Question: compilation of pi catalyst searches?

[Scorbie]

I have seen many catalyst discoveries that are based on known reactions. AFAIK, the rectifier is based on a known pi reaction (~ the transparent beehive, right?), and the syringe's beehive with tail catalyst was known, if I remember correctly.

Are there any other searches that dealt with the pi heptomino, since the search space is fairly large and it's more effective to avoid duplicate search spaces?

[dvgrn]

I have seen many catalyst discoveries that are based on known reactions. AFAIK, the rectifier is based on a known pi reaction (~ the transparent beehive, right?), and the syringe's beehive with tail catalyst was known, if I remember correctly.

Are there any other searches that dealt with the pi heptomino, since the search space is fairly large and it's more effective to avoid duplicate search spaces?

Paul Callahan did a lot of searching in 1996, and came up with the standard Herschel receiver and several of its variants. See the new elementary-conduits stamp collection -- some of the G-to-Gs are also from that era. The new variant from simsim314 and the boojum reflector are also based on the same initial reaction.

So it seems to be worth re-visiting old searches every now and then -- just adding a few new catalysts is enough to open up a lot of new search space.

[Kazyan]

I'm investigating the pi heptomino as well, in effort to find a spartan G-to-X. I've already found the Spartan beehive factory and that exploding-pond thing I posted in the CatForce thread.

In the earlier parts of the Herschel thread and the G-to-X thread, I posted several Bellman catalysts for the Pi, which I've investigated with 2-catalyst runs in CatForce, though I didn't try very hard.

One technique I'm trying now is to CatForce one side of a pi (2 catalysts), take two promising interactions, and place them on opposite sides of the Pi to see what happens when it runs (hopefully, a block/hive/boat/ship/pond will appear in just the right place). This gives a fairly large number of combinations to investigate, even if most of them are incompatible.

EDIT: The beehive-with-tail apparently has a few niche uses, but it was found by Bellman in my semi-manual exploration of branches of the search tree.

[simsim314]

I would recommend to run Pi with CatForce using 3 catalysts, and fit-in-width-height feature. This should bring up some interesting options.

[Scorbie]

Here's my attempt on finding the baseline for the spartan syringe. Any suggestions? (especially the search area)
picol.in

Code: Select all

max-gen 180
start-gen 68
last-gen 110
num-catalyst 3
stable-interval 15
search-area -10 -3 19 19
pat 2o$2o19$18bo$17b2o$17bobo!
cat 2o$2o! 60 0 0 . forbidden obo$b2o$bo2$2o$2o! 0 -4 forbidden 2o3bo$2ob2o$4b2o! 0 0 forbidden b2o$b2o2$bo$2o$obo! -1 0 forbidden 2o$b2ob2o$o3b2o! -4 -1 forbidden o3b2o$b2ob2o$2o! -4 0 forbidden obo$2o$bo2$b2o$b2o! -1 -4 forbidden 4b2o$2ob2o$2o3bo! 0 -1 forbidden 2o$2o2$bo$b2o$obo! 0 0 
cat 2o$o$b3o$3bo! 12 -2 -2 * forbidden 2o$o$b3ob2o$3bobobo$6bo! -2 -2 forbidden bo$obo$2o2$2o$o$b3o$3bo! -2 -6 forbidden bo$2bo$3o2$3b2o$3bo$4b3o$6bo! -5 -6 forbidden 2bo$obo$b2o2$3b2o$3bo$4b3o$6bo! -5 -6
cat bo$obo$obo$bo! 50 1 2 /
cat b2o$o2bo$bobo$2bo! 35 2 2 +
cat bo$obo$bo! 15 1 1 .
cat 2o$obo$bo! 25 1 1 +
cat 2o$obo$b2o! 35 1 1 |
fit-in-width-height 12 8
output picol.rle

[Scorbie]

It turns out that the pi search ended with no fruitful results. (And it also turns out that I did this search before and just deleted it.)

Code: Select all

x = 328, y = 521, rule = B3/S23
17b2o81bo$17bo82b3o$15bobo85bo$11b2o2b2o85b2o6b2o$11b2o97b2o$105b2o$7b
2o95bo2bo$7b2o96b2o2$102b2o$8bo92bo2bo$7bobo91bobo$7bobo92bo$8bo10$29b
o98bo$28b2o97b2o$28bobo96bobo77$11b2o$11b2o4$5b2o$5b2o3$b2o$2bo$2bobo$
3b2o$7b2o$6bobo$6b2o5$29bo$28b2o$28bobo77$11b2o97b2o97b2o97b2o$11b2o
97b2o97b2o97b2o10$b2obo5b2o88b2obo5b2o88b2obo6b2o87b2obo6b2o$obob3o3b
2o87bobob3o3b2o87bobob3o4b2o86bobob3o4b2o$bo5bo91b2o5bo92bo5bo91b2o5bo
$6b2o97b2o97b2o97b2o6$29bo98bo98bo98bo$28b2o97b2o97b2o97b2o$28bobo96bo
bo96bobo96bobo77$11b2o97b2o$11b2o97b2o5$4b2o97b2o$4b2o97b2o3$3b2o97b2o
$4bo5b2o91bo6b2o$b3o6b2o88b3o7b2o$bo98bo7$29bo98bo$28b2o97b2o$28bobo
96bobo77$11b2o$11b2o5$4b2o$4b2o3$3b2o$4bo$b3o$bo$10bo$9bobo$10bo4$29bo
$28b2o$28bobo77$11b2o97b2o$11b2o97b2o5$5b2o97b2o$5b2o97b2o3$4b2o97b2o$
5bo4b2o92bo5b2o$2b3o5b2o89b3o6b2o$2bo98bo7$29bo98bo$28b2o97b2o$28bobo
96bobo!

[simsim314]

It turns out that the pi search ended with no fruitful results.

Well I wouldn't call it "not fruitful" just mostly known. How about increasing the search area, say -20 -10 40 40. Try to fit in a bit larger square say 15 15, set last gen 160. If it takes too long remove some catalyst (like loaf or boat, as they take 4 symmetries, and not as useful as others), or fit in something a bit smaller.

I would also try something crazy as 4 catalyst leaving only blocks (or in smaller area blocks + eaters only). Play with fit-in-width-height parameter to find reasonable run-time.

[affamatodidio]

I absolutely despise necroposting. However,
the advent of Barrister and better searching tools means that a lot of searches will get lost in threads on the forums. Say, AlbertArmStain's sandbox or JP21's... whatever they do. A great forum thread for general searches of sequences could be made and would improve the quality of searches - no doing-over something someone else did. We have a honey farm thread - why not pi sequences?

[AlbertArmStain]

A great forum thread for general searches of sequences could be made and would improve the quality of searches - no doing-over something someone else did. We have a honey farm thread - why not pi sequences?

I did make a thread for pi catalysts, it died out though.

EDIT by dvgrn: Collection threads never die -- they just go dormant, sometimes indefinitely.

The thread that AlbertArmstain is talking about has only been dormant for a year and change; I think nobody should have the slightest worry that a new post to it will be seen as necroposting.

The title of the thread can be adjusted as needed, by changing the title of the first post.