Thread for basic questions

[C28]

Why has interest in conduits declined these days?

most likely because most of the objects that conduit searches are done on (herschel, r-pentomino, century, etc.) have been "squeezed dry" of conduits. to put it another way, all of the conduit searches have left virtually no conduits undiscovered, making almost all future searches kind of pointless. the more recently recognized objects, like the U-turner, Blonk-tie, two-glider octomino, etc. are not as throughtly searched yet, so there may still be some hope for conduit searches.

[GUYTU6J]

Why has interest in conduits declined these days?

most likely because most of the objects that conduit searches are done on (herschel, r-pentomino, century, etc.) have been "squeezed dry" of conduits. to put it another way, all of the conduit searches have left virtually no conduits undiscovered, making almost all future searches kind of pointless. the more recently recognized objects, like the U-turner, Blonk-tie, two-glider octomino, etc. are not as throughtly searched yet, so there may still be some hope for conduit searches.

My first thought: among other factors, the absence of leading efforts from some particular users, especially M&C.
My second thought: the sequence-based view is incomplete; stagnation of catalyst development, particularly stable ones, can also account for that phenomenon. These days program searches for genuinely novel catalytic reactions, or at least replacements for old failed catalysts, have been done less and less as far as I see.
From a less pessimistic perspective, perhaps other interesting aspects of GoL have caught more attention, and the classical set of conduits are already sufficient.

---
Can soneome have a look at my previous questions here?

[calcyman]

Why has interest in conduits declined these days?

Conduits have mainly been seen as useful components for building larger constructions (guns, glider-to-spaceship converters, signal circuitry, and suchlike), rather than as objects of independent interest. There have been sufficiently many conduits discovered that both the probability and marginal utility of finding another one has diminished.

Finally, a lot of Herschel track plumbing is used to solve the problem "move a signal from here to there in n generations", and (thanks to the snark, syringe, and bandersnatch) it's easier to use gliders as the base signal instead of Herschels. For example, when AbhpzTa discovered that marvellous conduit that splits a Herschel into two Herschels with reasonably low recovery time, it certainly caused a lot of well-deserved excitement -- but not as much excitement as it would have done had the syringe not existed at the time.

[mniemiec]

Why has interest in conduits declined these days?

most likely because most of the objects that conduit searches are done on (herschel, r-pentomino, century, etc.) have been "squeezed dry" of conduits. to put it another way, all of the conduit searches have left virtually no conduits undiscovered, making almost all future searches kind of pointless. the more recently recognized objects, like the U-turner, Blonk-tie, two-glider octomino, etc. are not as throughtly searched yet, so there may still be some hope for conduit searches.

I would think that there might be a fair bit of ground that has been previously searched, but then discarded as uninteresting, but might still be useful. E.g. before U-turners were considered important, B-to-U-turner reactions might have been found and discarded.

[C28]

I would think that there might be a fair bit of ground that has been previously searched, but then discarded as uninteresting, but might still be useful. E.g. before U-turners were considered important, B-to-U-turner reactions might have been found and discarded.

good point.

(although B-to-U-turner converters would be great to have, since the "U-turner as output" section of the U-turner gallery only has a single X-to-U, HBx63U)

[dvgrn]

Why has interest in conduits declined these days?

It might be because we already have conduits that solve most of the problems that we usually want to solve using conduits.

Many of the conduits that were found early on are relatively small and simple, which is why they were relatively easy to find. That means that your average new conduit tends to be larger and more complicated than known conduits, without necessarily solving any new problems that couldn't already be solved in some other way... so a lot of discoveries don't really find uses anywhere. That might tend to reduce interest quite a bit -- even if something "new" is found, it might well not turn out to be particularly "new and interesting".

There are definitely a quite a few missing conduits that would be great to find -- small Spartan 90-degree glider reflector, small color-changing 90-degree glider reflector, direct G-to-2G splitter, direct G-to-B / G-to-R / G-to-pi / etc., smaller or with different geometry than a syringe -- and so on. So there would be a lot of interest if any of those suddenly showed up -- but if past experience is any guide, such things are rather hard to find!

[wwei47]

What is yl8_1_3_1463956f33241332e72e4a2c9f4ecc61? It's in ikpx2_stdin, but no soups are saved.

[LLAMASKYWALKER]

How do i view the population graph of a pattern in golly

[dvgrn]

How do i view the population graph of a pattern in golly

Scripts/Lua/pop-plot.lua .

[NimbleRogue]

Weird question, but does anyone know how to remove all duplicate entries in excell
for example, I want to turn
ac
ac
.
.
.
into
ac
.
.
.
Any advice?

[hotdogPi]

Weird question, but does anyone know how to remove all duplicate entries in excell
for example, I want to turn
ac
ac
.
.
.
into
ac
.
.
.
Any advice?

Create a second column. In that column, put a formula that gets the row number of the first result that matches that row in column A. If the obtained value doesn't match the row number, it's a duplicate.

[dvgrn]

Weird question, but does anyone know how to remove all duplicate entries in excell

Search the Help in Excel for "Remove Duplicates". There's actually a dedicated function that does exactly that.

If your screen is wide enough, I believe the "Data" ribbon will include a labeled "Remove Duplicates" tile under Data Tools. On a normal laptop screen, all you might get is a cryptic icon, so look for this ...

remove-duplicates.png (895 Bytes) Viewed 1363 times

[Herschel of Ostropol]

Is there a shuttle (or other oscillator) that does the following?:
The oscillator places an object that stabilizes within oscillator's period, which does not disrupt the oscillator's sequence. But, the object does not destroy, or get destroyed by, the oscillator in any of the next cycles.

or

Is there a still life with an island which, when the island is destroyed, replaces the destroyed island?

[dvgrn]

Is there a shuttle (or other oscillator) that does the following?:
The oscillator places an object that stabilizes within oscillator's period, which does not disrupt the oscillator's sequence. But, the object does not destroy, or get destroyed by, the oscillator in any of the next cycles.

You could arrange any of the "keeper" conduits into a composite oscillator that matches those conditions. There might be a natural, "elementary" high-period oscillator or two that is an object-keeper along these lines, but I don't remember any such thing offhand.

This p518 oscillator has the "keeper" property, for example -- the block at the right.

Code: Select all

x = 174, y = 83, rule = LifeHistory
5.A18.A$5.3A7.2A5.3A$8.A6.2A4.A$7.2A12.2A3$20.2A$15.2A3.2A$15.2A4$34.
A14.2A$32.3A14.A24.2A$31.A18.3A6.2A13.A$31.2A19.A6.2A11.A.A$72.2A2$
92.A$46.D36.2A5.3A$A22.A21.D37.2A4.A$3A20.3A18.2D43.2A$3.A19.A.A15.2A
2.2D$2.2A21.A15.2A3.D26.D$71.D.D14.2A$71.3D9.2A3.2A$71.D11.2A3$44.2A$
24.2A19.A$24.A.A15.3A13.2A3.2A$26.A9.A5.A16.A3.A$26.3A5.3A19.3A5.3A$
29.A3.A16.A5.A9.A$28.2A3.2A13.3A15.A.A$47.A19.2A85.A$47.2A64.A38.3A$
111.3A37.A$110.A40.2A$8.2A11.D88.2A$3.2A3.2A9.3D76.A$3.2A14.D.D76.3A$
19.D26.D3.2A15.D21.2A10.A$46.2D2.2A15.D.D19.A10.2A$2.2A43.2D18.3D20.
3A$3.A4.2A37.D21.D16.A5.A$3A5.2A36.D39.3A25.2A11.A$A88.A24.2A11.3A$
88.2A40.A$19.2A108.2A$18.A.A11.2A6.A19.2A91.2A$18.A13.2A6.3A18.A91.2A
$17.2A24.A14.3A$42.2A14.A$138.D$136.3D$119.2A15.D.D33.2D$76.2A13.2A
26.2A15.D35.2D$71.2A3.2A13.2A$71.2A3$70.2A$71.A4.2A38.2A$68.3A5.2A38.
A19.2A$68.A48.A17.A.A$116.2A17.A$134.2A3$159.A$158.A.A$159.2A6$138.2A
5.2A$139.A6.A4.2A$136.3A4.3A5.2A$136.A6.A!

EDIT: Aha, hotdogPi is right -- this doesn't quite count either, because the block gets destroyed and re-formed. We need a structure that suppresses its own construction reaction. Such things definitely do exist, but it will take a bit longer to come up with an example.

[hotdogPi]

This p128 is a blinker keeper. However, it doesn't quite satisfy the initial question, as the blinker gets destroyed and reformed.

Code: Select all

x = 53, y = 47, rule = B3/S23
31b2o7b2o$31bo2b2ob2o2bo$32b2obobob2o$33bo5bo$33bo5bo$34b5o$35b3o$35bo
bo$35b3o$35b3o$36bo2$45bo$45b2ob2o$23bo22bobobo$23bo21b2ob2o$23bo23b2o
$46b3obo$42b3o4bob2o3$29b3o$29bobo$21bobo5bobo$21bobo$21b3o3$2obo4b3o$
2bob3o$4b2o23bo$3b2ob2o21bo$2bobobo22bo$3b2ob2o$7bo2$16bo$15b3o$15b3o$
15bobo$15b3o$14b5o$13bo5bo$13bo5bo$12b2obobob2o$11bo2b2ob2o2bo$11b2o7b
2o!

[confocaloid]

Is there a still life with an island which, when the island is destroyed, replaces the destroyed island?

The still life xs14_mmge13z1 has the following property: if you remove the block island then after 26 ticks the rest evolves to a R-pentomino (and a spark which does not interact with the R-pentomino).

In the following pattern, at generation 95 six gliders collide to form two SLs, a beehive and a xs14_mmge13z1. At generation 249, two more gliders (these gliders are added for illustrative purpose and should not actually be present) collide to remove the block island from the xs14_mmge13z1. 26 ticks later (generation 275) you can see the R-pentomino and a dying spark. At generation 303, the R-pentomino evolves to a B-heptaplet and another active region (the latter collides with the beehive to release a single glider before dying). At generation 362, the B-heptaplet is transformed into a Herschel. At generation 437 two NW gliders escaped and the rest is stabilized.

Code: Select all

x = 122, y = 129, rule = LifeHistory
120.A$119.A$119.3A42$91.A$90.A$90.3A$45.A.A$46.2A$46.A$33.A$31.A.A$
32.2A16$68.2D$64.2D.D.D27.D$64.2D.D27.3D$67.D10.2E15.D.D$64.3D11.2E
15.D$64.D41.E$59.D45.E.E$58.D.D44.E.E$58.D.D45.E$59.D$75.2E$75.E$76.
3E$78.E12$37.3A44.2A$39.A44.A.A$38.A45.A3$35.2A$34.A.A$36.A26$2A$.2A$
A!

If stable reflectors and duplicators are added to convert the two escaped gliders into six gliders which are needed to rebuild the beehive and the xs14_mmge13z1 again in the same location, the resulting pattern will be a constellation of still lives that eventually recovers when the block island of xs14_mmge13z1 is removed. If completed, would this be a solution?

[squareroot12621]

There are zero 1-glider collisions. (It can't collide with anything.)
There are seventy-one 2-glider collisions. (http://pentadecathlon.com/lifeNews/2005 ... sions.html)
There are _____ 3-glider collisions.
I want to fill in the _____, and I think I have an idea for how to do it.

1. Run through each 2-glider collision.
2. Count the number of ways to interact with the junk produced, accounting for oscillators and their periods.
3. Check them for invalid collisions.
4. Add the valid collisions up.

I know, it will take some work, but I have a gut feeling that it will take less than a year. (At least, I hope!)

[hotdogPi]

See viewtopic.php?f=9&t=2833

[mniemiec]

There are zero 1-glider collisions. (It can't collide with anything.)
There are seventy-one 2-glider collisions. (http://pentadecathlon.com/lifeNews/2005 ... sions.html)
There are _____ 3-glider collisions.
I want to fill in the _____, and I think I have an idea for how to do it.

1. Run through each 2-glider collision.
2. Count the number of ways to interact with the junk produced, accounting for oscillators and their periods.
3. Check them for invalid collisions.
4. Add the valid collisions up.

I know, it will take some work, but I have a gut feeling that it will take less than a year. (At least, I hope!)

Unfortunately, the number is "infinity", because several 2-glider collisions produce constellations that include both stationary objects and gliders, an the third glider can collide with the outgoing glider in many ways at an infinite number of different distances from the stationary objects. Granted, most of these are trivial; over half just kill the glider anyway, and most of the others just leave some stationary junk - which leads to an infinite number of different output constellations, with two lumps of junk separated by an arbitrary distance.

It gets more complicated if the second collision fires a glider back in the direction of the first constellation, possibly re-igniting it; the results can become non-trivial if the first constellation grows into a large mess that is close enough to the second constellation to impact it before settling down (and similar, if it shoots a glider back towards the second constellation).

These additional complications are why nobody has yet counted all the 3 glider collisions. Back in the '80s or '90s, David Buckingham manually tracked most of the trivial ones, but there were a few that slipped between the cracks (evidenced by the fact that over the next 30 years, a handful of new 3-glider syntheses have been found: pentadecathlon, bi-pond, and BLSE), and he didn't have the resources to track all the glider-hits-escaping-glider ones.

[hkoenig]

Following up on Niemiec's post that there are an infinite number of 3 Glider collisions with a concrete example.

Consider the following pair of 3 Glider collisions:

Code: Select all

x=49, y=32
33bo$34b2o$33b2o$o$b2o$2o18$46bo$11bo33bo$10bo34b3o$10b3o2$46b3o$11b3o32b
o$11bo35bo$12bo!

In both cases the collision starts with a pair of Gliders resulting in a Traffic Lights with an escaping Glider. A third Glider is positioned to collide with the escaping Glider, producing a Loaf. Reversing the third Glider by 8 generations causes the resulting Loaf to be removed on cell diagonally to the northwest. That process can be repeated indefinitely, each time resulting in a Loaf farther and farther from the Traffic Lights, all of the resulting patterns being unique.

Long time ago (early 1990s?) I wanted to try and enumerate all the 3 Glider collisions. I did come up with a quick way to reduce any pair of Gliders to either one of the 71 collisions or determine "no-collision" (but I seem to have lost my notes and programming on how I did that). I thought it could lead to a similar process with three, but I never could come up with a satisfactory way to recognize and stop this infinite removal process and specify that the part of result was dependent on the position of the incoming 3rd Glider.

(Similar problem with trying to enumerate what happens when two Gliders strike a single target object. How do you handle the Kickback reaction?)

[Book]

Can anyone point me to rle or apgcode for oscillators Merzenich's p31 on 37P7.1 and pentadecathlon on centinal. Thanks.

[hotdogPi]

Can anyone point me to rle or apgcode for oscillators Merzenich's p31 on 37P7.1 and pentadecathlon on centinal. Thanks.

Merzenich's p31 on 37P7.1 doesn't really deserve a mention. The only reason 217 has an entry is because of a larger p217 that exists and was very briefly mentioned. That row can go away.

[Book]

Can anyone point me to rle or apgcode for oscillators Merzenich's p31 on 37P7.1 and pentadecathlon on centinal. Thanks.

Merzenich's p31 on 37P7.1 doesn't really deserve a mention. The only reason 217 has an entry is because of a larger p217 that exists and was very briefly mentioned. That row can go away.

Can you elaborate on the reasoning? I'd like to understand the decision making process. Also, "very briefly mentioned" by whom and in what context? I don't question the judgements, but I also don't really understand them. Again, thanks.

[hotdogPi]

Can anyone point me to rle or apgcode for oscillators Merzenich's p31 on 37P7.1 and pentadecathlon on centinal. Thanks.

Merzenich's p31 on 37P7.1 doesn't really deserve a mention. The only reason 217 has an entry is because of a larger p217 that exists and was very briefly mentioned. That row can go away.

Can you elaborate on the reasoning? I'd like to understand the decision making process. Also, "very briefly mentioned" by whom and in what context? I don't question the judgements, but I also don't really understand them. Again, thanks.

On the main oscillator page, Raucci's p217 (you know, the one with the U-turners) was on the list because it was the only notable p217. However, it was quickly found that that wasn't the smallest p217, so it got changed to the LCM, while it would have been much more reasonable to delete the row entirely and leave it to the page that has the table going up to 384.

[LLAMASKYWALKER]

How do I run rules on a x by 1 grid in lifeviewer?