Thread for basic questions

[mniemiec]

2: How do I find billiard table oscillators? I was able to rediscover known ones (including a P17 from jslife) from the shape of the walls, but when I hollowed out a random still life and tried to get a billiard table, I couldn't find anything. Help?

Viable billiard table oscillators follow a very delicate dance, where all the rotor cells that are adjacent to any living stator cell must have a total population that never veries by more than one over the entire period of the oscillator (i.e. so the total neighbors for every living stator cell must always be exactly 2 or 3). E.g. if a stator cell has two living neighbors, and is adjacent to two rotor cells, then either zero or one of those rotor cells must be on in every phase; if they are ever both on simultaneously, they destroy the stator. It gets even more fragile if stator cell has three or more rotor neighbors. This balance gets harder and harder to maintain, the larger the stator, and the higher the period.

[GUYTU6J]

The discovery of Galloping Gliders, an INT rule with a 2-glider replicator, may give some insight into spaceships made up of gliders. How can we apply the method of constructing replicator-based spaceships (e.g. something about linear algebra mentioned in the basilisk article) in HighLife here?
A proof of concept:

Code: Select all

x = 111, y = 208, rule = B3-er4city/S23-a4city
40bobo33bobo$40b2o34b2o$41bo35bo3$41bo$40b2o$40bobo22$63b2o$62b2o$64bo
4$77bobo$77b2o$78bo27$62b2o$61b2o$63bo4$78bobo$78b2o$79bo27$61b2o$60b
2o$62bo4$79bobo$79b2o$80bo27$60b2o$59b2o$61bo4$80bobo$80b2o$81bo27$59b
2o$58b2o$60bo4$81bobo$81b2o$82bo4$22bobo31bobo$22b2o33b2o$23bo33bo3$
23bo33bo$22b2o33b2o$22bobo31bobo12$bo35bo71bo$o37bo69bo$3o33b3o69b3o3$
3o33b3o69b3o$o37bo69bo$bo35bo71bo!

Sadly there is no known push or pull reactions for termination as for now.

[Entity Valkyrie 2]

Can I ask a question:

"Is there a still life whose inversion is also a still life? (Other than the long ship)"

[wwei23]

Can I ask a question:

"Is there a still life whose inversion is also a still life? (Other than the long ship)"

Does the block/empty space count?

[Entity Valkyrie 2]

Can I ask a question:

"Is there a still life whose inversion is also a still life? (Other than the long ship)"

Does the block/empty space count?

Not really, since a still life is not a still "life" if it has no "life".

[mniemiec]

Can I ask a question: "Is there a still life whose inversion is also a still life? (Other than the long ship)"

Could you be a bit more precise in what, exactly, you mean by "inversion"? I can't think of any transformation that turns a long ship into another long ship (other than rotation or reflection, and since the Life rule is isotropic, under that definition, all objects sharing the same symmetries would be their own inversions by definition).

[Entity Valkyrie 2]

Take the long ship bounding box. if you invert the cells in the bounding box, you get another long ship.

Code: Select all

x = 13, y = 4, rule = LifeHistory
2D2C5.2C2D$DCDC5.CDCD$CDCD5.DCDC$2C2D5.2D2C!

[Chris857]

Can I ask a question:

"Is there a still life whose inversion is also a still life? (Other than the long ship)"

Infinite: The inversion of the Zebra stripes, Chicken wire, and Onion rings agars is also a stable agar.

Constellation: a pattern of blocks separated by two cells would meet your definition of inverting the cells within the bounding box, the below would still be a set of still-lifes (more blocks). Edit: formatted similar to post above.

Code: Select all

x = 13, y = 4, rule = LifeHistory
2C2D2C2D2C$2C2D2C2D2C!

[Entity Valkyrie 2]

Any more complex still lives?

[yujh]

Is there a stong P7 domino sparker?

also, anyone who wants to finish this?

Code: Select all

x = 349, y = 128, rule = B3/S23
66bo$67b2o$66b2o$276bo$276bobo$276b2o11$81bobo$82b2o$82bo178bo$260bo$
260b3o11$97bo$98b2o$97b2o$245bo$245bobo$245b2o11$112bobo$113b2o$113bo
116bo$138b2o89bo$138bo90b3o$135b2obo$135bobo$149b2o$148bo$145b2obo26b
2o$144bo4b3o22bo$144bo26b2obo26b2o$144b3o23bo4b3o22bo$124bo45bo26b2obo
$122bo2bo44b3o23bo4b3o48bo2bo58bo2bo$o2bo58bo2bo55bo4bo69bo27bo26bo30b
4o27bo30b4o$4bo27b4o30bo27b4o23bo4bo69b3o23b3obo24bo3bo26bo3bo26bo3bo
26bo3bo$o3bo26bo3bo26bo3bo26bo3bo24bob3o23b3o69bo4bo23b4o27bo30b4o27bo
$b4o30bo27b4o30bo26bo27bo69bo4bo55bo2bo58bo2bo$31bo2bo58bo2bo48b3o4bo
23b3o44bo2bo$148bob2o26bo45bo$148bo22b3o4bo23b3o$146b2o26bob2o26bo$
174bo22b3o4bo$172b2o26bob2o$200bo$198b2o$211bobo$210bob2o$117b3o90bo$
119bo89b2o$118bo116bo$234b2o$234bobo11$102b2o$101bobo$103bo$250b2o$
249b2o$251bo11$86b3o$88bo$87bo178bo$265b2o$265bobo11$71b2o$70bobo$72bo
$281b2o$280b2o$282bo!

[Moosey]

Is there a stong P7 domino sparker?

also, anyone who wants to finish this?

Code: Select all

x = 349, y = 128, rule = B3/S23
66bo$67b2o$66b2o$276bo$276bobo$276b2o11$81bobo$82b2o$82bo178bo$260bo$
260b3o11$97bo$98b2o$97b2o$245bo$245bobo$245b2o11$112bobo$113b2o$113bo
116bo$138b2o89bo$138bo90b3o$135b2obo$135bobo$149b2o$148bo$145b2obo26b
2o$144bo4b3o22bo$144bo26b2obo26b2o$144b3o23bo4b3o22bo$124bo45bo26b2obo
$122bo2bo44b3o23bo4b3o48bo2bo58bo2bo$o2bo58bo2bo55bo4bo69bo27bo26bo30b
4o27bo30b4o$4bo27b4o30bo27b4o23bo4bo69b3o23b3obo24bo3bo26bo3bo26bo3bo
26bo3bo$o3bo26bo3bo26bo3bo26bo3bo24bob3o23b3o69bo4bo23b4o27bo30b4o27bo
$b4o30bo27b4o30bo26bo27bo69bo4bo55bo2bo58bo2bo$31bo2bo58bo2bo48b3o4bo
23b3o44bo2bo$148bob2o26bo45bo$148bo22b3o4bo23b3o$146b2o26bob2o26bo$
174bo22b3o4bo$172b2o26bob2o$200bo$198b2o$211bobo$210bob2o$117b3o90bo$
119bo89b2o$118bo116bo$234b2o$234bobo11$102b2o$101bobo$103bo$250b2o$
249b2o$251bo11$86b3o$88bo$87bo178bo$265b2o$265bobo11$71b2o$70bobo$72bo
$281b2o$280b2o$282bo!

Immediate thought: make the wick really long and suppress most of the G outputs, then reflect some back with bumpers and bouncers and synthesize the lwsses/aim the gliders back

[dvgrn]

Immediate thought: make the wick really long and suppress most of the G outputs, then reflect some back with bumpers and bouncers and synthesize the lwsses/aim the gliders back...

Immediate thought about that immediate thought: is there known bumper and bouncer technology that works at p62? Merzenich's p31 can reflect gliders directly. Given that the gliders coming out of opposite sides of this reaction are different colors, that might actually be all that's needed to produce the five (?) input glider streams on each side. (The LWSS will need three gliders, unless there's a trick with a p31 providing a spark to reduce the cost to 2G. Anybody know offhand whether that might work?)

With such a prolific glider source at p62, a self-supporting oscillator along those lines could also be done without any periodic reflectors, only stable stuff (Snarks, semi-Snarks and semi-cenarks, and Bandersnatches). This would make a pseudo-period oscillator/gun, but it really wouldn't be particularly difficult.

[wwei23]

Immediate thought about that immediate thought: is there known bumper and bouncer technology that works at p62?

There's this from jslife (Though Snarks work at P62):

Code: Select all

x = 349, y = 128, rule = B3/S23
66bo$67b2o$66b2o$276bo$276bobo$276b2o11$81bobo$82b2o$82bo178bo$260bo$
260b3o11$97bo$98b2o$97b2o$245bo$245bobo$245b2o11$112bobo$113b2o$113bo
116bo$138b2o89bo$138bo90b3o$135b2obo$135bobo$149b2o$148bo$145b2obo26b
2o$144bo4b3o22bo$144bo26b2obo26b2o$144b3o23bo4b3o22bo$124bo45bo26b2obo
$122bo2bo44b3o23bo4b3o48bo2bo58bo2bo$o2bo58bo2bo55bo4bo69bo27bo26bo30b
4o27bo30b4o$4bo27b4o30bo27b4o23bo4bo69b3o23b3obo24bo3bo26bo3bo26bo3bo
26bo3bo$o3bo26bo3bo26bo3bo26bo3bo24bob3o23b3o69bo4bo23b4o27bo30b4o27bo
$b4o30bo27b4o30bo26bo27bo69bo4bo55bo2bo58bo2bo$31bo2bo58bo2bo48b3o4bo
23b3o44bo2bo$148bob2o26bo45bo$148bo22b3o4bo23b3o$146b2o26bob2o26bo$
174bo22b3o4bo$172b2o26bob2o$200bo$198b2o$211bobo$210bob2o$117b3o90bo$
119bo89b2o$118bo116bo$234b2o$234bobo11$102b2o$101bobo$103bo$250b2o$
249b2o$251bo11$86b3o$88bo$87bo178bo$265b2o$265bobo11$71b2o$70bobo$72bo
$281b2o$280b2o$282bo!

I would like to join the c/8 and 3c/8 spaceship searches, where can I start?

[dvgrn]

Immediate thought about that immediate thought: is there known bumper and bouncer technology that works at p62?

There's this from jslife (Though Snarks work at P62):

Code: Select all

x = 349, y = 128, rule = B3/S23...

This appears to be the exact same pattern that yujh posted, so I'm not sure why it needed to be posted again.

On the other hand, it's good to know that the source is jslife. That would have been a helpful detail for yujh to mention in the original post.

[wwei23]

This appears to be the exact same pattern that yujh posted, so I'm not sure why it needed to be posted again.

Whoops, I should have read the original message!

How should I search for P5+ oscillators in JLS? I find that there's a huge slowdown from P5+ compared to P4- when searching. By P7, it is almost impossible to search for me.

EDIT: Also, is there any collection of fuses? I'm digging up quite a bit of stuff that's not in jslife, but jslife doesn't really have much in the way of fuses.

EDIT 2: Is there such a thing as a true-period 2c/3 diagonal fuse? The signal wire doesn't count because it doesn't repeat every 2 cells, it's technically 4c/6 pseudo-period.

EDIT 3: Could a cubic sawtooth exist? It wouldn't need to rely on a cubic growth, it could be something that grew linearly to an increasing cubic limit then shrank back to the initial population.

[bubblegum]

Is there some kind of century injector like this?:

Code: Select all

x = 26, y = 23, rule = LifeHistory
6.A$7.A5.A$5.3A4.A$12.3A3$4.A.A4.B$5.2A4.B$5.A5.B$11.B$11.B$9.3B5.9F$
9.3B5.F$10.2B5.F$10.2B5.F$11.B5.F$7F4.B5.F$F16.F$F9.2D5.F$F8.D.D5.F$F
8.D7.F$F8.2D6.F!

[mniemiec]

EDIT 3: Could a cubic sawtooth exist? It wouldn't need to rely on a cubic growth, it could be something that grew linearly to an increasing cubic limit then shrank back to the initial population.

1. A cubic limit is impossible for any rules excluding B0. Assume you have some pattern P that exists within an NxN square, and runs under rule R. Regardless of what P and R are, this will necessarily be <= a solid NxN solid square under rule B12345678/S012345678. Since such a pattern must necessarily have a population <= (N+2T)^2 at time T, the growth of the upper limit is at most quadratic. A cubic limit would require a 3-dimensional rule.

2. Rules including B0+S8 are isomorphic to another rule excluding these, with living and death conditions reversed, equivalent to 1 above.

3. Rules including B0 but not S8 are equivalent to alternating rules on alternating generations, one with living and death conditions reversed, both under the same constraints as condition 1 above.

[wwei23]

I'm trying to run a local copy of Catagolue on my computer. Why isn't it committing hauls?

[Ian07]

I'm trying to run a local copy of Catagolue on my computer. Why isn't it committing hauls?

Could you please describe in more detail what you're trying to do? (e.g. what rule/symmetry, etc.)

[wwei23]

I'm noticing haul verification acting a bit weird on B3/S2/C1 in normal Catagolue (Every haul with a sufficiently rare object gets thrown away, and B3/S2/C1 is one of those rules with a few common objects and lots of rare ones, thanks to the fragility caused by only having one survival condition.). I suspect that the chi-squared test is realizing that these objects cause a statistical discrepancy, and thus further skewing census results, and worsening the damage over time. To test this hypothesis, I need to set up my own Catagolue copy to see what the console logs say about the chi-squared test results to see if that's really what's going on. Once I figure out the root cause of the issue, I can then try and code a fix and try to get it into the official Catagolue. However, I don't know where census data is stored, so I used apgsearch 1.x to submit a small haul to my local Catagolue, wait for it to commit the haul, then to turn off my local Catagolue, edit the census data to match the official Catagolue, then turn back on my local Catagolue. However, it doesn't seem to be able to commit hauls. There was also a message in the console that I think is related:

Code: Select all

[INFO] GCLOUD: SEVERE: File base storage: root directory doesn't exist

I also tried to submit a B3/S23/C1 haul to my local Catagolue to see if that would fix things, but apparently it doesn't seem to be able to commit hauls to that census either.

[Moosey]

Not exactly related to GoL but to heck with it:

How is "Goucher" to be pronounced? I've always pronounced it /gaʊt͡ʃəɹ/ (rhymes with voucher or sloucher), but have heard rumors that it's pronounced more like /guʃə/ or /guʃəɹ/ (I don't know any rhymes for this, maybe "moosher")

I've also seen that others pronounce it differently; sarp, for instance, pronounces it /gəʊt͡ʃə/, apparently

[dvgrn]

How is "Goucher" to be pronounced?

I asked that in an email a year ago June, and got this response (from calcyman):

> ... How _do_ you pronounce your last name, anyway? I've been assuming
> non-French people will rhyme it with "voucher", but I suppose there
> are some other options.

Yes, indeed -- even though it's derived from French (my father conducted
some genealogical research, and apparently our patrilineal ancestors were
French Huguenots who emigrated to England) so should originally have been
pronounced to rhyme with the French word 'boucher' (ergo the automatic
transcription error), I almost always pronounce it to rhyme with the
English word 'voucher'.

The "almost always" is interesting -- leaves a little bit of room for negotiation, I suppose.

[Moosey]

<snip>

...should originally have been
pronounced to rhyme with the French word 'boucher' (ergo the automatic
transcription error), I almost always pronounce it to rhyme with the
English word 'voucher'.

The "almost always" is interesting -- leaves a little bit of room for negotiation, I suppose.

Interesting! So, according to that email, both /gaʊt͡ʃə(ɹ)/ and/guʃe/ could be argued correct. That was a possibility I did not consider.

[wwei23]

Is it possible to build a rake gun where the gliders used to synthesize each rake come from the previous rake?

[dvgrn]

Is it possible to build a rake gun where the gliders used to synthesize each rake come from the previous rake?

Sure, there's no theoretical difficulty with that at all. Just use various reflectors to route the earlier gliders on longer signal paths, so they all get back to the construction site at the right time.

It's very easy if you can use Bandersnatches to change glider colors, and if you can just use eaters on some of the rake outputs to get a separation between lanes of at least 18hd, so that connections can be made with Snarks. Assuming that it's considered cheating to use Herschel-based stages -- that seems to be the standard arbitrary constraint recently -- then a combination of Snarks, Bandersnatches, and rectifiers allows for universal timing adjustment.