Gustavo Ramos Rehermann's patterns

[Gustavo6046]

I don't know what happened! There were two eaters!

[mniemiec]

I don't know what happened! There were two eaters!

It's usually a good idea, when posting, to first use the "Preview" button, to ensure that the post has the correct syntax (e.g. there aren't any errors in hand-crafted tags), and to also check that every pattern posted works correctly. (I used to use Golly to do this, but since the Viewer feature has been added, this is much easier to do.)

[Gustavo6046]

Just for fun I evoluted this:

Code: Select all

x = 46, y = 32, rule = LifeHistory
34.2B$33.4B$33.4B$13.2A17.6B$14.A17.7B$14.A.AB14.6B$15.2AB15.6B$17.3B
12.8B$4.7B.B4.4B12.8B$4.13B.4B10.9B$2.E26B2.9B$2.BE36B$.2B2E35B$2B2E
2B2A22B.11B$2BE3B2A36B$2.11B3.2B2.24B$2.10B11.21B$3.6B16.19B$4.3B19.
18B$4.B21.16B$3.2A19.17B$3.A18.A.2A3.B2.7B$4.A15.3AB2A7.5B$3.2A14.A4.
B9.5B$20.3A.2A9.4B$22.A.A11.B.B2A.A$22.A.A13.B2AB3A$23.A16.B4.A$39.2A
.3A$40.A.A$40.A.A$41.A!

Any transparent catalyst ptbsearch can add in the norteast huge spark to something useful?

[Gustavo6046]

I really need to ptbsearch that huge spark that appears in this pattern at northeast!

Code: Select all

x = 33, y = 32, rule = LifeHistory
21.2B$20.4B$20.4B$2A17.6B$.A17.7B$.A.AB14.8B$2.2AB15.6B$4.3B12.8B$4.
4B12.8B$5.4B10.9B$6.10B2.9B$7.5BE14B$8.2B3E14B$7.3BEBE4B.11B$8.2BE20B
$8.23B$10.21B$12.19B$13.18B$13.16B$11.17B$9.A.2A3.B2.7B$7.3AB2A7.5B$
6.A4.B9.5B$7.3A.2A9.4B$9.A.A11.B.B2A.A$9.A.A13.B2AB3A$10.A16.B4.A$26.
2A.3A$27.A.A$27.A.A$28.A!

[Gustavo6046]

H-to-block:

Code: Select all

x = 26, y = 23, rule = LifeHistory
24.2A$24.A$21.BA.A$21.B2A$19.3B$18.4B$17.4B$16.4B$9.10B$3.B4.4BD5B$2.
3B2.5B3D2B$2.3B.6BDBD3B$.13BD2B$.16B$.8B2D4B$2.7B2DB$3.9B$5.5B.B2A$4.
4B3.BA.A$2.2AB.B7.A$.A.AB9.2A$.A$2A!

[codeholic]

H-to-block:

It seems that something is wrong with your pattern.

[Gustavo6046]

Maybe if ptbsearch can put transparent catalysts in the huge spark that appears in the northeast of the Herschel eater a couple posts ago, it could be more useful.

[Gustavo6046]

Is this conduit elementary?

Code: Select all

x = 20, y = 26, rule = LifeHistory
B$2B$3B$4B$.4B$2.4B$3.4B.B$4.7B8.A$5.6B6.3A$5.2B3D2B4.A$5.2BD4B4.2A$
5.B3D4B.4B$5.11B$5.12B$5.12B$6.11B$9.7B$8.8B$7.8B$8.7B$9.5B$9.5B$7.2A
B.2B$6.A.AB$6.A$5.2A!

And this HBx317H, is it useful?

Code: Select all

x = 49, y = 65, rule = LifeHistory
23.A$23.3A$26.A$25.2AB$25.4B$26.4B.B$27.7B8.A$3.2A23.6B6.3A$2.A2.A2.
2A4.B13.2B3E2B4.A$3.2A2.A.A3.3D12.2BE4B4.2A$5.2A.B4.BDB12.B3E4B.4B$5.
A2.2B2.2B3D11.11B$2.2A.A.BA2B.5B11.12B$2.A.2A.A.A8B10.12B$6.A.A.8B11.
11B$3.2A2.A4.5B15.7B$.3A.2A4.6B14.8B5.2A$A4.B6.6B12.8B6.A$.3AB2AB3.7B
13.8B2.BA.A$3.A.2AB.8B13.10B.B2A$7.10B13.12B$7.6B2A3B12.12B$7.6B2A2B
5.2A4.2AB2.9B$7.10B5.A4.A.AB2.8B$6.11B2.BA.A4.A6.6B$6.12B.B2A4.2A5.6B
$5.15B14.5B$4.16B14.7B$.2B.16B14.8B2.2A.A$2A18B12.12BA.2A$2AB.17B11.
11B$.B.4B.8B2.4B10.5B2A4B$8.7B4.4B9.5B2A3B$9.6B5.4B8.10B$11.4B6.4B7.
9B$13.3BA5.4B6.9B$14.BA.A5.4B6.8B$15.A.A6.4B4.8B$16.A8.4B4.6B$17.3A6.
4B3.6B$19.A7.4B2.6B$28.4B.7B$29.4B.6B$30.9B$31.8B$32.8B$33.7B8.A$34.
6B6.3A$34.7B4.A$34.7B4.2A$34.8B.4B$34.11B$34.12B$34.12B$35.11B$38.7B$
37.8B$36.8B$37.7B$38.5B$38.5B$36.2AB.2B$35.A.AB$35.A$34.2A!

[dvgrn]

Is this conduit elementary?...
And this HBx317H, is it useful?

Not really, and no.

You could perfectly well answer these questions yourself, but you're still not paying attention.

The first pattern is yet another Herschel eater. The glider it emits is the same old SW glider on lane -2 that you've "discovered" several times before. Herschels produce that glider naturally.

A catgl or ptbsearch search will find hundreds or thousands of ways to cleanly kill the Herschel, but there's no reason to keep track of the big complicated ones when we can just use the small fast ones (a beehive, or two blocks, or one of the extra-compact versions that you can find on the new gun collection thread.)

The second conduit is not elementary. It's an F117 conduit plus a syringe. It doesn't make any sense to keep track of it as a separate conduit. If we did that, we'd also have to track Bx313 and Bx321 and Bx325 and Bx329 and Bx333 and so on -- not to mention a similarly infinite number of right-turn conduits using a mirror-image syringe, and an infinite series starting from an Fx77 and every other known conduit.

If there's anything about this that you don't understand, please ask specific questions until you do understand. It's kind of silly of you to keep posting patterns and questions like these, without seeming to learn anything from people's responses.

[Gustavo6046]

I didn't expect a bakery here:

Code: Select all

x = 18, y = 9, rule = LifeHistory
10.2A$10.2A2$.2A$A2.A6.2A$.A.A6.2A$2.A12.2A$15.A.A$15.A!

Anyway which is the bakery predecessor?

[Freywa]

Asking too many questions, eh? The bakery and half-bakery are themselves quite common in soups.

[Gustavo6046]

I have faith!

Code: Select all

x = 10, y = 12, rule = LifeHistory
.2A$.2A2$.2A$A2.A$A2.A$.2A$6.2B$.2A2.B.2B$.2A3.B3A$7.A2B$8.A!

[dvgrn]

New interests come! I am planning a G-to-pi to complete all! I already have the original reaction:

Code: Select all

x = 7, y = 4, rule = LifeHistory
3D$D2A2.2A$2DCA.2A$.A!

Also souldn't we be searcing for reflectors?

Code: Select all

x = 14, y = 9, rule = LifeHistory
9.A$8.A.A.2A$8.A.A.2A$9.A3$2A$.2A$A!

@Gustavo, you're continuing to post ideas that don't make sense, to threads where you've been asked repeatedly not to post. Please don't do that. You'll make much better contributions on other threads, after you've shown here on this thread that you can tell the difference between sense and nonsense.

You've certainly learned quite a bit about building Life signal circuitry in the last few weeks. Unfortunately there's no sign yet that you're paying any attention to people's responses to your postings.

That means it's probably time for another warning. Until further notice, if you post to any active thread besides this one, and if it's as badly thought out, un-researched, and irrelevant as the message quoted above, then I'll put you back on the banlist for a week.

I won't bother with further warnings. Next time it will be two weeks, then three, and so on. A new thread isn't necessarily as much of an interruption, but please be careful there also. A new topic with content like the above might also get you banned for a week.

It should be easy enough for you to limit yourself to this thread for now. Your ideas have been getting plenty of attention here. I'm sorry, but your current method posting dozens of ideas, one after another, without thinking about them first, is not really compatible with the way everyone else uses these forums.

Once again:

If there's anything about this that you don't understand, please ask specific questions [here on this thread] until you do understand. It's kind of silly of you to keep posting patterns and questions like these, without seeming to learn anything from people's responses.

[Gustavo6046]

I know most of one-still life glider reactions are already know, but does that includes crashes with mangos?

Code: Select all

x = 12, y = 13, rule = LifeHistory
2.A$.A.A$A2.A$A.A$.A6$9.2A$9.A.A$9.A!

[dvgrn]

I know most of one-still life glider reactions are already know, but does that includes crashes with mangos?

Code: Select all

x = 12, y = 13, rule = LifeHistory
2.A$.A.A$A2.A$A.A$.A6$9.2A$9.A.A$9.A!

Heinrich Koenig has tried every possible way for a glider to hit every possible stable object with up to 16 ON cells. I believe that work was finished sometime in 2007.

Mangoes have only 8 ON cells, so you can be very sure that they were explored a long time before that.

Now, it's certainly possible that everyone missed something interesting in some of those collisions. For example, maybe if you wait until your glider+mango explosion settles, and you hit it with one more glider from the same direction, it will turn back into a mango in the exact same location.

That would certainly be very interesting... but unfortunately it's also very very very very unlikely!

So there doesn't seem to be much point in posting random single object+glider collisions that are 100% certain to have already been tested many years ago -- until you actually find a lucky coincidence like that, and can point it out and explain clearly why it's interesting.

Koenig did his searches by using a program to enumerate all the possible N-bit still lifes, and then another program to try all possible glider collisions with each. It would take you several thousand years to do that much testing, experimenting by hand as you've been doing so far -- and even that would be just to duplicate work that's been done before, much more reliably.

[BlinkerSpawn]

Gustavo, if you'd like to work with conduit-searching, is there any chance you could run a search program to attempt to restore the loaf?

Code: Select all

x = 16, y = 6, rule = LifeHistory
2.C$.C.C$C2.C$.2C6.C4.2C$7.3C4.2C$8.C!

...let's just say you've got a lot of room to work here; almost a 25 by 30 rectangle to the left of the R.

[codeholic]

Why a loaf and why at this exact place?

[Scorbie]

@Blinkerspawn Is that part of a known conduit? If not, I think it would be one of the not-so-likely-but-you-never-know type of searches...

From my experience with ptbsearch, it usually takes 3 or 4 catalysts to restore a block or a hive. If there aren't any solutions in that range, new solutions don't tend to appear even if you add more catalysts. (Well, maybe 5 or 6 for something big and open like a pi explosion.)
I think this happens because the space there is kinda limited.
(Ptbsearch experts, please tell me if I am wrong about my observations!)

If you look at your reaction the catalysts could only approach on one side, so I am not expecting anything over four catalysts. However, it is pretty unlikely to restore a damaged loaf with <= 3 catalysts, I think. (Edit: make that four, the explosion looks not bad, but I doubt it would be a clean one even a 4-cat-solution exists.)

But note that these are from my non-general experience and there might be a solution.

By the way, why did you put the loaf over there? It is kinda close to the input R... Why not a block or a hive farther away?

[dvgrn]

By the way, why did you put the loaf over there? It is kinda close to the input R... Why not a block or a hive farther away?

Yes, between the unlikeliness of the loaf restoration, and the closeness of the block, I think this isn't exactly a good first conduit search to attempt.

There's a first step that I'd strongly recommend for every search: make sure you know where your input is coming from.

Ideally there should be multiple known conduits in the elementary-conduits collection that can place an input -- in this case an R-pentomino -- in the starting location, without getting in the way of any of the objects that you're planning to place as catalysts.

In this case I can't find even one conduit that will work. Maybe there's one hiding somewhere... but as a general rule, for an R input you need to leave the conduit open either above or below the R. There are R-output conduits that bring in the R from one side or another, but none that make it magically pop into existence in the middle of a constellation like this.

So pick a likely looking R-output conduit, and make sure that your experimental placements of catalysts don't get in the way of that conduit's reaction envelope.

If you don't do this, then once you're done solving your new conduit, you have to invent another new conduit that can sneak the R into such a tight space. That theoretical new conduit would actually be enormously more useful than this one, in any case, so you might as well save yourself a lot of time and look for it first!

[Gustavo6046]

But I'm sure Koenigsegg still didn't saw e.g. this collision:

Code: Select all

x = 20, y = 12, rule = LifeHistory
9.A$8.A.A6.A.A$2A3.A2.A2.A5.2A$A.A.A.A.3A7.A$2.A.A.A$.2A.A.3A$5.A3.A$
6.3A2$6.3A$5.A2.A$5.2A!

I also just made a script that allows getting the population of the selection:

Code: Select all

import golly as g

def getSquarePop(x, y, width, height):
	selcells = len(g.getcells([x, y, width, height]))
	stts = g.numstates()
	if stts < 3:
		return selcells / 2
	else:
		return selcells / 3

g.show('Selection population is: ' + str(getSquarePop(g.getselrect()[0], g.getselrect()[1], g.getselrect()[2], g.getselrect()[3])))

[dvgrn]

But I'm sure Koenigsegg still didn't saw e.g. this collision...

Absolutely right. There are on the rough order of 100 billion 36-bit still lifes more or less interchangeable with that one -- in the sense that if you hit them with a glider, you'll generally make a random mess that's never been seen before. So if you pick a random one to post here every minute, you'll have tried them all in less than 200,000 years, by which time you'll have seen a lot of random messes.

That's just one glider per 36-bit still life, though. For a complete survey you should really try all of the 100 or so valid collision lanes for each still life. But that would take 20 million years to finish... Basically, anything you can find by experimenting manually like this, you can almost certainly find a lot faster by writing a program to find it for you.

I also just made a script that allows getting the population of the selection...

Looks good! The only trick I'd suggest is to unpack the selection-rectangle list to get the four numbers out, instead of retrieving the selection four times:

Code: Select all

g.show('Selection population is: ' + str(getSquarePop(*g.getselrect())))

Or at least store g.getselrect() in temporary variable.

[Gustavo6046]

I don't know if it is interesting, but two gliders convert a boat into a block with the help of a catalyst:

Code: Select all

x = 10, y = 16, rule = LifeHistory
5.A$5.A.A$5.2A3$8.2A$8.2A$.2A$A.A$.A4$7.A$6.2A$6.A.A!

[mniemiec]

I don't know if it is interesting, but two gliders convert a boat into a block with the help of a catalyst: ...

Considering that there are several ways for two gliders to convert (nothing) into a block, without the need for a catalyst either, this isn't really interesting at all. There are doubtless countless ways of adding sacrificial bait objects and useless catalysts to increase the complexity of an object synthesis, but almost none of them are interesting. Some MIGHT occasionally be interesting, if they produce the result in an unusual way, or an unusual place (e.g. the recent soup-inspired synthesis of a Phoenix, which is much more expensive than traditional syntheses, but which also emits it almost on an edge), but for simple objects like blocks, there are many simpler and cheaper ways to do it.

[Gustavo6046]

Maybe interesting, this produces three gliders in different directions and consists of only a block and a bakery:

Code: Select all

x = 18, y = 20, rule = LifeHistory
.A$A.A$A2.A$.2A.A$3.A.A$3.A2.A$4.2A$9.2A$9.2A9$15.2E$15.E.E$15.E!

[M. I. Wright]

Are you just not reading? At all? Scroll up a few posts to find out why that isn't interesting in the slightest.