of an example of how to run a set of RLEs through apgsearch and upload them to Catagolue. With enough encouragement I might try doing that for some sample database -- the current set of octohash collisions, let's say, just to see if any likely-looking expensive objects show up where a synthesis might be improved.

If you're familiar with piping the output of one executable to the input of another, then a thread like this one, plus a little experimentation, should be enough to get something working.
Thank you for the reply! To clarify, as long as the symmetry is not recognized, the program will read RLEs from stdin instead of generating soups from SHA of a random seed, is that right? Does the symmetry have to end with 'stdin'?

wwei47
Posts: 928
Joined: February 18th, 2021, 11:18 am

Re: Thread for basic questions

Post by wwei47 » June 21st, 2021, 12:41 am

I just tell the Python apgsearch to open a file for the first soup then use a blank grid for the other soups. It's useful for sorting through JLS search results.
https://catagolue.hatsya.com/census/b3s ... 3_patterns

User avatar
bubblegum
Posts: 938
Joined: August 25th, 2019, 11:59 pm
Location: click here to do nothing

Re: Thread for basic questions

Post by bubblegum » June 21st, 2021, 12:46 am

Selena Silverstep wrote:
June 21st, 2021, 12:30 am
To clarify, as long as the symmetry is not recognized, the program will read RLEs from stdin instead of generating soups from SHA of a random seed, is that right? Does the symmetry have to end with 'stdin'?
"Unofficial" symmetries are used for apgsearch hacks, mainly. They do generate soups from SHA, but the source code is modified to interpret the result differently. Of course, this serious use of these symmetries is not the only one.

Any symmetry containing "stdin" will instead be a stdin symmetry.
Each day is a hidden opportunity, a frozen waterfall that's waiting to be realised, and one that I'll probably be ignoring
sonata wrote:
July 2nd, 2020, 8:33 pm
conwaylife signatures are amazing[citation needed]
anything

User avatar
yujh
Posts: 2385
Joined: February 27th, 2020, 11:23 pm
Location: Toronto, On, Canada
Contact:

Re: Thread for basic questions

Post by yujh » June 21st, 2021, 3:44 am

How can I make apgsearch(python or compiled) make the 'stdin' soups? (e.g. 4g)

User avatar
JP21
Posts: 1495
Joined: October 26th, 2019, 5:39 am
Location: PH

Re: Thread for basic questions

Post by JP21 » June 21st, 2021, 1:03 pm

Why has this pattern been ignored for so long? (7 months)

Code: Select all

x = 20, y = 29, rule = B3/S23
o$3o$3bo$2b2o3$3b2o$2bobo4bo$2bo5bobo$3bo4bobo$3o6bo$o17b2o$18bo$16bob
o$16b2o2$6b2o$5bo2bo$5bo2bo$4bo4bo$4bo4bo$5b4o$4b2o2b2o$4bo4bo$4bo4bo
3$5b4o$6b2o!
Is it that useless?

User avatar
dvgrn
Moderator
Posts: 8110
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Thread for basic questions

Post by dvgrn » June 21st, 2021, 1:32 pm

JP21 wrote:
June 21st, 2021, 1:03 pm
Why has this pattern been ignored for so long? (7 months)

Code: Select all

x = 20, y = 29, rule = B3/S23
o$3o$3bo$2b2o3$3b2o$2bobo4bo$2bo5bobo$3bo4bobo$3o6bo$o17b2o$18bo$16bob
o$16b2o2$6b2o$5bo2bo$5bo2bo$4bo4bo$4bo4bo$5b4o$4b2o2b2o$4bo4bo$4bo4bo
3$5b4o$6b2o!
Is it that useless?
It hasn't really been ignored, has it? You put a link to the original copy of it on the Spaceship Stable Circuitry, so it can be found when it's needed. No need to re-post -- though you can certainly post links to anything I've forgotten to link to in the Stable Signal Circuitry first post. I've done that now for the above copperhead-to-G converter.

It's certainly not useless -- 149-tick recovery beats the previous copperhead-to-G that I had linked, by a nice wide margin. The old recovery time was 495 ticks. because a long boat had to be cleaned up. No cleanup beats even single-stage cleanup any day!

MathAndCode
Posts: 4804
Joined: August 31st, 2020, 5:58 pm

Re: Thread for basic questions

Post by MathAndCode » June 21st, 2021, 1:37 pm

JP21 wrote:
June 21st, 2021, 1:03 pm

Code: Select all

x = 20, y = 29, rule = B3/S23
o$3o$3bo$2b2o3$3b2o$2bobo4bo$2bo5bobo$3bo4bobo$3o6bo$o17b2o$18bo$16bob
o$16b2o2$6b2o$5bo2bo$5bo2bo$4bo4bo$4bo4bo$5b4o$4b2o2b2o$4bo4bo$4bo4bo
3$5b4o$6b2o!
There's a replacement catalyst for when a block turns into a P-pentomino in that way? I was unaware. Thank you for informing me.

Code: Select all

x = 21, y = 10, rule = B3/S23
3bo13bo$3bo13bo$obo11bobo$bo13bo2$b2o12b2o$b2o8b2o2bo2bo$10bobo3b2obo
$10bo8bo$9b2o8b2o!


Edit: I have added it to the catalyses tutorial.
I have historically worked on conduits, but recently, I've been working on glider syntheses and investigating SnakeLife.

User avatar
dvgrn
Moderator
Posts: 8110
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Thread for basic questions

Post by dvgrn » June 21st, 2021, 7:17 pm

yujh wrote:
June 21st, 2021, 3:44 am
How can I make apgsearch(python or compiled) make the 'stdin' soups? (e.g. 4g)
Selena Silverstep wrote:
June 21st, 2021, 12:30 am
To clarify, as long as the symmetry is not recognized, the program will read RLEs from stdin instead of generating soups from SHA of a random seed, is that right? Does the symmetry have to end with 'stdin'?
There's a new draft of a tutorial about stdin symmetry on LifeWiki now. If any regular users of apgluxe's stdin functionality want to look it over and clarify anything I said that was wrong, confusing, or downright silly, that would be a Good Thing.

The tutorial doesn't entirely answer the question about re-creating the 4g soups. I think there's a brief walkthrough of that -- piping the output of a program into apgluxe, rather than piping the contents of a text file into it -- at this link that I gave before.

However, the tutorial does document the steps I had to go through to make my own custom "stdin" symmetry to census all 455,380 collisions in the octohash database. It seemed very straightforward, so I hope I've successfully explained how to do this kind of thing for similar census projects.

(As far as the octohash database census results go, I don't think anything particularly exciting showed up; I only checked the biggest 19-bit and 20-bit still lifes that appeared, and they certainly aren't going to set any records for a synthesis of those objects. The sample results are all posted at the bottom of the tutorial if anyone wants to look through them.)

User avatar
Ian07
Posts: 698
Joined: September 22nd, 2018, 8:48 am

Re: Thread for basic questions

Post by Ian07 » June 21st, 2021, 8:51 pm

dvgrn wrote:
June 21st, 2021, 7:17 pm
(As far as the octohash database census results go, I don't think anything particularly exciting showed up; I only checked the biggest 19-bit and 20-bit still lifes that appeared, and they certainly aren't going to set any records for a synthesis of those objects. The sample results are all posted at the bottom of the tutorial if anyone wants to look through them.)
Yep, can confirm there aren't any improvements in there.

Post Reply