Thread for basic questions

[dvgrn]

This method is indeed simple enough to tell the relative color (preserved/changed) of the input&output gliders of a reflector. It does not rely on the definition of absolute color given in the wiki for glider. Then, what's the point of defining the "absolute" color of a glider, if its difference is actually what we care about?

Well, the method I gave requires being in some kind of LifeHistory-like mode, and being able to see the intersection point between two 90-degree glider paths. A lot of times Herschel circuitry or something else kind of gets in the way, in practice.

And you also have to remember which direction the gliders are going. If you put the output glider on the same lane but going the other way, it suddenly becomes a same-color glider instead of an opposite-color glider.

So... in terms of what glider color actually gets used for, there isn't much point in knowing absolute color. But in a lot of cases it's much easier to determine absolute color for two gliders than to figure out their relative colors -- e.g., you can't use the above trick for two gliders unless they're traveling on perpendicular paths.

Instead, the usual thing if you're looking at a couple of gliders in Golly is to pick a leading cell in one of them and take the sum of the X and Y coordinates. Then find the same cell in the other glider when it's in a phase that's a rotation of the first one. Even coordinate sums are one absolute color, odd sums are the other absolute color.

-- You don't really have to add anything, of course. (Odd, odd) and (even, even) coordinates are one absolute color, (even, odd) and (odd, even) coordinates are the other color, and the gliders are the same relative color if they're both in the red group or both in the blue group.

[GUYTU6J]

...
Instead, the usual thing if you're looking at a couple of gliders in Golly is to pick a leading cell in one of them and take the sum of the X and Y coordinates. Then find the same cell in the other glider when it's in a phase that's a rotation of the first one. Even coordinate sums are one absolute color, odd sums are the other absolute color.

-- You don't really have to add anything, of course. (Odd, odd) and (even, even) coordinates are one absolute color, (even, odd) and (odd, even) coordinates are the other color, and the gliders are the same relative color if they're both in the red group or both in the blue group.

Okay, I got it that defining absolute color is for easy computation of relative colors. Makes sense because we often do so for things like altitude or potential energy. There's of course some confusion similar to "why do we define the imaginary unit to be √(-1) instead of -√(-1)", but this is not a big issue.
The reason I asked is that the "rotation-but-no-reflection" part of the definition is not so obvious. The previous method with the help of LifeHistory for reflector input/output seems to be a workaround.

[dvgrn]

The reason I asked is that the "rotation-but-no-reflection" part of the definition is not so obvious. The previous method with the help of LifeHistory for reflector input/output seems to be a workaround.

There's kind of a nice mnemonic that goes with it. If you look at the cells on the inside corner of the turn, and mentally replace each cell with a "C", you get either "CC" for color-changing, or just one "C" for color-preserving.

[gameoflifemaniac]

Any with smaller population?

Code: Select all

x = 8, y = 12, rule = B3/S23
3b2o$3b2o$2bo2bo$b2o2b2o$3b2o$2bo2bo$o6bo$o6bo$bo4bo2$4bo$3b2o!

Code: Select all

x = 2, y = 5, rule = B3/S23
2o$o$o$o$2o!

[Hunting]

Code: Select all

x = 2, y = 5, rule = B3/S23
2o$o$o$o$2o!

This was confirmed to be the smallest predecessor of Pulsar back in the Guokr days, I think.

[gameoflifemaniac]

Is it possible to make an oscillator search program to look for oscillators like Rob's p16?

[Saka]

One more question about QuickLife and it's bricks: Do the bricks overlap with each other? Im assuming not.

[calcyman]

One more question about QuickLife and it's [sic] bricks: Do the bricks overlap with each other? Im assuming not.

Yes, they're disjoint.

[Saka]

One more question about QuickLife and it's [sic] bricks: Do the bricks overlap with each other? Im assuming not.

Yes, they're disjoint.

Alright, thanks. And sorry for the bad grammar (My excuse is "It was 4 AM when I wrote that").

[gameoflifemaniac]

How much would the 0E0P metapillar weigh in RLE format?

[calcyman]

How much would the 0E0P metapillar weigh in RLE format?

An individual B3/S23 metacell is 66162592 bytes in RLE format, so a meta-Caterpillar would probably be about 780 terabytes.

[Moosey]

How much would the 0E0P metapillar weigh in RLE format?

An individual B3/S23 metacell is 66162592 bytes in RLE format, so a meta-Caterpillar would probably be about 780 terabytes.

When you need 780 of your own computer to store a large object in RLE format you know it's truly large

[Hdjensofjfnen]

How much would the 0E0P metapillar weigh in RLE format?

An individual B3/S23 metacell is 66162592 bytes in RLE format, so a meta-Caterpillar would probably be about 780 terabytes.

Alternatively, make a meta-metacell, which will take up around five petabytes. Luckily, after simulating it for enough generations, it'll eventually take up zero bytes.

[Hunting]

How did they find Hertz in 1970? There's absolutely no available search programs at that time.

[Ian07]

How did they find Hertz in 1970? There's absolutely no available search programs at that time.

Manual experimentation. Just make an O shape with a 3x4 inside, place a dot inside it in the right spot, and stabilize the edges as you would normally and voila, Hertz oscillator.

[Hunting]

How did they find Hertz in 1970? There's absolutely no available search programs at that time.

Manual experimentation. Just make an O shape with a 3x4 inside, place a dot inside it in the right spot, and stabilize the edges as you would normally and voila, Hertz oscillator.

Huh. I wonder what will I get in LeapLife if I do that with a script. The pinwheel was find with the same method I suppose?

[Hunting]

What is sngdetect?

[Hunting]

I'm mainly searching with gfind, which uses diagonal width rather than orthogonal width that LeapLife Spaceship Search Status uses. Are there any programs using orthogonal width? Like, a nt modification of knight2?

BTW, what does "Searching for speed c/7, width 7, glide-reflect symmetry." in gfind mean? A 2c/14 spaceship that flips itself every 7 gen?

[praosylen]

I'm mainly searching with gfind, which uses diagonal width rather than orthogonal width that LeapLife Spaceship Search Status uses. Are there any programs using orthogonal width? Like, a nt modification of knight2?

rlifesrc and some modifications of ntzfind are the only ones I know of. (I don't know much about those particular ntzfind modifications, but if you're interested you could ask in the ntzfind thread.)

BTW, what does "Searching for speed c/7, width 7, glide-reflect symmetry." in gfind mean? A 2c/14 spaceship that flips itself every 7 gen?

Yes, that's correct.

[Hunting]

I'm mainly searching with gfind, which uses diagonal width rather than orthogonal width that LeapLife Spaceship Search Status uses. Are there any programs using orthogonal width? Like, a nt modification of knight2?

rlifesrc and some modifications of ntzfind are the only ones I know of. (I don't know much about those particular ntzfind modifications, but if you're interested you could ask in the ntzfind thread.)

BTW, what does "Searching for speed c/7, width 7, glide-reflect symmetry." in gfind mean? A 2c/14 spaceship that flips itself every 7 gen?

Yes, that's correct.

Ah, alright.

New question:

How does dr work?

[wildmyron]

What is sngdetect?

From the readme:

About sngdetect

This program generates patterns that are combinations of common objects, and
for each one, it determines what happens when a glider hits it. In particular,
it checks whether the original pattern re-appears, possibly translated but at
its original orientation, after the reaction settles down. While the pattern is
running, the program detects and eliminates escaped gliders in order to
increase the speed at which the pattern is run.

In other words, it is aiming to find glider reflectors which optionally may be shifted by the reflection reaction. Finding a matching pair of push reflect and pull reflect reactions would allow construction of simple adjustable speed spaceships such as those in the Rules with small adjustable spaceships thread. More details in the readme.

Original: Stable and Glider, also available on Github with added Makefile to simplify compilation.

New question:

How does dr work?

I can't really answer this question very well, but have you read the documentation provided with the program? It has an in depth explanation of the algorithm and can be found in the zip file attached to the first post in the dr thread.

[Hunting]

May I ask:

1. How did they find Iwona, Lidka, etc.?

2. Much more cryptic is, what kind of restriction did this page use to prevent Trivial Glider+Blinker Collision Pseudo-Methuselah(TM) from getting into the collision?

[wildmyron]

2. Much more cryptic is, what kind of restriction did this page use to prevent Trivial Glider+Blinker Collision Pseudo-Methuselah(TM) from getting into the collision?

All of the methuselae on that page are contained within small bounding boxes so presumably all the candidate patterns were also contained within those bounding boxes. A Glider+Blinker collision within a small bounding box has no chance of competing with other methuselah lifespan.

As for the methuselae on the linked pages, different methods would have to have been used. Nick Gotts, I believe, used predefined lists of small clusters and tested the result of pairs of clusters at large distances - it would be easy to exclude gliders from those lists of clusters. An other option is to simply exclude glider collisions from the results in a post-processing step.

[Hunting]

2. Much more cryptic is, what kind of restriction did this page use to prevent Trivial Glider+Blinker Collision Pseudo-Methuselah(TM) from getting into the collision?

All of the methuselae on that page are contained within small bounding boxes so presumably all the candidate patterns were also contained within those bounding boxes. A Glider+Blinker collision within a small bounding box has no chance of competing with other methuselah lifespan.

As for the methuselae on the linked pages, different methods would have to have been used. Nick Gotts, I believe, used predefined lists of small clusters and tested the result of pairs of clusters at large distances - it would be easy to exclude gliders from those lists of clusters. An other option is to simply exclude glider collisions from the results in a post-processing step.

How do you define 'small'? And as far as I know Iwona is definitely not small. Or are you saying that list is using the measure mentioned below?

[wildmyron]

How do you define 'small'? And as far as I know Iwona is definitely not small. Or are you saying that list is using the measure mentioned below?

I was referring solely to the patterns discovered by Rokicki. I agree, Iwona is not contained within a small bounding box, though it's not particularly large either. I didn't really attempt to answer your question about Iwona and Lidka because I don't know the answer, but I suspect it would have involved some targeted survey of combination of small patterns based on long lasting interactions which had been previously discovered.