New honey farm catalyst

[MikeP]

Here's an interesting one: a still life that reacts with an evolving honey farm, producing a long-lived spark that eventually decays to a pair of blocks.

Code: Select all

x = 137, y = 19, rule = B3/S23
13b2o112b2o$12bobo113bo$2bo9bo29bo39bo39bo5bob2o3b2o$obo8b2o27bobo37bo
bo37bobo6bobo4bo$b2o38b2o38b2o38b2o8bob3o$11b4o115b2obo$11bo2bo$13bo$
14b3o$16bo39bo$54b3o$53bo$6b2o38b2o3bo2bo31b2o38b2o$6b2o38b2o3b4o31b2o
2b2obo32b2o$91bob3o$51b2o36bobo4bo$52bo35bob2o3b2o$52bobo33bo$53b2o32b
2o!

The reaction works in 4 different orientations and the spark looks very promising. When my current search run finishes I'll set one up to look for ways of perturbing it, and I think it might be worthwhile to run ptbsearch or something on it too.

[Sokwe]

As usual, your new catalyst is quite amazing. Unfortunately, I didn't find any practical use for it. The following reaction converts a glider to a small unstable pattern, but it is too close to the new catalyst to be useful:

Code: Select all

x = 17, y = 19, rule = B3/S23
12bo$10b3o$9bo$9b2o2$6b2o$6b2o3$16bo$14b3o$13bo$11bo2bo$11b4o$b2o$obo
8b2o$2bo9bo$12bobo$13b2o!

I would like to see more of these novel catalysts. Excluding block, eater, and eater2 reactions, these are all of the 'simple' catalyst reactions that I know of.

Code: Select all

x = 102, y = 13, rule = B3/S23
89b2o$89b2o$o11bo16bo22bo39b2o$bo11b3o9bob2o9bo23b2o2b2o12b2o9bo2bo2b
2o$2o12bo11bo25b2o10b3o9bob2o11bobo4bo$o38b3o11b2o22bo14bo5bob2o$3bo9b
obo11b2o38bo12bo14b2obobo$obobo8b2obo11bo12bo11bo13b3o9bobo13bo2bo2bo$
2o2bo11bo10bo12bobo9bobo15bo9bo11bo4bo2b2o$4b2o10b2o9b2o12bo11b2o14b2o
21b5o2$94b2obo$94bob2o!

If you find anything that is not one of these reactions, it is probably new.

[MikeP]

My searcher finds reactions like this occasionally too.

Code: Select all

x = 7, y = 11, rule = B3/S23
2o3b2o$obobobo$2bobo$b2ob2o3$3bo$2b3o$o4bo$3b2o$3bo!

Does this belong in your collection?

[Sokwe]

I'm not sure if I would include reactions like this because they are not significantly different from block reactions. The 'basic' block reactions seem to be these:

Code: Select all

x = 17, y = 7, rule = B3/S23
2b2o11b2o$2b2o11b2o2$2bo12bo$bobo10b2o$o3bo9bo$2bo12bo!

There are a few reactions using other catalysts that are essentially equivalent, but are not used often because the block reaction is smaller. For example, the first 'type' of reaction in the above pattern can be done by the mirrored bookends (as you noted), as well as an eater 2:

Code: Select all

x = 31, y = 12, rule = B3/S23
28bo$27bobo$27bobo$12b2o3b2o6b3ob2o$12bobobobo5bo$2b2o10bobo8b3ob2o$2b
2o9b2ob2o9bob2o2$2bo12bo12bo$bobo10bobo10bobo$o3bo8bo3bo8bo3bo$2bo12bo
12bo!

And the second 'type' of reaction can often be done by a long hook with tail (this is sometimes useful, because it gives slightly more room on one side):

Code: Select all

x = 17, y = 8, rule = B3/S23
11b2o$b2o9bo2b2o$b2o9bobobo$13bo$bo14bo$2o13b2o$o14bo$bo14bo!

Something to note about the dual bookends is that it seems to be a somewhat versatile catalyst for symmetric reactions, but it is rarely used in catalyst searches, where reactions tend to be asymmetrical.

Also, here are common boat and tub reactions that I neglected to include in my previous post:

Code: Select all

x = 17, y = 6, rule = B3/S23
3b2o6b2obo$o2bo8bobo$2bo9bo$o3bo10bo$3bobo6bobobo$4b2o9bo!

Some questions about your program:

• Does the 'drifter' part of your program attempt to find all of the possible still life catalysts that fit the search parameters, or is there some stochastic element to the search?
• How often does it find new catalyst reactions (i.e. reactions that you haven't seen before)?

[MikeP]

Some questions about your program:

• Does the 'drifter' part of your program attempt to find all of the possible still life catalysts that fit the search parameters, or is there some stochastic element to the search?

If I run it to completion it'll find everything.

• How often does it find new catalyst reactions (i.e. reactions that you haven't seen before)?

Maybe once a week or so, but it varies a lot.

It mostly finds blocks, eaters and eater2s, which in many cases I ignore since I assume this space has been well covered by ptbsearch etc.

It also misses things that I do know about - that is it's rare for it to find an eater3. I restrict the search space in two main ways: a limit on the number of live cells that the drifter phase adds, and a limit on the number of cells in the catalyst which are allowed to differ from the stable configuration; to find an eater3 requires both of these parameters to be set fairly high, which is impractical for larger sparks.