Soup-searching for potentially conduit-yielding reactions

[MathAndCode]

EDIT by dvgrn Related links:

• 1599: Thread for posting stable Herschel conduits, excluding Herschel-to-glider converters
• 1682: Thread for posting stable Herschel-to-glider converters
• 1849: Thread for posting stable elementary conduits
• 4790: Thread for posting periodic Herschel conduits
• 4996: Thread for posting periodic elementary conduits
• 4993: Thread for posting speculative elementary conduits using common active reactions (known signals)
• 4809: Thread for posting speculative conduits with uncommon active reactions
• 4891: Thread for posting speculative reactions that might allow for future conduits
• 4868: Thread for collecting speculative chaos-edgeshooting reactions
• 4879 (this thread): Thread for collecting speculative conduit-yielding reactions, e.g., from soup search results
• 4994: Thread for posting dirty conduits that recover but leave extra junk
• 4971: Thread for collecting bottlenecks (low-population active configurations in known conduits)

When I was trying to compact a ∏-accepting conduit partial earlier today, I saw one side of a ∏-sequence crash into and destroy a couple of fishhooks, but I noticed that the reaction caused another ∏-sequence to sprout out of the opposite side in such a short amount of time that any different in cells from the interaction with the fishhooks would have had to travel at over 80% of light.

By the way, is that ∏ seen facing upwards at generation 235 extractable? I think that it's extractable. A signal from the fishhooks would have had to travel 16 rows in 19 generations in order to reach the bottom of the newly-formed ∏ (and 16 rows in 18 generations in order to reach the bottom of the ∏'s hexaplet parent), which is nearly the speed of light, and I think that that means that the ∏ is extractable. (The Elementary Conduits Collection only has two ∏→∏ conduits, and neither has a step time of 31.)

As mentioned in the quote, this looked promising for a conduit. This gave me an idea: Let's search random soups specifically looking for reactions like that that could potentially yield conduits.

Here is a similar example from this soup.

Code: Select all

x = 56, y = 55, rule = B3/S23
obobobo4bob2o2bob4o2bo5b4o3bobo2b2ob2o2bo4bo$3bobob4ob5obobo3bobo3bobo3bobobob2o6b4obo$2b5obobo5b4ob3obobo6b2ob2ob3obobo4bo2b2o$bo3b2o2bob3obo2b2o3b2ob4o4b6obobo3bob2ob2obo$bo4bobob7o2b2o2b2o3bo2b2o3bo3bobo4bo2bobobo$obob6obo2bo5bo3bob7ob5o3b3o3b2o4b2o$o2bo5b2obobo2bo2bob4o7bobo2bo4b4o4bob3o$4ob6o6b3obo2b6obo2b2obo2b3obobo2b2obob2o$2o2bob3o2bob2o2b2obobob4o5bobo6bob3o3b2obobo$b2obo2b2ob2ob2ob4o2bo2bob2obo4bob3ob2obob2obob2obo$b4obo4bobob2o2b3ob2o2b2ob2o4bo2b5obo2b3o$bob2o3b4ob2o2bo2b2obob2obobob3obo3bobo8b3obo$b2obob2o2b7o4bo2bo4bo4b3o4bob2obo2bo2bo$6b3obobob2o3b12ob2o5bobo2bob2obo2bo$5o3bo4b2o4bo2bobobobob7obob3ob2obo2bobob2o$o3bobo3b2obo2b3ob3o6bob2obobo6b3ob2o3bo$2bob3o3bo3b7ob3o2b2o2bo2b2obo2b2obobo4b2obo$2o6bo2b3obobo5bob2ob2ob7ob2obobob2o3bo$bo3b2ob2o2b2obob2o2b2obobobob3obobo2bo7b2o2bob2o$3o3b2ob2o5b2ob2obo7b4ob2o2bobo4bob4o3bo$b2obobo4b5o6b2obob5obo4bob9obo$bob3o2b3o2bo11b2ob3ob5o4b3o4bobo3bo$b5o3b4o2bo2b2o3b3o4bobo2bobobo2bobobobo3b4o$2bo3bo2b2o2b2ob2o3b2obob4ob2ob8o3bo7bobo$bo2b2ob5o2bo2bo3b2o2bo2b11o3b3obo5bobo$6ob5o2b3o4b3o2bob3o3bobob2o2b2o2b3ob2o$3o3bobo2bo2bo2bo4bo2bob3obo2b2o2bo3bob3obo5bo$3ob2o2b2o2b2o3bob2ob2ob3o2bo3b2obobo4b2o5bo2bo$o2b5obo3b2o2b4o3bo4bobo2bobo2b3o3bo4bo$2ob4obo3b5o2b2ob3o4bob2ob2o3b2o3b2obobo3bobo$obobobo2bobo3b2ob2o2b3o2b4ob3ob6o5bo3b2o2bo$3bo3b2o3b2o2bob2o4b2o2bob3obobo4b3ob4o4bo$bob3o4b3o3b3ob2ob3ob4o2bobob3ob3obobobo3bobo$6o2b7o4b2o3bo5b2o2b3o2bob2o3bo3bobobo$3ob4obobo2b2ob2o2b3o2bob4obo7b4o3bo2bo2b2o$b5o2b3obo4bo2b2o2b3o2b3o4b3ob3o4bo5bo$4b3ob2ob2obob2o2b4o3b5o2b3o10b2o3bobo$o2b3ob3o3b3ob4obo3b2o5b4ob3ob5o2bobo$2o3bo4bobo2b3obobobob7o2b4obo5bobob2obo$3b2obo2bo5bob4obob2ob2o2bo2bo4bo2b3ob3o2b2o$ob3o4b3o5b2o4b2o7b2ob5o2bobobo2bo2b2obo$3o2bobo2b3o2b2ob2ob5o2bo2b2o3b2obob3o2b2o3b2ob2o$5b3o3bobo6b2ob5ob2obobob2o3b4obo2b3obobo$ob3o3bo2b5obo2b4obob3o5bobo2bo2b3obo5b3o$2obobo2bo4b2ob2o3bo4bo2b3o3bo4bo2bo4b5o2bo$ob2ob4o3b7o2b3obo3bobo2b5ob5o2b2obo4bo$bo7b2ob3o2b2obob5ob4ob3o4b3obob2o2b2obo$b2o2b2o2bo2b2obo2bo2b2obo4b3ob2o3bobobo4bo2b2o2b2o$o2bob5obo4bo6bobo6bob4ob3o4b3o6bo$b3o3bob2o5b4obo2bo2bo2b3obo2b6obo2b3o$4obo2bo3b2ob2o5bo3b3o3b6obo2b7obobobo$3o2b2obob7o2b3ob4ob5ob2ob2ob4o3b5o3bo$2b2o2b3o5b2ob4ob2ob2obobo5bobo4bo2bob3o3b2o$2b6o2bob3obo2bobo2b4o2bo4b3o2b3ob4ob7o$o2bo2b4obobo3b2o4b3ob3obob2ob3o2bobobobobobobo!

The relevant ∏-sequence starts at generation 262. Here is the reaction. (The nearby century and one of the nearby blocks are extraneous.)

Code: Select all

x = 3, y = 13, rule = B3/S23
b2o$b2o9$3o$obo$obo!

Twenty-six generations after the initial generation, two ∏-heptominoes are visible, one on each side. In order to reach the hexaplet grandparents, the signal from the block would have had to travel at 6c/9 for the ∏ grandparent on the left and 7c/9 for the ∏ grandparent on the right, so this could yield a ∏T26 or even a signal splitter.

The same soup also yields this reaction starting at generation 1,194.

Code: Select all

x = 6, y = 13, rule = B3/S23
2o$2o$2o8$3b3o$3bobo$3bobo!

The necessary speed of information transmission from the beehive to the glider may appear to be 9c/20, but because the region that eventually forms the glider detaches from the rest of the chaos earlier, the necessary speed is actually 8c/10, so this could yield a ∏→G.

Here is another soup.

Code: Select all

x = 56, y = 55, rule = B3/S23
o2bob2o9bo5bobob3o2bobob2o3bob2o4b5o$3obo4bob6ob5obob2o2b2obob7o5b2o4bobo$2b3obobobob2ob4o4b2o2b2o5bo3bo2b3ob3obo2b3o$2bo3b2o8bob2o4b2o4bob5o2b2obo2b2o4b4o$2obo2b2ob3o7bo5bob5o3bo2bobobobob2o2bobo$bob2ob2ob3ob2o3bo3b3ob2o3b3o2b2obobobo2b2o6b2o$2obo2bob5o4bob2obob2o3b3ob3obobo2b2ob2o2bob2o2bo$2ob4ob3obo2b2obo3b2o2bob2ob4o3bob2o2bob7ob2o$b3ob2ob2o2b2obobob2ob2obo2bo3b2o3b5ob3o2bobo2bobo$obo2bobo3b3ob3o3bobob2ob2ob2o3bo2b9o2bobo2bo$2ob3obo2b3ob2o2bobo3bo2bo3b2obo2b2ob2obobo3b2obo2bo$2ob2o2bo2bobo2b4ob2o3b2ob2o5b2obobob2o3bo2b4obo$4o2b2o3b2o2bo2bobo2bobobob2ob3ob2ob3obobo2bo4bo$o3bo3b3ob2o2b6obobob2o3bobobo3bo4bob2o3b2obo$bo3bo2bob2ob3ob2ob3o2bobob6ob2o2b4o2b2obob2o2bo$2bo2bob3ob6o6b4o5bo3bobob4o4b3o3bo$b2o5bobo2b7obo4bo4b2o2b2obo2b2ob4o2b6o$3b2o2b3obo3bobobo3b2o3bo2b3ob3obo2bob3o2bob4o$bobo4b3o5b4ob2ob3o4b4o4bobo2b2o2b4ob2o$3b2ob5o2bo2b5obobo2bob3o5b3o2bo2b2obo2bo3bo$ob3ob7o2bob8ob6obobo2bob3o3b3o$6o3b2obo2bob2o2b2o2b2ob3o2bob2o5bo3bo2b2obo$2o5bobobo2bob3o2bobo2bo2bo4bob4o2b2ob5o2bo2bo$obo3b3o3bob4o2b2o3b2o2bo2bo2b4o3b3o5bob2o$2b2o2bobo3b4o3b2obob4ob2o4bob3o3bo3b2ob2obobo$o2b4o2bo7b3ob2obo3bobobo4b2o2b5o3bo2bob2o$bo3b2o5bo3b5obo2b3ob3obobob3o2bobob2o2bo2bo$b2obo2bo2b2o2bob3o3b4o5bob2obo6bo4bobo2b2o$o2b2obo2b2ob3o2b2o2bob2obob2o2b2ob2ob4ob3obobo3bobo$ob3o3bo5b2ob3ob2o3bobobo2bobobob2ob2o2b2o2b3obo$bo2b2obob3ob2o3bo3b4o4bo2bob2o2bobo4b4o2b4o$3obob2obob2o2bo2bo4bo4bo7b2o2b2o2bobo2b3o2bo$bob3o2bob2obobo7bo3b2o2b2obo2b4o5b2ob2ob4o$4bobobo2b2o2b3ob2ob3ob2o2bo3bobo3b3o2b4obobo$3b2o2b9obo3b4o2b4ob2obo2bob3o2b4ob2o3bo$bo2b3o2b2obobob4ob7ob3o2bo2b2o2bob4ob2obob3o$2obo10b3ob2o2bobo3bo2bo6b2o2b4ob2obobo$2o3bo2b2ob3o5b3o2b2o4bo2bo2bo5b3ob2obobobobo$o8bo3bo2bobo2b2ob3o4bo2b4ob2o3bo2bobobobobo$o5bob2obo2bo2bo4bo2b2obob2ob3obob4ob3ob3o4bo$bo2b3o2b2o3bo2bob2o2b2o7b5o2bo2bo2bo3bobobobo$b4o4bobob3o4b2o3b3o2bo2bobobo4b2obobob3o2bo$2b2obob3o2bo3bo2b2o3b4o2b3obo2bob2o2b5o3bo2bo$obo2b4ob2o4b4o3bob3o3bo2bo4bo3bobo7bo$7o2bo7bob2o3b2o3b2ob2o2b3o4bob3obob2o$o3bo5bo5b3ob2o7bob3o2bobo5b2o2bo2b2ob2o$b6o4b4o3b6ob2o2bobob4o2bob4o2b2ob5o$5o3b2obo2b3obo3bo5b2obob2ob5ob4obob4o2bo$2b3ob2o5bobobob2ob3o4b2obo4bobob3o2bo3bobobo$o3bobob3o2b3o2bo2b4o3b2ob2ob2o3bob2ob2obobo5bo$b5o2bo2b2obob3ob2obo2bo3bobobobob4ob2ob3o2bobo$2o2b3ob2o2bo3bo4b7o3b2obob6ob3obob2o2bo$3ob2obob2ob3obo5bo2bo3bo2bobo3bobob2o2bobo2b5o$ob2obobobo3b6obo4bo2b4ob3o3bo2b5o4bob2o$2ob2o4bo2b3ob3ob2obo2bob2obo2b3obobob2o2b5o3b2o!

Here is part of generation 1,252.

Code: Select all

x = 8, y = 5, rule = B3/S23
2o2b3o$2o2bobo$4bo2bo$5bobo$5b3o!

The necessary speed of information transmission from the block to the ∏'s pentaplet grandparent is 5c/7, so this could yield a S→∏.

[MathAndCode]

I found a potentially useful ∏→B reaction with necessary information transmission speed 22c/30.

Code: Select all

x = 36, y = 20, rule = LifeHistory
20.3D$8.D.2D8.D.D$8.3D9.D.D$9.D4$10.2A$2A6.2A.2A$2A6.A2.A$8.A2.A$9.2A
2$9.2A10.3A.3A$8.A2.A8.A.2A.2A.A$2A6.A2.A7.2A7.2A$2A6.2A.2A7.A.2A.2A.
A$10.2A9.3A.3A5.3A$33.A$34.2A!

The B does interact with some trailing junk behind it soon after forming, but I'm not concerned about that because it looks possible to perturb the ∏ closer (both distance-wise and time-wise) to the B's formation in order to suppress said junk.

[MathAndCode]

Here is a partial ∏T26∏ with necessary information transmission speed 6c/6.

Code: Select all

x = 15, y = 12, rule = LifeHistory
3D$D.D$D.D5$12.2A$3.3C4.2A2.A$3.C5.2A2.2A$3.3C3.2A.2A$11.3A!

[bubblegum]

Here is a partial ∏T26∏ with necessary information transmission speed 6c/6.

Code: Select all

x = 15, y = 12, rule = LifeHistory
3D$D.D$D.D5$12.2A$3.3C4.2A2.A$3.C5.2A2.2A$3.3C3.2A.2A$11.3A!

For convenience could you rrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr use the standard pi abbreviation (P) in the designations, because 1) it's nice to just be able to use ASCII and not have to deal with that extra byte of UTF-8 2) that isn't pi anyway that's the continued product and 3) it seriously looks like you're decorating "T26" in dingbat style.

[MathAndCode]

For convenience could you rrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr use the standard pi abbreviation (P) in the designations, because 1) it's nice to just be able to use ASCII and not have to deal with that extra byte of UTF-8 2) that isn't pi anyway that's the continued product and 3) it seriously looks like you're decorating "T26" in dingbat style.

Okay; I shall do that in the future.

[MathAndCode]

Code: Select all

x = 8, y = 7, rule = B3/S23
5bo$2o2b3o$2o4bo3$6b2o$6b2o!

This suggests a P→R with necessary information transmission speed 8c/10. (It also suggests an R→P, but that does not look at useful and has a slower necessary information transmission speed.)

[MathAndCode]

Code: Select all

x = 45, y = 44, rule = B3/S23
35b2o$35bobo$36bo6b2o$43bo$41bobo$41b2o9$24b2o$8bob2o13bo17b2o$8b3o11b
3o18b2o$9bo12bo$35b2o$36bo$33b3o$33bo3$14b2o$14b2o$2o$2o4$3b2o8b2o$4b
o8bobo$b3o11bo$bo13b2o2$10bo$7b2ob2o$7b5o$10b2o$10bo2$10bobo$11bo!

Observe the interaction in the upper-right starting at generation 227.

Code: Select all

x = 16, y = 9, rule = LifeHistory
14.2A$14.A$12.A.A$12.2A$9.C$.D6.2C$2D7.2C$.2D$2.D!

Can the fishhook be replaced by a permanent catalyst?

[dvgrn]

Observe the interaction in the upper-right starting at generation 227...

At T=227 you have an R-pentomino heading leftward, and the fishhook turns it into a century, but the century shows up in the middle of a bunch of already-used territory from T=140 to T=160 or so. Why would a century be useful there?

To get the R at T=227 you have to sacrifice a boat catalyst.

If you're not planning on actually using this example conduit to create an input R, then is there another conduit that can put an R in that position without interfering with that fishhook -- let alone the unlikely hypothetical replacement for the fishhook?

The replacement might take up more space if more space is available, but we won't know what space is available unless we first find an actual working R input conduit to ask the question about... that doesn't eat boats, and doesn't cross the space where the century is showing up.

(Or have I just got the idea all wrong, as usual?)

[MathAndCode]

At T=227 you have an R-pentomino heading leftward, and the fishhook turns it into a century, but the century shows up in the middle of a bunch of already-used territory from T=140 to T=160 or so. Why would a century be useful there?

To get the R at T=227 you have to sacrifice a boat catalyst.

If you're not planning on actually using this example conduit to create an input R, then is there another conduit that can put an R in that position without interfering with that fishhook -- let alone the unlikely hypothetical replacement for the fishhook?

The R-pentomino doesn't have to be formed that way. I tend to include the original reaction that I found the reaction in, but should I stop doing that? I didn't worry very much about clearance.



Edit: I just found this:

Code: Select all

x = 33, y = 15, rule = LifeHistory
13.2A3.2A$13.A.A.A.A$15.A.A$14.2A.2A6$3D27.3D$D31.D$3D27.3D$15.3C$15.
C.C$15.C.C!

Is there a way to prevent the bookends from getting destroyed or to replace them with another catalyst that doesn't get destroyed? If so, this could yield a p41 gun.

[dvgrn]

The R-pentomino doesn't have to be formed that way. I tend to include the original reaction that I found the reaction in, but should I stop doing that?

No need to stop, but maybe make it a little more parenthetical? "(originally found in this reaction at T=227): [ RLE ]". This one had a particular lot of extraneous clutter.

I guess It would be less confusing to me if the initial introduction to an idea was boiled down to simplest possible form. You don't necessarily have to worry about not confusing me, it's just that I might complain less often.

[MathAndCode]

No need to stop, but maybe make it a little more parenthetical? "(originally found in this reaction at T=227): [ RLE ]". This one had a particular lot of extraneous clutter.

I guess It would be less confusing to me if the initial introduction to an idea was boiled down to simplest possible form. You don't necessarily have to worry about not confusing me, it's just that I might complain less often.

Okay; I'll try to remember to do that in the future.

[MathAndCode]

Is there a suitable replacement for the block at the right?

Code: Select all

x = 21, y = 14, rule = LifeHistory
8.2A$8.2A6$8.3C8.2A$9.C9.2A$9.3C$2.D$.2D$2D$.2D!

This might yield a P→B. Here is my attempt at using catalysts:

Code: Select all

x = 11, y = 11, rule = LifeHistory
.A$D.D3.A$CAC2.A.A$DCD3.2A3$5.2A$5.A2.A$6.2A.A$9.A$9.2A!

[MathAndCode]

Here's a potential PT31P with necessary information transmission speed 15c/19 from this soup:

Code: Select all

x = 15, y = 20, rule = B3/S23
3o$2bo$3o5$10b3o$8b2o3bo$8bo2bo2bo$8bo4b2o$9b3ob2o$14bo$12b2o4$10b2o$10bobo$10bo!

[wwei47]

Here's a potential PT31P with necessary information transmission speed 15c/19 from this soup:

The interactions look fairly light.
EDIT: LWSS?

Code: Select all

x = 17, y = 13, rule = B3/S23
15bo$3o11bobo$obo11bobo$3o12bo3$5b2o3b2o$4bobo3bobo$4bo7bo$4bo3bo3bo$
5bo5bo$6bo3bo$7b3o!