Finally trying out stable Herschel tracks...

For discussion of specific patterns or specific families of patterns, both newly-discovered and well-known.
Awesomeness
Posts: 126
Joined: April 5th, 2009, 7:30 am

Finally trying out stable Herschel tracks...

Post by Awesomeness » July 9th, 2009, 11:10 am

I've been playing around with Life for a while, and I've made lots of interesting patterns, but I've never gotten into Herschel tracks. I've decided to try because of the computation and complicated things you can create with them. After lots of searching for non-".lif" resources for Herschel tracks, (Golly cant open ".lif" file extensions.) I've played around and made a few things.

Here's my first try:

Code: Select all

x = 318, y = 226, rule = B3/S23
159bo2$156b2o$157bo$157bobo$158b2o2$317bo$315b3o$289b2o23bo$290bo23b2o
$290bobo$291b2o3$130bo$130b3o$133bo$132b2o2$300b3o$270b2o7b2o19bo$269b
obo7b2o18b3o$149b2o116b3obobo$150bo115bo5b2o$150bobo113b2o$151b2o3$
122bo190b2o$122bobo29bo103bo54b2o$122b3o29bo103b3o29bo$124bo29b3o103bo
29b3o$156bo103bo31bo$292bo15b2o$308b2o$312b2o$262b2o48b2o$147b2o113bob
o$141b2o5bo115bo$141bobob3o116b2o40b2o$134b2o7bobo160b2o$134b2o7b2o2$
300b2o$107b2o172b2o17bobo$108bo172bo20bo$105b3o174b3o17b2o$105bo178bo
12$257b2o$257bobo$259bo$12b2o245b2o$13bo18b2o$13bobo16b2o5b2o133bo82bo
$14b2o23b2o$171b2o$172bo$8b2o27b2o133bobo$8b2o27b2o134b2o$43b2o$43b2o$
2b2o$2b2o$6b2o42bo$6b2o15bo24b3o$23bobo21bo$23b3o21b2o$25bo$b2o$b2o
161bo$162b3o$138bo22bo$138b3o20b2o$141bo$35b3o102b2o$35bo$15b2o17b2o
85bo$15bo105b3o$13b3o108bo$123b2o11b2o$136b2o3$48b2o$48b2o$25bo$2b2o
21b3o94bo49bo$3bo23bo94bobo47bo$3o24bo15b2o77b3o36b2o9b3o$o42b2o79bo
23b2o11b2o11bo$47b2o100bo$47b2o97b3o$6b2o138bo$6b2o$12b2o27b2o72b2o$
12b2o27b2o73bo$113b3o20b2o$113bo22bo$10b2o23b2o100b3o17bo3b2o$10b2o5b
2o16bobo101bo22bo$17b2o18bo121b3o$37b2o120bo27$78bo$59b2o17b3o$60bo20b
o$60bobo17b2o29b2o$61b2o48b2o5b2o$118b2o2$55b2o$55b2o40b2o17b2o$98bo
17b2o$98bobo21b2o$49b2o48b2o21b2o$49b2o$53b2o$53b2o15bo58bo$70bo31bo
24b3o$70b3o29bobo21bo$72bo29b3o21b2o$48b2o54bo$48b2o4$95b2o$89b2o5bo$
89bobob3o18b3o$62b2o18b2o7bobo20bo$62bo19b2o7b2o20b2o$60b3o2$55b2o$56b
o$53b3o$53bo75b2o$68b2o59bo$68bobo56bobo$70bo56b2o$70b2o31b2o$72bo30b
2o$70b3o$69bo$69b2o3$105b2o$106bo$103b3o$71b2o30bo$71b2o31b2o$47b2o56b
o$46bobo56bobo$46bo59b2o$45b2o75bo$120b3o$119bo$119b2o2$113b3o$61b2o
20b2o7b2o19bo$61bo20bobo7b2o18b2o$59b3o18b3obobo$79bo5b2o$79b2o4$126b
2o$71bo54b2o$48b2o21b3o29bo$49bo23bo29b3o$46b3o24bo31bo$46bo58bo15b2o$
121b2o$125b2o$52b2o21b2o48b2o$52b2o21bobo$58b2o17bo$58b2o17b2o40b2o$
119b2o2$56b2o$56b2o5b2o48b2o$63b2o29b2o17bobo$94bo20bo$95b3o17b2o$97bo
!
Its just 2 different loops and 3 experimental conduits. (At least I think they're called that) The eaters with dots near them are not part of the conduit, just to make sure gliders don't get away.

I still don't quite understand how people make stuff like glider-to-LWSS converters and patterns that do cool things with this.

User avatar
PM 2Ring
Posts: 152
Joined: March 26th, 2009, 11:18 am

Re: Finally trying out stable Herschel tracks...

Post by PM 2Ring » July 9th, 2009, 11:51 am

(Golly cant open ".lif" file extensions.)
Depending on the version, Golly may be able to open it. There's a program that comes with xlife called lifeconv (or something like that) that can convert many older .lif files into Life 1.05 format, which looks like this:

Code: Select all

#Life 1.05
#D Acorn
#D The most vigorously growing 7-cell
#D "methuselah" pattern.  See also RABBITS.
#N
#P -3 -1
.*
...*
**..***
Golly can easily cope with that sort of file, although it'll ignore the (-3, -1) pattern origin info. Some .lif files use a slightly different system, but can be fixed by hand, or a simple script. I managed to convert most of the Life32 pattern collection. Some .lif files are more complex, and define patterns in terms of sub-components, some of which may be defined in other files. The conversion program doesn't work properly on these. :(

User avatar
Extrementhusiast
Posts: 1966
Joined: June 16th, 2009, 11:24 pm
Location: USA

Re: Finally trying out stable Herschel tracks...

Post by Extrementhusiast » July 11th, 2009, 10:04 pm

Is it possible to finish this? I got so far...only for it to fail in the end.

Code: Select all

x=23,y=18,rule=B3/S23
2o$2o10$6b3o7b2obo$7bo8b2ob3o$7b3o12bo$16b2ob3o$17bobo$17bobo$18bo!
Last edited by Extrementhusiast on July 16th, 2009, 7:38 pm, edited 1 time in total.
I Like My Heisenburps! (and others)

Sokwe
Moderator
Posts: 2645
Joined: July 9th, 2009, 2:44 pm

Re: Finally trying out stable Herschel tracks...

Post by Sokwe » July 12th, 2009, 2:20 am

Extrementhusiast wrote:Is it possible to finish this? I got so far...only for it to fail in the end.

Code: Select all

x = 23, y = 18, rule = B3/S23
2o$2o10$6b3o7b2obo$7bo8b2ob3o$7b3o12bo$16b2ob3o$17bobo$17bobo$18bo!
What you have there is a Herschel-to-R converter that releases a glider. You could try manipulating the R-pentomino in some known way to whatever effect.
-Matthias Merzenich

User avatar
dvgrn
Moderator
Posts: 10612
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Finally trying out stable Herschel tracks...

Post by dvgrn » July 12th, 2009, 6:38 am

Extrementhusiast wrote:Is it possible to finish this? I got so far...only for it to fail in the end.
It might be possible to get a Herschel-to-glider or Herschel-to-junk converter out of this, by adding more catalysts in the twenty-odd ticks between the time when the R-pentomino was created and when the R's products destroy the eater. But I'm not too optimistic: there are very few places to put new catalysts where the active pattern wouldn't already have destroyed them.
Sokwe wrote:What you have there is a Herschel-to-R converter that releases a glider. You could try manipulating the R-pentomino in some known way to whatever effect.
The biggest problem is that the R-pentomino shows up in the wrong orientation, travelling back toward the block, so none of the known conversions are going to work -- you'd have to magically shoehorn something in there _after_ the block has done its initial catalyzing job.

The eater catalyst is in a known position, and it turns out to be possible to substitute a regular fishhook eater for the eater2. Here's a comparison with one of the variants of the Fx77 conduit -- run this step by step to see where the differences show up in the reaction:

Code: Select all

x = 32, y = 70, rule = B3/S23
20bo$20b3o$23bo$22b2o2$3bo$3b3o$6bo$5b2o11b2o11bo$18b2o9b3o$29bo$29bo
4$4bo$4bobo$4b3o$6bo$24b2o$21b2o2bo2b2o$21b2obo3bo$24bobobo$21b2obob2o
$15b2o4bo2bo$15b2o6b2o22$4bo$3bobo$3bobo$b3ob2o$o$b3ob2o$3bob2o7$4bo$
4bobo$4b3o$6bo5$16b2o$16b2o!
So the block catalyst is in a slightly different location, but that's partly because in the Fx77 there's an additional "transparent block" catalyst that modifies the reaction very significantly, before the block does its work.

I'm going to add a new topic describing the tools that are available for working on this kind of problem. As folks have mentioned elsewhere, it's a little odd that only one new Herschel conduit has been discovered this century, after the Cambrian explosion of new inventions in the mid-nineties.

Apparently people stopped spending the necessary search time, because the known conduits got to be enough to make a "universal" toolkit that can build pretty much anything (given enough space and time)... as Brice Due's eaters-only F171 shows, it's not as if there are no more conduits waiting to be discovered! Quite likely there are dozens more stable conduits out there, and hundreds of periodic ones.

knightlife
Posts: 566
Joined: May 31st, 2009, 12:08 am

Re: Finally trying out stable Herschel tracks...

Post by knightlife » July 12th, 2009, 6:44 pm

Extrementhusiast wrote:
Is it possible to finish this? I got so far...only for it to fail in the end.
I tried it out and ended up with a partial H-to-G that creates an extra block.

Image

Using a different approach I found a partial H-to-G that deletes a block.
Both are shown here:

Code: Select all

x = 77, y = 16, rule = B3/S23
2o$2o$64b2o$64b2o$21b2o$21bobo$23bo$23b2o4$6b3o7b2o45b3o$7bo8bobo45bo
8b2o$7b3o8bo45b3o6bo$18b2o54b3o$76bo!
Since they complement each other I combined them like this, using F171 to delete the block:

Code: Select all

x = 162, y = 65, rule = B3/S23
26b2o$27bo8b3o$27bobo8bo$28b2o7b3o$20bo$14b2o4b3o$15bo7bo$15bobo4b2o$
16b2o25b2o$43bo$10b2o29bobo$11bo29b2o$11bobo$12b2o15$9bo99bo$9b3o97b3o
20b2o$12bo6b3o90bo19bo18bo$11b2o8bo89b2o17bobo16b3o$21b2o107b2o16bo$
148b2o$156bo$154b3o$153bo$153b2o2$131bo$2o98b2o29bo24b2o$2o98b2o27b3o
24bo$129bo24bobo$154b2o$21b2o98b2o$21bobo97bobo$23bo99bo37bo$23b2o98b
2o34b3o$158bo$158b2o2$6b3o7b2o88b3o7b2o$7bo8bobo88bo8bobo$7b3o8bo88b3o
8bo$18b2o98b2o2$160bo$158bobo$158b3o$158bo2$150b2o$150bo$151b3o$153bo!
The F171 is rotated 90 degrees clockwise in the pattern on the right.
It just happens that the two converters don't interfere with each other's still life in either case.
Of course, the limitation is that the block must be formed before it can be deleted, as shown.
The F171 conduit was added just as an example, to allow time for the block to be created.

There is a very simple stable H-to-G converter, however, that is commonly used:

Image

Code: Select all

x = 8, y = 13, rule = B3/S23
5b2o$4bo2bo$5b2o8$3o$bo$b3o!

Awesomeness
Posts: 126
Joined: April 5th, 2009, 7:30 am

Re: Finally trying out stable Herschel tracks...

Post by Awesomeness » July 12th, 2009, 9:56 pm

knightlife wrote: I tried it out and ended up with a partial H-to-G that creates an extra block.
I've just found one with 30 seconds of tinkering which deletes a block. Another almost-H-to-G converter.

Code: Select all

x = 23, y = 18, rule = B3/S23
2o19b2o$2o19b2o10$6b3o7b2obo$7bo8b2ob3o$7b3o12bo$16b2ob3o$17bobo$17bob
o$18bo!

User avatar
calcyman
Moderator
Posts: 2932
Joined: June 1st, 2009, 4:32 pm

Re: Finally trying out stable Herschel tracks...

Post by calcyman » July 13th, 2009, 2:51 am

Another almost-H-to-G converter.

However, there is not much point stabilising these partial H-to-G devices, since they do not achieve a different path or timing to the beehive termination.

If you manage to emit a glider on a different path (to the 25 or so already discovered), it might prove very useful in completing stable reflectors.
What do you do with ill crystallographers? Take them to the mono-clinic!

User avatar
dvgrn
Moderator
Posts: 10612
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Finally trying out stable Herschel tracks...

Post by dvgrn » July 16th, 2009, 7:11 am

PM 2Ring wrote:
(Golly cant open ".lif" file extensions.)
Depending on the version, Golly may be able to open it... Some .lif files are more complex, and define patterns in terms of sub-components, some of which may be defined in other files. The conversion program doesn't work properly on these. :(
Once upon a time I wrote up some notes on how to semi-automatically convert these Xlife #I-format files into Golly scripts.

Unfortunately it's not altogether easy, and for the small volume of #I-format files that seem to be in existence, writing a converter utility wasn't quite worth the effort. Golly scripts definitely seem to be the way to do this kind of structured pattern representation nowadays.

User avatar
Extrementhusiast
Posts: 1966
Joined: June 16th, 2009, 11:24 pm
Location: USA

Re: Finally trying out stable Herschel tracks...

Post by Extrementhusiast » July 16th, 2009, 7:18 pm

YAY! It's nice to see that people are helping me with my problems!
I Like My Heisenburps! (and others)

Awesomeness
Posts: 126
Joined: April 5th, 2009, 7:30 am

Re: Finally trying out stable Herschel tracks...

Post by Awesomeness » July 26th, 2009, 1:23 pm

Awesomeness wrote:I still don't quite understand how people make stuff like glider-to-LWSS converters and patterns that do cool things with this.
Yeah... My question remains. It seems this thread has changed topic to partial H-to-G converters. Or has this question been answered, and I can't find it in the thread, buried inside the other conversation that's been going on?

knightlife
Posts: 566
Joined: May 31st, 2009, 12:08 am

Re: Finally trying out stable Herschel tracks...

Post by knightlife » August 1st, 2009, 5:05 am

Awesomeness wrote:
Awesomeness wrote:I still don't quite understand how people make stuff like glider-to-LWSS converters and patterns that do cool things with this.
Yeah... My question remains. It seems this thread has changed topic to partial H-to-G converters. Or has this question been answered, and I can't find it in the thread, buried inside the other conversation that's been going on?
You need a good imagination, a very good memory for patterns you have seen before and where to find them, and last but not least, perseverance. What you have is a construction kit that allows you to engineer new ideas from existing pieces. Even if you have a very good idea that is superficially plausible, you will run into snags where the details don't let you build what you intended. But with perseverance you can finish the construction. Herschel tracks let you create glider salvos that can then synthesize spaceships and objects, so I would say glider synthesis is a must for constructing "cool things". Typically the patterns that are interesting have a repeating theme (oscillators and loops) and make use of converters (like H-to-G and G-to-H converters) or grow infinitely (guns) but not necessarily: that is where the good imagination comes in.

I can only guess how many snags there were and how much perseverance it took to overcome them when constructing the giant caterpillar spaceship. I think its size indicates the large number of snags encountered, with each solution requiring more pi crawlers and increasing complexity.

This has been an attempt to answer your question. Not sure it has.

User avatar
dvgrn
Moderator
Posts: 10612
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Finally trying out stable Herschel tracks...

Post by dvgrn » August 1st, 2009, 11:18 am

Awesomeness wrote:I still don't quite understand how people make stuff like glider-to-LWSS converters and patterns that do cool things with this.
This question has a couple of possible interpretations:

1) How do people go about discovering new conduits -- i.e., conduits that make use of reactions that were not previously known, such as a new stable conduit that converts a glider, or maybe a Herschel, directly into an LWSS?

2) How do people go about stringing together known conduits to do new things -- e.g., accept a glider input, produce an output LWSS, and fix everything up to be ready for another glider input?

#1 is something I've been sort of planning to write about sometime -- some kind of "best practices" document for the 'ptbsearch' and 'catalyst' utilities would be good to have, among other things -- but it's probably going to take a while to get a good summary together. It's really easy to spend a lot of time trying to fix the unfixable, following likely dead ends that other people have already tried -- but there are also huge areas of stable-conduit search space, easily within reach of today's computers and search programs, that haven't even been looked at yet.

I tried to address #2 in a separate topic last month -- that was a first swipe at an introductory explanation of Hersrch, the most powerful search tool currently available to help put together synchronized Herschel circuitry. Let me know where this needs more (or better) explanation, and I'll see what I can do...

Known stable Herschel technology is versatile enough to construct any pattern that can be produced by colliding gliders. That is, if you have a glider recipe to construct a given object, then with Hersrch's help you could build a contraption that takes (say) an input glider as a trigger, produces that object as output, and returns to its original state afterwards, ready to produce another object at the same location... or if the object is stationary, at the next location along a line, or whatever.

The circuitry may take a lot of time to recover, and use up a lot of space, but it can always be done somehow. If anyone thinks they have a counterexample, I'll be happy to try to shoot it down...!

knightlife
Posts: 566
Joined: May 31st, 2009, 12:08 am

Re: Finally trying out stable Herschel tracks...

Post by knightlife » August 1st, 2009, 5:57 pm

dvgrn wrote:Known stable Herschel technology is versatile enough to construct any pattern that can be produced by colliding gliders.
What if the pattern can only be constructed by using tightly packed salvos that are not themselves constructible? Has it been proven that any glider salvo is constructible? I suppose there is usually a different way to get the same result.

I would like to see a herschel track lay down tracks in front of itself so it can continue to lay down more new tracks indefinitely. It would be extremely slow but I think that one could be done.

User avatar
dvgrn
Moderator
Posts: 10612
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Finally trying out stable Herschel tracks...

Post by dvgrn » August 1st, 2009, 6:58 pm

knightlife wrote:
dvgrn wrote:Known stable Herschel technology is versatile enough to construct any pattern that can be produced by colliding gliders.
What if the pattern can only be constructed by using tightly packed salvos that are not themselves constructible? Has it been proven that any glider salvo is constructible? I suppose there is usually a different way to get the same result.
That's what I'm counting on, in making my overconfident conjecture that Herschel Conduits Can Do Anything... certainly it's easy enough to produce gliders following each other at the minimum possible spacing -- just use a variant of the reactions in the pseudo-p14 gun, for example.

But no, I haven't formally proved by induction that for any patch of N gliders, it's always possible to find at least one that can be added to the rest of the patch using Herschel conduits. Might not be hard to do, though -- the best trick might be to find a small constellation that can be constructed way out in front of an oncoming fleet, that can be hit by a glider or *WSS from the side at the last moment to produce an edge-of-envelope output glider to add to the front of the fleet.

Let's see, a couple of the turners in Golly's Syntheses/blockish-and-blockic-seeds.rle work fine at p15:

Code: Select all

x = 37, y = 23, rule = B3/S23
21b2ob2o$21b2ob2o$16b2o$16b2o5$34b3o$34bo$35bo10$2o$b2o$o!
For that matter, a simple kickback reaction could produce the same spacing. So that just leaves p14, and maybe a few slightly worrisome cases with gliders right next to each other -- I bet something can be found to handle each possible case. In which case I wouldn't even have to use the different-way-to-get-the-same-result defense (though I bet that's also valid!)
knightlife wrote:I would like to see a herschel track lay down tracks in front of itself so it can continue to lay down more new tracks indefinitely. It would be extremely slow but I think that one could be done.
Couldn't be slower than a Herschel moving through the circuit, could it? Or are you thinking of using multiple Herschels somehow?

You need at least a couple of still-lifes per conduit to get a glider out of a Herschel, and at least a couple of gliders per still-life to build more conduits at the front. So I don't immediately see how you'd generate enough gliders to keep the thing going, no matter how many parallel tracks you had. And among other problems, it's also surprisingly difficult to get a forward glider out of a constructible Herschel circuit. Otherwise we might have a Caterpillar spaceship by this time that uses Herschel tracks, instead of pi heptominoes climbing on blinker trails.

The other good thing about blinker trails was that they could be moved to let glider streams pass through them -- Herschel conduits are much less forgiving, though they are permeable here and there...

knightlife
Posts: 566
Joined: May 31st, 2009, 12:08 am

Re: Finally trying out stable Herschel tracks...

Post by knightlife » August 1st, 2009, 10:27 pm

dvgrn wrote:Couldn't be slower than a Herschel moving through the circuit, could it?
If the Herschel has to keep going straight, I thought, then that is true. Parallel Herschel circuits would just make things worse. Then I thought a track could be built out in front using parallel circuits if there was enough time before needing to use it. Of course the new track needs the parallel circuits built also so that the sequence can be repeated. The delay, I thought, could be a glider that goes to a distant 180 degree reflector and back before converting back to a Herschel and continuing on its track. In other words, build a specialized replicator triggered by a spaceship that builds another one in front of itself, finally triggering the new one with the same type of spaceship. This is cheating because the tracks would be discontinuous but it would certainly be a new kind of puffer! There is the problem of building the reflector for the delay, but maybe if it were placed behind rather than to the side then the same one could be used over and over. The delay would keep getting longer and longer as new tracks are laid in front.

You are right that there aren't enough gliders. A Herschel track does not make enough gliders to build itself. That is where this breaks down. Another source of gliders would be needed such as Herscel loops, guns or rakes launched to the side. Block laying puffers could create corridors of blocks on either side of the track that consist of those cool one-time glider replicators made of blocks. I know, this is getting far fetched.

All this makes me wonder if a Herschel can merge into an existing Herschel loop for delay purposes.

User avatar
dvgrn
Moderator
Posts: 10612
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Finally trying out stable Herschel tracks...

Post by dvgrn » August 2nd, 2009, 1:01 am

knightlife wrote:You are right that there aren't enough gliders. A Herschel track does not make enough gliders to build itself. That is where this breaks down. Another source of gliders would be needed such as Herschel loops, guns or rakes launched to the side.
Once you allow extra rakes and such off to the side, you're kind of back in known territory, though. Have you seen Puffers/p100-H-track-puffer.rle in Golly's pattern collection? There was also another rake-based track-builder breeder pattern where Herschels were continuously created at the front of the track, and travelled to the back. Could dig that up if you're interested and can't locate it.
knightlife wrote:All this makes me wonder if a Herschel can merge into an existing Herschel loop for delay purposes.
Technically, no, not in while the signal is in Herschel form anyway: the catalysts that keep a Herschel from misbehaving when it's coming in from one direction are never far enough apart to allow another Herschel to come in from another direction and wind up in the same place. Not in currently known conduits, anyway.

But there are lots of workarounds... most of which involve gliders. Two parallel gliders separated by 2, 5 or 6 cells can be caught by a Herschel receiver and turned into a Herschel, quickly and easily as these things go. A receiver can also catch just about any pair of gliders on crossing paths, often in several different ways.

So suppose that a set of these "tandem gliders" is produced by edge-shooting H-to-G circuits or permeable conduits -- e.g., the Fx77 and F166 let a glider pass through on the same path as their standard output glider, and several others can be made to work as well. Another set of tandem gliders farther upstream can be created with standard Herschel conduits... and there's your merge circuit.

But if repeat time isn't an issue, the simplest trick is to convert the Herschel signal to a glider, and merge glider streams instead and convert back to a Herschel later. For example, here's a startable/stoppable p1024 Herschel loop that I put together for Calcyman a year ago, when he was working on his stable non-totalistic unit-cell metapixel project (!)

Code: Select all

#C P1024 Herschel loop gun w/ glider merge input and boat-bit stopper
#C Send another glider in at the bottom to cleanly shut down the loop
x = 118, y = 114, rule = B3/S23
18bo$19bo$17b3o3$15b2o$15b2o6$84bo$84b3o$87bo23b2o$2o84b2o23bo$2o107bo
bo$67bo37b2o2b2o$41bo9bo15b3o35b2o$41b3o5b3o18bo$44bo3bo20b2o11b2o$43b
2o3b2o32b2o5$20b2o$20bo$21b3o$23bo33b2o52b2o$57b2o34b2o16bobo$45b2o45b
obo18bo$44bo2bo44bo20b2o$4b2o33b2o4b2o44b2o4b2o$3bobo32bobo54bo2bo$3bo
34bo56b2o$2b2o33b2o65b2o$47b2o33b2o20b2o$47bo34bo$48b3o32b3o$50bo34bo
3$12b2o39b2o$12b2o40bo59b2o$51b3o60bo$19bob2o28bo60bobo$19b2obo89b2o4$
2b2o$2bobo89b2o$4bo90bo$4b2o89bobo$96b2o3$23b2o$23bo$21bobo$4b2o15b2o$
4b2o3$21b2o$21bobo92b2o$23bo73b2o17b2o$23b2o72b2o6$5b2o$6bo$3b3o$3bo2$
73bo$37b2o34b3o$21b2o15bo10bo26bo$21bobo13bo9b3o25b2o$23bo9bo3b2o7bo$
23b3o5b3o12b2o$26bo3bo$25b2o3b2o6b3o25bo$38bo25b3o44b2o$39bo23bo47b2o$
50b2o11b2o$50b2o$13b2o$13b2o91b2o$106b2o$4b2obo102b2o$4bob2o102b2o3$
75b2o27b2o$75b2o11b2o14b2o$10b2o76bo$9bobo77b3o$9bo81bo$8b2o26b2o$36bo
34b2o$30b2obo3b3o9b2o20bo$30b2ob4o2bo8bobo21b3o$36bo11bo25bo$30b2ob3o
11b2o$29bo2b2o$29b2o$63b3o$63bo$64bo!
A little farther out on the lunatic fringe, here's a merge circuit using two synchronized gliders that sneak right into a standard block-snake catalyst in a Herschel conduit, and add another Herschel on the fly:

Code: Select all

#C non-optimal pseudo-P39 gun
#C Includes an unusual 2-glider Herschel merge,
#C and also a standard H-to-G merge circuit
x = 252, y = 163, rule = B3/S23
151b2o$151b2o3$153b2o$153b2o2b2o3b2o$157b2o2bo2bo$143bo18b2o$137bo6bo$
136bobo4bo12bo$136bobo16bobo$137bo16b2obo$154bobo$152b3o$154b3o9b2o$
155b2o4b2o3b2o$157b2obo2bo$157b2o2b2o$158b4o5b2o$159bo7b2o$131b2o26bo
11b2o$131b2o11bo26b2o$135b2o7bo$135b2o5b4o$141b2o2b2o$140bo2bob2o$99b
2o35b2o3b2o4b2o$99b2o35b2o9b3o$149b3o$147bobo$146bob2o14b2o$100b2o8b2o
34bobo15bo$100b2o7bo2bo34bo17b3o$99bo10b2o55bo$85bo14b2o38b2o$84bobo
13b2o37bo2bo2b2o15bobo$84bobo53b2o3b2o2b2o12b2o$85bo63b2o12bo3$151b2o$
151b2o4$111b2ob2o$79b2o5bo24b2ob2o3b2o$79b2o3b2ob2o24bo5b2o$84b2ob2o3$
170b2o$163b2o5b2o$163b2o$199b2o$199bo$112b2o51b2o33bo$112bo52b2o32b2o$
98b2o13b3o43b2o$98b2o15bo43b2o$88b2o10bo$87bo2bo7b2o$88b2o8b2o$196b2o$
196b2o2$99b2o$99b2o3$174b2ob2o$173bobobobo$156b2o16bo2bobo$157bo19bob
2o7b2o$157bobo16b2o3bo6bo$158b2o13bobo2bobo8b3o$173b2o2b2ob2o9bo$118b
2o$111b2o5b2o$111b2o$147b2o$147bo6b2o$113b2o33bo6bo$113b2o32b2o6bobo
72b2o$107b2o47b2o72b2o$107b2o$36bo20bo$34b3o11bo7bobo110b2o61b2o$33bo
14b3o6bo111bobo60b2o2b2o3b2o$33b2o16bo16bo75b2o25bo64b2o2bo2bo$50b2o
14b3o75b2o25b2o68b2o$10bo31bo22bo149bo$10b3o27bo3bo20b2o148b3o$13bo31b
o172bo$3b2o7b2o24b2o5b2o15bo104b2o48b2o5b3o6bo$4bo33bo8bo13b2o104bobo
54bo6b5o$4bobo31bo4bobo14bob2o6b2o50b2ob2o36b2o4bo55bo5b2ob3o$5b2o32b
3o3bo2bo11bo2bo6bo50bobobobo35bo5b2o50b2o8b5o9b2o$44bo2b2o9b2obobo4bob
o33b2o16bo2bobo30bo5b3o54bobo9bo4bo6b2o$10b3o31b3o12b2o7b2o35bo19bob2o
7b2o19bobo6bo54bo15b3o$8b3o34bo59bobo16b2o3bo6bo21bo78b3o$9bo2bo46b2o
45b2o13bobo2bobo8b3o96b2ob2o5b2o$10b2obob2o38b2o2b2o60b2o2b2ob2o9bo8b
2o87b3o6b2o$12b2ob2o29b2o6bobo92bo60b2o25b3o10b2o$bo10bo12b2o20bo6bo
91b3o61b2o10b3o25b2o$obo22bo20bo6b2o91bo67b2o6b3o$bo14b2o8b3o17b2o166b
2o5b2ob2o$16bo11bo193b3o$17b3o202b3o15bo$19bo158b2o35b2o6bo4bo9bobo$
178b2o35b2o9b5o8b2o$3b2o52b2o166b3ob2o5bo$2bobo51bobo167b5o6bo$2bo53bo
50b2o15b2o54b2o46bo6b3o$b2o8b2o42b2o50b2o15bobo53b2o2b2o3b2o$9bo2bo4b
2obo78b2o25bo57b2o2bo2bo$8bo4bo3bob2o79bo25b2o61b2o$7b2ob2ob2o85bobo
60bo55b2o20b2o$6bo5b2o59bo27b2o60b3o52bo2bo2b2o14bo2bo$10bo60bobo92bo
52b2o3b2o2b2o11b2o$5bo66b2o91b2o61b2o$11bo171b2o$6bo4bo170bob2o$8bo2bo
107b2o60bo4bo6b2o35b2o$8b3o55b3o50bo62b2o2b2o5b2o35b2o$65bo2bo48bobo
64bo2bo$2o63bo4bo46b2o65bobo$bo63bo15b2o102bo8b2o$bobo67bo8bo2bo110b2o
$2b2o62bo13bobo38bo36b2o38b2o$63b2o5bo10bo38bobo35b2o38b2o$62b2ob2ob2o
50bobo39b2o$56b2obo3bo4bo17b2o31b2ob3o37b2o8bo$56bob2o4bo2bo17bo2bo36b
o45bobo$20b2o42b2o8b2o9bobo31b2ob3o45bo2bo$20bo53bo11bo32b2obo40b2o5b
2o2b2o$18bobo51bobo88b2o6bo4bo$18b2o52b2o17b2o79b2obo15bo$90bo2bo12b2o
65b2o17bo$57bo32bobo12bobo82b3o$57b3o31bo13bo$48bo11bo43b2o$29b2o17b3o
8b2o14bo111b2o$22b2o6bo20bo22bobo90b2o17b3o$22bo6bo20b2o12bo10bo49b2o
39bo2bo2b2o16bo$20bobo6b2o29b2ob2o60bobo39b2o3b2o2b2o12bo$16b2o2b2o38b
2obob2o60bo48b2o$16b2o46bo2bo59b2o$31bo34b3o$7b2o7b2o12b3o31b3o48bo62b
2o$6bobo4bobob2o9b2o2bo81bobo61b2o$6bo6bo2bo11bo2bo3b3o32b2o42bobo$5b
2o6b2obo14bobo4bo31bobo42bo$14b2o13bo8bo33bo27b2o17b2o$14bo15b2o5b2o
24b2o7b2o25bobo17bobo$31bo31bo35bo21bo$10b2o20bo3bo27b3o31b2o21b2o$11b
o22bo31bo$8b3o14b2o$8bo16bo16b2o$19bo6b3o14bo$18bobo7bo11b3o$19bo20bo!

User avatar
Extrementhusiast
Posts: 1966
Joined: June 16th, 2009, 11:24 pm
Location: USA

Re: Finally trying out stable Herschel tracks...

Post by Extrementhusiast » August 29th, 2009, 12:22 am

My best attempt:

Code: Select all

x = 32, y = 55, rule = B3/S23
20bo$19bobo$19bobo$18b2ob3o$24bo$18b2ob3o$18b2obo8b2o$30bo$28bobo$28b
2o7$29bo$27bobo$27b3o$27bo11$14b2o$14b2o12$bobobobo$o$7bo$o2b2o$3b2o2b
obobobo$o13bo2$obobobo3b2o2bo$7bo2b2o$14bo$7bo$8bobobobo!
I Like My Heisenburps! (and others)

User avatar
calcyman
Moderator
Posts: 2932
Joined: June 1st, 2009, 4:32 pm

Re: Finally trying out stable Herschel tracks...

Post by calcyman » August 29th, 2009, 5:57 am

I've made an elementary Herschel conduit recently - the F268:

Code: Select all

x = 131, y = 90, rule = B3/S23
17boo$17boo42bo$60bobo$60bobo$61bo$50bo$50bo$50b3o$52bo$10boo$9bobo$9b
o$8boo4boo$15bo$11bobo$9b3oboo$8bo$9b3oboo$11boboo$$9bo$9b3o$12bo$11bo
bo$12bo$$10bo$9bobobo$9bobboo$8boo3$14bo$13bobo$13boo$51boo$7boo42boo$
8bo$8bobo$9boo6$29boo8boo$9bo19boo8boobboo$9bobo31bobo$9b3o33bo$11bo
33boo$26boo$26boo3$bboo$3bo$3o$o16boo82bo$18bo81bobo$15b3o82bobo$15bo
85bo12$110boo$110boo$$92boo$91bobo$91bo$90boo$121boo$120bobbobboo$120b
obo4bo$101boo18bo5boboo$100bobo21boobobo$100bo23bobbobbo$99boo20bo4bo
bboo$121b5o$$123boobo$123boboo!
It starts off very chaotic, but then cleans up all of the mess that it creates.
What do you do with ill crystallographers? Take them to the mono-clinic!

knightlife
Posts: 566
Joined: May 31st, 2009, 12:08 am

Re: Finally trying out stable Herschel tracks...

Post by knightlife » August 29th, 2009, 5:10 pm

Nice. The use of the rectifier to clean up reminds me...
I found the rectifier can clean up the pesky block that a B-heptomino creates before becoming a Herschel:

Stable B to H using the rectifier:

Code: Select all

x = 106, y = 64, rule = B3/S23
5b2o$4bo2bo$5b2o8$2o$bo$b3o$56b2o$55bo2bo$56b2o8$61b2o$61bo$59b3o3$2b
2o$2b2o2$bo7bob2o48bo14bo$3o6b2obo47b3o12bobo$ob2o56bob2o11bobo$76bo
12$85b2o$85b2o2$67b2o$66bobo$66bo$65b2o$96b2o$95bo2bo2b2o$95bobo4bo$
76b2o18bo5bob2o$75bobo21b2obobo$75bo23bo2bo2bo$74b2o20bo4bo2b2o$96b5o
2$98b2obo$98bob2o!
The stable B-to-H from the life lexicon is shown for reference. The Herschels are mirror images of each other.

If the rectifier is in the way it can be moved further away in the SE diagonal directiion (or closer in the NW diagonal direction if possible), afecting recovery time.

Guam
Posts: 37
Joined: October 7th, 2011, 3:50 pm

Re: Finally trying out stable Herschel tracks...

Post by Guam » October 7th, 2011, 5:58 pm

I found 3 new stable Herschel conduits:
F_266, Ff_266 (can be combined with F_166 or Lf_200) and L_309:

Code: Select all

x = 29, y = 47
bbo$bb3o$5bo21boo$4boo21bo$25bobo$25boo4$oo$oo5$25bo$24bobo$25bo4$20b
oo$20bobo$22bobboo$22b4obo$27bo$24b3o$23bo$23boo3$boo19boo$boo19boo5$
4boo$3bobo$3bo$bboo$$18b3o$9boo8bo$10bo6b3o$7b3o$7bo!

Code: Select all

x = 34, y = 47
26bo$24b3o$oo21bo$bo21boo$bobo$bboo4$27boo$27boo5$3bo$bbobo$3bo4$7boo$
6bobo20boo$6bo22bo3bo$5boo23b4o$$28b4o$27bo3bo$27boo3$5boo19boo$5boo
19boo5$8boo$7bobo$7bo$6boo$$22b3o$13boo8bo$14bo6b3o$11b3o$11bo!

Code: Select all

x = 39, y = 59
28boo$27bobo$21boo5bo$14bo6boo$5bo7bobo$5b3o6bo$8bo$7boo$$22boo8boo$
21bobbo8bo$22boo9bobo$34boo$oo$oo6$3boo$4bo$3bo$3boo10$31boo$31bobo$
33bobboo$33b4obo$38bo$35b3o$34bo$34boo3$12boo19boo$12boo19boo5$15boo$
14bobo$14bo$13boo$$29b3o$20boo8bo$21bo6b3o$18b3o$18bo!
Modified Ff_119 (can be combined with F_166 or Lf_200) for binary counter conduit:

Code: Select all

x = 38, y = 20
bboo$bobo25boo4boo$bo27boo4boo$oo3$30boo$30boobboo$34bobo$15b3o18bo$6b
oo8bo19boo$7bo6b3o$4b3o$4bo3$21boo6boo$21bobo5bobo$23bo7bo$23boo6boo!
Combined Ff158+Rf202 conduit:

Code: Select all

x = 55, y = 55
14boobo$14boboo$$15b5o$14bo4bo$10boobobbo$11boboboo27bo$11bobo5bo10boo
10b3o$10boobo4bobo9boo9bo$11boboobbobbo20boo8boo$11bo6boo31bo$9b3o37bo
bo$8bo40boo$8boo$$53boo$49boobobo$49boobo$o52bo$3o46booboo$bbo15boo30b
obo$bbo15boo30bobo$51bo5$21boo$18bo3bo$18b4o$$16b4o$16bo3bo$19boo3$44b
o$42b3o$41bo$16boo23boo$16bobboo$17boobo$18bo$18bo$17booboboo18boo$16b
obbooboo18bobo$17bo26bo$18b3oboo20boo$20bobo$22bo$21booboo$23bobo$23bo
13b3o$22boo14bo$36b3o!

Sokwe
Moderator
Posts: 2645
Joined: July 9th, 2009, 2:44 pm

Re: Finally trying out stable Herschel tracks...

Post by Sokwe » October 23rd, 2011, 4:47 am

I've been playing around with ptbsearch and have found a few things of interest. First, a periodic R135 Herschel conduit (shown here at periods 4, 5, 6, and 8):

Code: Select all

x = 140, y = 51, rule = B3/S23
17b2o$17bobo$19bo75b2o2b2o$8b2o5b5ob2o72bobo2bo$2b2o5bo4bo6bobo18b2o
38b2o8b2obob2obobob2ob2o13b2o$2bobob3o5b2o2b2obobo18bobob2obo32bobob2o
bo2bo2bo3b2ob2obob2o13bobob2o$4bobo5b3o3bobobo21bobob2o34bobob2o4bob2o
3bo22bobo2bo$4b2o5b3o6bo23b2o9b2o27b2o7b2obob2o2b2ob4o15b2o2b2o$14bo4b
2o31b2o2bo35bo2bo2bo2bobo4bo23b2o$14bo3b3o30b2ob2o36bo2bo2bo3bo2b3o$
10bo3bo3b3o30b2o39bo9bob2o24bo3bo$10bo3bo3b3o30b2o39bo3bob3o29bo4bo$
14bo3b3o30b2ob2o36bo3bo5bob2o26bobobo$14bo4b2o31b2o2bo35bo3bob2ob2ob2o
27bobobo$11b3o6bo34b2o36b2obo2bobo32bo4bo$12b3o3bobobo71bobobo2bo33bo
3bo$2bo11b2o2b2obobo18bo39bo11bobo2b2o21bo$2bobo9bo6bobo18bobo37bobo
10b2o25bobo11b2o$2b3o5b2o3b5ob2o19b3o5b2o30b3o5b2o30b3o5b2o$4bo5b2o7bo
24bo5b2o32bo5b2o32bo5b2o$17bobo$17b2o7$9b2obo36b2obo36b2obo36b2obo$9bo
b2o36bob2o36bob2o36bob2o2$2b2o38b2o38b2o38b2o$2b2o38b2o38b2o38b2o16$b
3o37b3o37b3o37b3o$bo39bo39bo39bo$2o38b2o38b2o38b2o!
It is of particular interest because it uses a well known still life that is part of many low period oscillators but was not used to catalyze larger reactions until a search by MikeP found it. I have had some success in using this still life with ptbsearch. The only use for the new conduit that I could see at the moment is a slight improvement of the p316 gun (using modified p4 sparkers to reduce size further):

Code: Select all

x = 94, y = 81, rule = B3/S23
b2ob2obo$obobob4o5b2o$o2bo6bo4bo5b2o$bobob2o2b2o5b3obobo$2bobobo11bobo
$4bo7b2o5b2o20b2o$3b2o4b2obo28bo$2b3o3bo3bo29bo11bo$2b3o7bo28b2o9b3o$
2b3o7bo38bo$2b3o3bo3bo38b2o10b2o$3b2o4b2obo50bo$4bo7b2o7b2o38bobo$2bob
obo13b2obo37b2o13b2o$bobob2o2b2o6b2o2bobo14b2o36b2o$o2bo6bo6b2o3bo15b
2o$obobob4o3b2o$b2ob2obo5b2o67b2o$82b2o$78b2o$78b2o3$47b2o$48bo34b2o$
45b3o35b2o$12bob2o29bo$12b2obo2$21b2o$21b2o7$31b2o$31bo$29bobo$29b2o
32b2o$14bo47bobo$14b3o45bo$17bo43b2o$16b2o6$71b2o$71b2o20bo$91bobo$78b
ob2o10b2o$48bo29b2obo$9b2o35b3o$9b2o34bo$45b2o3$14b2o$14b2o$10b2o$10b
2o67b2o5bob2ob2o$79b2o3b4obobobo$54b2o15bo3b2o6bo6bo2bo$16b2o36b2o14bo
bo2b2o6b2o2b2obobo$16b2o13b2o37bob2o13bobobo$30bobo38b2o7b2o7bo$30bo
50bob2o4b2o$29b2o10b2o38bo3bo3b3o$42bo38bo7b3o$39b3o9b2o28bo7b3o$39bo
11bo29bo3bo3b3o$52bo28bob2o4b2o$51b2o20b2o5b2o7bo$73bobo11bobobo$71bob
ob3o5b2o2b2obobo$71b2o5bo4bo6bo2bo$77b2o5b4obobobo$86bob2ob2o!
I have also recently been looking at stable circuitry that converts one common pattern to another. As far as I could tell, there were no known century-to-Herschel converters, so here are two (I would not be surprised if these are already known):

Code: Select all

x = 72, y = 28, rule = B3/S23
63b2o$63b2o2$54b2obo$54bob2o$2o$bo28b2o38b2o$bobo26b2o38b2o$2b2o42b2o$
47bo$47bobo$48b2o4$20bo39bo$19b3o37b3o$21b2o38b2o3$11b2o$10bobo$10bo$
9b2o$50b2o$51bo$48b3o$48bo!
The only patterns I can think of where the century features prominently are the p246 gun, bistable switch, and Diuresis. The only one of these patterns in which a century-to-Herschel converter would be applicable would be the bistable switch, but the above converters do not work with it.

With a little bit of searching I came up with this (very slow) pi-to-century converter (this also uses MikeP's still life catalyst):

Code: Select all

x = 33, y = 38, rule = B3/S23
32bo$30b3o$15bo13bo$15b3o11b2o$18bo$17b2o14$6b2o$2b2obobo$o2bobo$2o2b
2o3$5b2o$5b2o6$13b3o$13bobo$5b2o6bobo$4bobo$4bo$3b2o!
This works with one of the century-to-Herschel converters above, and, when combined with the known Herschel-to-pi converter, gives a new conduit, Lx496:

Code: Select all

x = 60, y = 61, rule = B3/S23
42b2o$43bo$43b3o8$18b2o$19bo28b2o$19bobo26b2o$20b2o9$22b2o$21bobo$21bo
$20b2o7b2o$28bobo$28bo29b2o$27b2o29bo$56bobo$56b2o4$14b2o$14b2o5$2o$bo
52b2o$bobo50bobo$2b2o52bo$56b2o6$2bo$2bobo$2b3o34b2o$4bo19b2o7b2o5bo$
25bo7b2o2b3o$14b2o6b3o12bo$14b2o2b2o2bo15b2o$18bobobo16bo$20bob2o13bo$
20bo2bo13b2o$21b2o!
It's so slow and bulky that it probably will not improve any guns, unfortunately.

Also, one of Guam's new conduits can reduce the size of the p421 (and related p842) gun:

Code: Select all

x = 101, y = 76, rule = B3/S23
23bo7bob2o28b2o$21b3o7b2obo27bo2bo$5bo14bo41bob2o$5b3o12b2o38bobobo23b
2o$8bo47b2o2b2o2bo23bo$7b2o47b2o6b3o22b3o$67bo23bo$66b2o$8b2o$8b2o17b
2o70b2o$27b2o69bobo$99bo2$55bobo$55bo4b2o28bo$59bo29bobo$24b2o30b2obob
2o26bobo5b2o$24bo19b2o11b4o29bo6b2o$26bo16bobo10b3o$25b2o16bo11b2obo$
21b2o19b2o11bo2b2o$21bo33b2ob2o$22b3o25b2o4bobo$24bo26bo5bo38bo$48b3o
44bobo$48bo7b2o38bo$56b2o$5b2o55b2o$4bobo56bo$4bo55b3o$3b2o55bo32b2o$
93bobo$95bo$95b2o$77bob2o$77b2obo2$13b2o71b2o$13b2o71b2o2$20bob2o$20b
2obo$4b2o$5bo$5bobo$6b2o32bo55b2o$38b3o55bo$37bo56bobo$37b2o55b2o$43b
2o$4bo38b2o7bo$3bobo44b3o$4bo38bo5bo26bo$42bobo4b2o25b3o$41b2ob2o33bo$
41b2o2bo11b2o19b2o$42bob2o11bo16b2o$42b3o10bobo16bo$2b2o6bo29b4o11b2o
19bo$2b2o5bobo26b2obob2o30b2o$9bobo29bo$10bo28b2o4bo$43bobo2$bo$obo69b
2o$2o70b2o17b2o$91b2o$33b2o$9bo23bo$9b3o22b3o6b2o47b2o$12bo23bo2b2o2b
2o47bo$11b2o23bobobo38b2o12b3o$35b2obo25bo15bo14bo$35bo2bo26bo11b3o$
36b2o25b3o11bo!
There are likely other gun improvements from these new conduits, but I have not tried much.

Request: Does anyone have a thorough collection of the known stable converters. I am especially interested in circuits that convert common objects to other common objects, as well as converters of common objects to gliders (or gliders to common objects). Below is a collection of all of the somewhat interesting converters that I can find (excluding Herschel-to-X, glider-to-X, and X-to-glider converters). Is it missing anything?

Code: Select all

x = 422, y = 364, rule = B3/S23
9b2o$9b2o8$2b2o$bobo$bo$2o4b2o$7bo$3bobo$b3ob2o$o$b3ob2o$3bob2o2$bo$b
3o$4bo$3bobo$4bo7bo$11b3o$2bo7b2obo$bobobo154b2o$bo2b2o154b2o$2o$151b
2obo7bo$151bob2o6b3o$160b2obo3$43b2o$43b2o9$31b2o$31b2o2b2o$35bobo$37b
o$37b2o8$93bo$92bobo$92bobo$93bo12$102b2o$102b2o2$84b2o$83bobo$83bo$
82b2o$113b2o$112bo2bo2b2o$112bobo4bo$93b2o18bo5bob2o$92bobo21b2obobo$
92bo23bo2bo2bo$91b2o20bo4bo2b2o$113b5o2$115b2obo$115bob2o31$284b2o38b
2o38b2o$284b2o38b2o38b2o3$283bo39bo39bo$218b2o28b2o32bobo37bobo37bobo$
5b2o211b2o28b2o33bo39bo39bo$5bo274b3o37b3o37b3o$3bobo203b2obo26b2obo
37bo39bo39bo$3b2o204bob2o26bob2o34b2obo36b2obo$7b2o128b2o138b2ob3o14b
2o18b2ob3o14b2o$7bo129bo145bo13bo25bo13bo$5bobo127bobo144b2o11bobo24b
2o11bobo$5b2o4bo123b2o158b2o4bo33b2o4bo$9b3o287b3o37b3o$8bo289bo39bo$
9bo84bo204bo39bo$10bo82bobo204bo39bo$9b2o82b2o43bo29bo130b2o38b2o$137b
obo27bobo109bo$bo69bo29bo29bo6bo22bo6bo52bo29bo26bobo10bo39bo39bo39bo$
3o67b3o27b3o27b3o27b3o57b3o27b3o25b2o10b3o37b3o37b3o37b3o$o69bo29bo29b
o29bo59bo29bo39bo39bo39bo39bo2$64b2o99b2o46b2o28b2o121bo$63bo2bo98bobo
44bobo27bobo120bobo44b2o$63bobo101bo44bo29bo121bo2bo44b2o3b2o$64bo102b
2o42b2o28b2o76b2o44b2o50b2o$225b2o91bobo$225bo92bo$226b3o28b2o58b2o99b
2o$228bo28bo154b2o4bo$258b3o151b2o5b3o$260bo160bo56$67b2o$67b2o232bo$
104b2o193b3o$104bobo177bo13bo$105bo5b2o171b3o11b2o$111b2o174bo$286b2o
5$21b2o77b2o8b2o90b2o$21b2o77bo8bo2bo89b2o$98bobo9b2o$80bo17b2o$20bo
57b3o120bo$19bobo32b2o21bo122bobo$20bo34bo21b2o122bo$17b3o35bobo140b3o
$17bo38b2o140bo$275b2o$271b2obobo$269bo2bobo$26b2o53b2o55bo6bo61b2o60b
2o2b2o$26b2o53b2o54bobo5b3o59b2o$137bobo8bo$135b3ob2o6b2o125b2o$35b2o
97bo81b2o56b2o$35b2o98b3ob2o75b2o$2o55bo79bob2o40b2o$2o54bobo122b2o$
57bo2$19b3o12b2o33b3o37b3o37b3o48b3o12b2o65b3o$19bobo12bo34bobo37bobo
22b2o13bobo6bob2o38bobo12bo66bobo$19bobo8b2o3b3o23b2o6bobo29b2o6bobo
23bo13bobo6b2obo38bobo8b2o3b3o55b2o6bobo$30bobo4bo22bobo37bobo32bobo
73bobo4bo54bobo$32bo27bo39bo35b2o75bo59bo$32b2o25b2o38b2o112b2o57b2o2$
5b2o179b2o$6bo180bo$3b3o178b3o$3bo180bo33$63b2o$63b2o2$54b2obo$54bob2o
$2o$bo28b2o38b2o$bobo26b2o38b2o$2b2o42b2o$47bo$47bobo$48b2o4$20bo39bo$
19b3o37b3o$21b2o38b2o3$11b2o$10bobo$10bo$9b2o$50b2o$51bo$48b3o$48bo38$
2o$2o3$9b2o$b2o6b2o$bobo$3bo$3bo2$8b2o$8bo$9b3o$11bo!
A side note: Guam's new Herschel receiver (posted here) needs two parallel gliders with a path separation of 4, which fits nicely between the possible separations for Callahan's receiver (2, 5, and 6).
-Matthias Merzenich

User avatar
calcyman
Moderator
Posts: 2932
Joined: June 1st, 2009, 4:32 pm

Re: Finally trying out stable Herschel tracks...

Post by calcyman » October 23rd, 2011, 5:36 am

Impressive work! I may even squeeze a LifeNews article out of these discoveries.
Is it missing anything?
Yes, there's a R-to-H converter from the paleo-Herschel era:

Code: Select all

x = 22, y = 20, rule = LifeHistory
A$3A$3.A$2.2A6.2A$10.2A6$20.2A$9.2A9.2A$10.2A$10.A2$20.2A$20.2A2$11.
2A$11.2A!
What do you do with ill crystallographers? Take them to the mono-clinic!

User avatar
Extrementhusiast
Posts: 1966
Joined: June 16th, 2009, 11:24 pm
Location: USA

Re: Finally trying out stable Herschel tracks...

Post by Extrementhusiast » October 23rd, 2011, 1:37 pm

This, as well as the new stable HWSS Heisenburp (please include that in the article), has re-fired my interest in stable circuitry. However, I'm thinking of using two non-standard reactants: an R-Mango (how common is that, again?), and also an Octomino methuselah (the one that can be synthesized with two gliders).

EDIT: Some promising octomino reactions:

Code: Select all

x = 106, y = 29, rule = LifeHistory
81.2A7.C$81.2A6.3C$89.C.C$91.2C$102.2A$102.A.A$104.A$104.2A4$34.A8.C$
33.A.A6.3C$33.2A7.C.C$44.2C11$2A7.C$2A6.3C$8.C.C$10.2C!
I Like My Heisenburps! (and others)

User avatar
Extrementhusiast
Posts: 1966
Joined: June 16th, 2009, 11:24 pm
Location: USA

Re: Finally trying out stable Herschel tracks...

Post by Extrementhusiast » October 23rd, 2011, 5:30 pm

This looks to be the best candidate for a reaction:

Code: Select all

x = 33, y = 33, rule = LifeHistory
19.2A$18.A.A$19.A3$8.A$8.3A$11.A$10.2A18$2A7.C$2A6.3C$8.C.C$10.2C17.
2A$29.A.A$31.A$31.2A!
I Like My Heisenburps! (and others)

Post Reply