The Hunting of the Elementary Conduits

[dvgrn]

The latest up-to-datest elementary conduits collection and script are attached. Still very much a work in progress -- this is a placeholder post and will be updated again soon. Still have to apply names to the rest of the R->X converters, for example, and also write R->G and W->G recognizers, though that only takes a few minutes.

See also the H-to-G and H-to-Gn collections.

EDIT: Things are looking much better now! There's the latest complete 5 August 2016 update from thunk, which includes the conduit compiler ZIP file. And now there's a November 2017 update from Calcyman, but without the conduit compiler.
EDIT2: I've finally (17 May 2018) gotten around to getting the conduit compiler working and up-to-date again.

EDIT3: Freywa has posted a new and improved version of the ECC (7 December 2018).

Stamp collection pattern:

ElementaryConduits_7December2018.rle.gz

ECC with updates as of December 2018

(134 KiB) Downloaded 135 times

Individual conduit files, with the compilation script that produces the above stamp collection:

ec-7-12-18.zip

Elementary Conduits compiler script, 7 December 2018

(188.76 KiB) Downloaded 117 times

Here's the current center-cell diagram:

Code: Select all

x = 84, y = 21, rule = LifeHistory
.3D7.3D6.D2.D7.2D7.3D7.3D8.2D7.D3.D6.2D$.D2.D6.D2.D5.D2.D6.D2.D6.D2.D
6.D2.D6.D2.D6.D3.D5.D2.D$.D2.D6.D2.D5.D2.D6.D9.D2.D6.D2.D6.D2.D6.D3.D
5.D$.3D7.3D6.4D6.D9.D2.D6.D2.D6.D2.D6.D3.D5.D.2D$.D2.D6.D2.D5.D2.D6.D
9.D2.D6.3D7.D2.D6.D.D.D5.D2.D$.D2.D6.D2.D5.D2.D6.D2.D6.D2.D6.D9.D.D7.
2D.2D5.D2.D$.D2.D6.3D6.D2.D7.2D7.3D7.D10.D.D6.D3.D6.2D3$.4B$6B$8B$10B
$10B4.2B2.2B40.D$2BD10BD7BD38.2D$2BCD10BD6BDBD7.3D8.2D6.3D8.2D9.2D7.D
$B2D7B2.2BCD5BDCDB7.CD7.DCD7.CD8.DCD8.CD7.CD$10B2.B2D8BD6.D.2D6.D2.D
6.3D8.2D8.2D7.3D$13BD8B8.2D8.D.D17.2D9.D$15B.B24.D18.D$5B.B3.B!
#C [[ HEIGHT 240 ]]

EDIT: List of conduit names in text format, from the 17 May 2018 collection:

Code: Select all

BFx157B, BRx161B, BRx161B_Lx303H, BRx46B, BFx199B, BRx204B, BR24B, BB309B, BSW6T73_SW10T285, BSE8T-22, BNW-12T9_SE22T31, BSE22T31, BNE5T66, BSE48T33, BNE17T68, BNW-4T72, BNW5T30, BSW-5T37, BNW39T172, BSE15T2, BSW1T250_BSE15T2, BSW6T73, BNE14T30, BSE1T1, BNW-11T124, BFx59Ha, BFx59Ha, BFx59Hb, BFx59Hc, BFx59Hd, BFx59He, BR146H, BR84H, BBx187H, BF187Ha, BF187Hb, BRx161B_Lx303H, BF20Ha, BF20Hb, BR292H, BBx251H, BLx291H, BF137H, BF22P, BR241P, BF221P, BLx19R, BBx19R, BFx176R, BFx183R, CFx53B, CF85B, CRx28B, CFx50C, CBx37C, CF106C, CNW10T73a, CNW10T73b, CNW-1T27a, CNW-1T27b, CSE12T35, CNE0T0, CNE-1T9, CSW-7T74, CSW-4T34, CNW-12T13, CFx73H, CB148P, CL26R, G000NE_quadri-Snark, G00NE_tremi-Snark, G0C_semi-cenark-base, G0NE_semi-Snark, G0NE_CP_semi-cenark, G0NE_CC_semi-cenark, G11R, G13SE9, G13FxR, G18SW_SEa, G18SW_SEb, G2SWG3, G2SEa, G2SEb, G2NWa, G2NWb, G2H, G2Ra, G2Rb, G3SWG3, G4SWG3, G4SWa, G4SWb, G4R, G5SW, G5SE, G5H, G5R, G6SE2a, G6SE2b, G6H, G6Ra, G6Rb, GNEo_Snark, GNW8a_rectifier, GNW8b_boojum_reflector, GNEe_SWo_NW54_Silver_reflector, G-15FHb_syringe_depend, G-15FHa_syringe, G-2RH_SilverGtoH575rep, G-29RH_SilverGtoH497rep, HFx58Ba, HFx58Bb, HFx107B, HF94B, HRx131B, HLx53B, HL141B, HL79B, HR203B, HR44Ba, HR44Bb, HR44Bc, HR143B, HR117B, HFx197B, HRx32B, HF122B, HLx59B, HBx138B, HF110B, HB56B, HL98B, HL153B, HL226B, HLx103B, HR48B, HLx185B, HF214B, HB62B, HFx236B, HF123B, HB119B, HF144B_adjustable, HRx160C, HR160C, HF104C, HR43C, HSW-2T145_SW-2T21, HSW-2T140_SW-2T21, HSW-2T345_SW-2T21_NW38T161, HSW-1T110_SW-2T21, HSW-1T177_SW-2T21, HNW31T120_NW30T234, HSW0T230_SW-2T21, HSW-4T237_SW-2T21, HSW0T320_SW-2T21, HSW-2T21_SW-4T104, HNW31T120_NW29T213, HNW19T102_NW21T247, HSE39T59_SE41T149, HSW1T118_SW-2T21, HSW1T43_SW-2T21, HSW2T98_SW-2T21, HNW31T207_NW35T419, HSW-7T38_SW-2T21, HSW3T188_SW-2T21, HNW31T120_NW26T163, HSW3T118_SW-2T21, HSW4T188_SW-2T21, HSW-8T89_SW-2T21, HNW25T126_NW31T120, HNW24T386_NW31T120, HSW5T160_SW-2T21_SE22T-8, HSE22T-8_SE29T-2, HSW5T105_SW-2T21, HSW-10T202_SW-2T21, HSW6T167_SW-2T21, HSW7T255_SW-2T21, HSE7T14_SE16T36, HSW10T108_SW19T319, HSW7T94_SW-2T21, HSW-2T21a, HSW-2T21b, HSW10T108, HSW10T97, HSW25T179, HNW19T102, HNW31T120, HNW18T106, HNW8T186, HNW30T138, HNW17T130, HNW16T108, HNW2T132, HNW22T200, HSE-2T47, HNE36T22, HNE18T32, HNE23T140, HNE16T14a, HNE16T14b, HNE8T3, HSE22T-8, HSE-2T80, HSE5T227, HSE31T60, HSE5T23, HSE39T59, HSE7T14a, HSE7T14b, HNE5T-4, HSE23T-6, HNW-5T80, HSE5T12, HNW6T106, HNW16T96, HSW4T44, HSE16T100, HSE29T12, HNE2T56, HNW26T109, HSE-17T106, HSE30T14, HSE21T-4, HSE19T106, HNE1T22, HSW8T86, HSW28T1784, HSW15T116, HSE10T19, HSE-1T41, HNW36T128, HNE34T49, HSE18T9, HNW16T124, HSW3T118, HSW13T112, HNW17T90, HSW15T118, HNW36T126, HNE37T12, HSE10T29, HNW2T196, HNW1T192, HNE-27T176, HSW7T94, HNE15T22, HNW34T204, HNW39T203, HNE-14T138, HSE16T36, HSE19T42, HF116Ha, HF116Hb, HF171H, HFx77Ha, HFx77Hb, HFx119H, HB60H, HR64H, HR126H, HRx140H, HFx158H, HBx125H, HRx162H, HRx191H, HLx86H, HF150H, HR197H, HBx106H_Fx119H, HLx226H, HLx231H, HF271H, HF244H, HBx362H, HR321H, HR246H, HS-15T152M, HF95P, HF2P, HL75P, HL167P, HB62P, HL227P, HF134P, HF135Q, HLx69R, HLx86R, HLx111R, HF157R, HB91R, HF124R, HRx334R, HR107R, HL83R, HLx30R, HL156R, HR149R, HF73R, HL166W, HL163W, HFx135W, HF198W, HLx36W, LNEe_FomichevLtoG, MNWe_MerzenichMtoG, MNEe_FomichevMtoG, PLx9B, PR9B, PB68B, PBx68B, PR112B, PLx112B, PB90B, PBx90B, PF40B, PFx40B, PB173B, PBx173B, PL328C, PRx328C, PLx142C_R142C, PB146C, PBx146C, PF39C, PFx39C, PNW6T138, PNE9T138, PSE24T144, PNW11T160, PNW6T184, PNW-3T177, PNW7T126, PSE20T111, PSE34T59, PSE3T-5, PNE-6T40_SE-6T40, PSE-14T66, PNE19T123, PNW-8T28, PNE8T63, PNW-16T5, PNW6T114_NE17T206, PF171H, PFx171H, PF81Ha, PFx81Ha, PF81Hb, PFx81Hb, PF81Hc, PFx81Hc, PF81Hd, PFx81Hd, PF99H, PFx99H, PLx196H, PR196H, PS12T565L, PL8P, PR8P, PF31Qa, PF31Qb, PLx127R, PR127R, PF140R, PFx140R, PL124R, PRx124R, PF291R, PFx291R, PL153R, PRx153R, PBx30R, PB30R, PB41R, PBx41R, PF35Wa, PFx35Wa, PF35Wa, PFx35Wb, PF56W, PFx56W, QF116B, QFx116B, QLx149B, QR149B, QNE6T144, QNW-10T57, QNE8T54, QNE9T144, QLx232Ha, QR232Ha, QLx232Hb, QR232Hb, QL226H, QRx226H, QF138W, QFx138W, QF157W, QFx157W, RF28Ba, RF28Bb, RF28Bc, RF28Bd, RF28Be, RF28Bf, RF28Bg, RF28Bh, RF28Bi, RBx22B, RLx70B, RLx20C, RSW0T41, RNE-19T84, RNW3T46, RSW5T38, RSW3T62, RNW-16T104, RSW3T42, RSW4T37, RSW-1T33, RNE2T28, RSE-3T46, RSE-3T41, RNW-1T43, RSW13T28a, RSW13T28b, RNE6T46, RF48Ha, RF48Hb, RF48Hc, RF48Hd, RF48He, RF48Hf, RF48H_B164R, RR56Ha, RR56Hb, RR56Hc, RR174H, RR112H, RF29Pa, RF29Pb, RB57P, RF57Ra, RF57Rb, RF48H_B164R, RFx36R, WNW-3T100, WNW-18T15, WFx46H

Names from the August 2016 collection, for comparison -- there are some simplifications in the G(n) converters:

Code: Select all

BB309B, BBx187H, BBx19R, BBx251H, BF137H, BF187Ha, BF187Hb, BF20Ha, BF20Hb, BF221P, BF22P, BFx157B, BFx176R, BFx183R, BFx199B, BFx59H, BLx19R, BLx291H, BNE14T30, BNE17T68, BNE5T66, BNW-11T124, BNW-12T9_SE22T31, BNW-4T72, BNW39T172, BNW5T30, BR146H, BR241P, BR24B, BR292H, BR84H, BRx161B, BRx161B_Lx303H, BRx161B_Lx303H, BRx204B, BRx46B, BSE15T2, BSE1T1, BSE22T31, BSE48T33, BSE8T-22, BSW-5T37, BSW1T250_BSE15T2, BSW6T73, BSW6T73_SW10T285, CB148P, CBx37C, CF106C, CF85B, CFx50C, CFx53B, CFx73H, CL26R, CNE-1T9, CNE0T0, CNW-12T13, CNW-1T27a, CNW-1T27b, CNW10T73a, CNW10T73b, CRx28B, CSE12T35, CSW-4T34, CSW-7T74, G-10G1LR, G-15FHa_syringe, G-15FHb_syringe_depend, G-19G-17FxR, G-22G-17FxR, G-23G-17FxRa, G-23G-17FxRb, G-29RH_SilverGtoH497rep, G-2RH_SilverGtoH575rep, G-30G-17FxR, G-35G-30LxH, G-9G-5LxR, G17G19RR, G25G31BH, G26G31BH, G29G31BH, GG0NEe_Semi-Snark, GG13SE9, GG18SWo_SE10a, GG18SWo_SE10b, GG2NW-8, GG2NW10, GG2SE-2, GG2SE6, GG2SWoG3, GG3SWeG3, GG4SWea, GG4SWeb, GG4SWoG3, GG5SE1, GG5SWo, GG6SE2a, GG6SE2b, GNEe_SWo_NW54_Silver_reflector, GNEo_Snark, GNW8a_rectifier, GNW8b_boojum_reflector, HB119B, HB56B, HB60H, HB62B, HB62P, HB91R, HBx106H_Fx119H, HBx125H, HBx138B, HBx362H, HF104C, HF110B, HF116Ha, HF116Hb, HF122B, HF124R, HF134P, HF135Q, HF144B_adjustable, HF150H, HF157R, HF171H, HF198W, HF214B, HF244H, HF271H, HF2P, HF94B, HF95P, HFx107B, HFx119H, HFx135W, HFx158H, HFx197B, HFx236B, HFx58Ba, HFx58Bb, HFx77Ha, HFx77Hb, HL141B, HL153B, HL156R, HL163W, HL166W, HL167P, HL226B, HL227P, HL75P, HL79B, HL83R, HL98B, HLx103B, HLx111R, HLx185B, HLx226H, HLx231H, HLx30R, HLx36W, HLx53B, HLx59B, HLx69R, HLx86H, HLx86R, HNE-14T138, HNE16T14a, HNE16T14b, HNE18T32, HNE1T22, HNE23T140, HNE2T56, HNE36T22, HNE5T-4, HNE8T3, HNW-5T80, HNW16T108, HNW16T96, HNW17T130, HNW18T106, HNW19T102, HNW19T102_NW21T247, HNW22T200, HNW24T386_NW31T120, HNW25T126_NW31T120, HNW26T109, HNW2T132, HNW30T138, HNW31T120, HNW31T207_NW35T419, HNW34T204, HNW39T203, HNW6T106, HNW8T186, HR107R, HR117B, HR126H, HR143B, HR149R, HR160C, HR197H, HR203B, HR246H, HR321H, HR43C, HR44Ba, HR44Bb, HR44Bc, HR48B, HR64H, HRx131B, HRx140H, HRx160C, HRx162H, HRx191H, HRx32B, HRx334R, HS-15T152M, HSE-17T106, HSE-2T47, HSE-2T80, HSE16T100, HSE16T36, HSE19T106, HSE19T42, HSE21T-4, HSE22T-8, HSE22T-8_SE29T-2, HSE23T-6, HSE29T12, HSE30T14, HSE31T60, HSE39T59, HSE39T59_SE41T149, HSE5T12, HSE5T227, HSE5T23, HSE7T14_SE16T36_NW31T120, HSE7T14a, HSE7T14b, HSW-10T202_SW-2T21, HSW-1T110_SW-2T21, HSW-1T177_SW-2T21, HSW-2T140_SW-2T21, HSW-2T145_SW-2T21, HSW-2T21_SW-4T104, HSW-2T21a, HSW-2T21b, HSW-2T345_SW-2T21_NW38T161, HSW-4T237_SW-2T21, HSW-7T38_SW-2T21, HSW-8T89_SW-2T21, HSW0T230_SW-2T21, HSW0T320_SW-2T21, HSW10T108, HSW10T108_SW19T319, HSW10T97, HSW1T118_SW-2T21, HSW1T43_SW-2T21, HSW25T179, HSW2T98_SW-2T21, HSW3T188_SW-2T21, HSW4T188_SW-2T21, HSW4T44, HSW5T160_SW-2T21_SE22T-8, HSW6T167_SW-2T21_NW31T120, HSW7T255_SW-2T21, HSW8T86, LNEe_FomichevLtoG, MNEe_FomichevMtoG, MNWe_MerzenichMtoG, PB146C, PB173B, PB30R, PB41R, PB68B, PB90B, PBx146C, PBx173B, PBx30R, PBx41R, PBx68B, PBx90B, PF140R, PF171H, PF291R, PF31Qa, PF31Qb, PF35W, PF39C, PF40B, PF56W, PF81Ha, PF81Hb, PF81Hc, PF81Hd, PF99H, PFx140R, PFx171H, PFx291R, PFx35W, PFx39C, PFx40B, PFx56W, PFx81Ha, PFx81Hb, PFx81Hc, PFx81Hd, PFx99H, PL124R, PL153R, PL328C, PL8P, PLx112B, PLx127R, PLx142C_R142C, PLx196H, PLx9B, PNE-6T40_SE-6T40, PNE19T123, PNE8T63, PNE9T138, PNW-16T5, PNW-3T177, PNW-8T28, PNW11T160, PNW6T138, PNW6T184, PNW7T126, PR112B, PR127R, PR196H, PR8P, PR9B, PRx124R, PRx153R, PRx328C, PS12T565L, PSE-14T66, PSE20T111, PSE24T144, PSE34T59, PSE3T-5, QF116B, QF138W, QF157W, QFx116B, QFx138W, QFx157W, QL226H, QLx149B, QLx232Ha, QLx232Hb, QNE6T144, QNE8T54, QNE9T144, QNW-10T57, QR149B, QR232Ha, QR232Hb, QRx226H, RB57P, RBx22B, RF28Ba, RF28Bb, RF28Bc, RF28Bd, RF28Be, RF28Bf, RF28Bg, RF28Bh, RF28Bi, RF29Pa, RF29Pb, RF48H_B164R, RF48H_B164R, RF48Ha, RF48Hb, RF48Hc, RF48Hd, RF48He, RF48Hf, RF57Ra, RF57Rb, RFx36R, RLx20C, RLx70B, RNE-19T84, RNE2T28, RNE6T46, RNW-16T104, RNW-1T43, RNW3T46, RR112H, RR174H, RR56Ha, RR56Hb, RR56Hc, RSE-3T41, RSE-3T46, RSW-1T33, RSW0T41, RSW3T42, RSW3T62, RSW4T37, RSW5T38, WFx46H, WNW-18T15, WNW-3T100

[Sphenocorona]

Probably the main issue is that "Pi" is such a common name and I think there are still many people who wish to call it such. Lowercase "I" doesn't mean anything so it's not like still supporting it as input would produce any ambiguity yet. Additionally, didn't Kazyan say the Pi center would be better closer to the center if the pi's explosion rather than of the pi itself?

[dvgrn]

Probably the main issue is that "Pi" is such a common name and I think there are still many people who wish to call it such.

Sure! And just as we call a Herschel a Herschel and a century a century and so on, we will continue to call a pi-heptomino "pi". Tradition shall be upheld.

In filenames and in LifeHistory labels for this particular project, H is short for Herschel, C is short for Century, and P can unambiguously stand for Pi -- a nice clean extension of Kazyan's single-letter suggestion. Now it's single-ASCII-letter, very convenient for parsing by future scripts.

When the labeling is done, all the label names will be extracted from the files, and they'll be posted in a nice list in a code box in the first post. The idea is that whenever someone finds a conduit that they think is new, they can easily figure out what the name is -- and they'll be able to do a quick text search for it and see if a variant of the same conduit has been found already.

I'd rather search for "HL167P" than "HL167Π", myself. But that was closer to Calcyman's objection. I think the implied request above would be to call the conduit "HL167Pi". So far we have three people expressing three different preferences. Clearly, like the donkey owners in the Aesop's fable, I won't be able to keep everybody happy...!

In this collection and in Hersrch's library, lowercase letters after the conduit name generally mean that there are several useful variants of a conduit -- e.g., Fx119a and Fx119b, or now HFx58Ba and HFx58Bb.

Now, probably no one will ever see "HL167Pi" and think that it's the tenth useful version of a HL167P conduit. But really, what's so wrong with a nice simple one-ASCII-letter abbreviation for each target? I'm betting everyone will get used to it pretty quick.

[simsim314]

Just jumped to my eyes, two missing R->G (they're very useful as they're located at some distance from R, becoming many times the only working connectors):

Code: Select all

x = 78, y = 45, rule = LifeHistory
8$25.A$23.3A$22.A$22.2A3$63.2A$63.A$61.A.A$23.A37.2A$22.A.A32.2A$23.A
33.A.A$58.2A5$12.A$12.2A40.A$11.2A7.2A32.2A$20.2A31.2A7.2A$62.2A5$17.
2A$17.2A40.2A$59.2A!


Great job for the maintenance! And I like the single (english) letter convention it makes it clear and simple to read and use.

[Sphenocorona]

I found a Q->W:

Code: Select all

x = 42, y = 33, rule = LifeHistory
23.2A$21.2B2AB$21.4B5.B$20.4B5.2B$19.6B3.3B$16.9B2.4B$15.10B.4B$12.
17B$12.16B$11.3B2D11B$10.3BDBD12B$11.B2D16B9.2A$10.20B4.2A2.A2.A$9.
21B4.A.A2.2A$8.21B6.B.2A$7.26B.2B2.A$6.31B.A.2A$5.30B2A.A2.A$4.2A16BC
12B2A.2A$4.2A15B3C9B.2B$5.B3.11B5C8B$10.9B2C3B2C7B$8.2A23B$3.2A2.A2BA
2.19B$3.A3.BABA4.16B$2A.A.4BA11.3B$.A.A.2AB13.3B$A2.A.BAB14.B$2A2.A4.
A$5.5A2$4.A.2A$4.2A.A!

It only works with the single-block W->G, but doing that and attaching the H->Q produces an H->G1:

Code: Select all

x = 57, y = 57, rule = LifeHistory
11.2A$10.B2AB$10.3B$10.2B$9.4B$9.4B$9.4B.B$8.8B7.2A$7.9B5.2B2AB$6.12B
3.4B5.B$6.13B.4B5.2B$6.19B3.3B$6.19B2.4B$6.19B.4B$6.23B$6.22B$7.7B2D
11B$8.5BDBD12B$7.5B2D16B9.2A$6.24B4.2A2.A2.A$7.23B4.A.A2.2A$8.21B6.B.
2A$7.26B.2B2.A$6.31B.A.2A$5.30B2A.A2.A$4.2A16BD12B2A.2A$4.2A15B3D9B.
2B$5.B3.11B5D8B$10.9B2D3B2D7B$8.2A23B$3.2A2.A2BA2.19B$3.A3.BABA4.16B$
2A.A.4BA5.11B.B8.2A3.2A$.A.A.2AB7.14B6.B2AB.B2AB$A2.A.BAB7.14B6.4B.3B
$2A2.A4.A5.16B5.2B3.B6.2B$5.5A5.37B$13.2A39B$4.A.2A4.A.A39B.B$4.2A.A
4.A4.38B2A$11.2A5.B.35B2A$20.31B.4B$21.27B4.4B$21.24B$21.B.21B$22.2B
7.3B2.11B$21.B2AB11.12B$22.2A12.12B$37.11B$35.4B.4B3CB$35.2A4.4BC2B$
36.A4.2B3C2B$33.3A6.7B$33.A8.8B$44.B.5B$47.5B$48.5B!

[gmc_nxtman]

Currently useless 28-tick, left-turn E-to-C using a transparent beehive:

Code: Select all

x = 13, y = 13, rule = LifeHistory
2.3DB$.BDAD2B2.BE$2BDABC3BEBE$.2BA2C3BEBE$.3BA5BE2B$.11B$3.9B$7.3B$5.
4B$5.CE$6.C$3.3C$3.C!


I don't think that the output C can connect to anything, though. I guess that if the C can't be used for anything, it has to be one of those almost-conduits that displaces something by one cell.

The input is not too common, unfortunately, so could someone try to search for an H-to-E or a C-to-E?

[dvgrn]

I don't think that the output C can connect to anything, though. I guess that if the C can't be used for anything, it has to be one of those almost-conduits that displaces something by one cell.

The input is not too common, unfortunately, so could someone try to search for an H-to-E or a C-to-E?

I'll certainly be keeping an eye out for E outputs, but they're a little hard to run a dedicated search for -- except maybe with Paul Chapman's search utility or Brice Due's planned rebuild of ptbsearch. Both of those could pattern-match interesting small targets tick by tick as the pattern evolved. But neither project seems to have made it out of pre-alpha stage, and apparently the source code was too incomplete and/or embarrassing to make available.

I haven't added either the E or the H'-that-Freywa-wants-to-call-Q-but-Q-means-queen-bee to the stamp collection of canonical targets in the first post. That won't happen until at least two successful X-to-E-to-Y's have been found, or two X-to-H'-to-Y's that aren't really just X-to-H-to-H'-to-Y.

I think that will make a very reasonable rule of thumb to avoid proliferating targets too much. Canonical objects have to be legitimate "branch points", where it's possible to connect two or more different conduits and send the signal in different directions.

I'm pretty happy with the labeling so far. It's starting to come in handy to double-check things. Maybe this is stating the obvious, but there's a workable algebra of elementary conduits. I know I labeled simeks' Fx169 and Bx297 correctly, because Fx169 = HF110B + BFx59H, and Bx297 = HF110B + BBx187H.

The next stage is to build a database and collect all the center-to-center X,Y offsets for each elementary conduit. Then it will be trivial to write a script that connects each conduit's output to every known conduit with a compatible input, and tests to see if it works (maybe with user assistance in welding).

Along the same lines as Hersrch, we'll need a designation for "hopeless -- never show me this again", versus "maybe somebody can weld those catalysts together, but I don't see a solution". Eventually this might result in a sizable pre-built database of known-to-work composite conduits, which would speed up searches considerably.

Ideally a new Elsrch would be able to handle possibilities like H-to-G-to-syringe-to-transmitter-to-receiver, to get to a specific spacetime location for an output signal -- a Herschel or glider or whatever. We'll see if that turns out to be an unmanageably large search space.

[dvgrn]

I found a Q->W...
It only works with the single-block W->G, but doing that and attaching the H->Q produces an H->G1...

The H-to-G1 has been added to the H-to-Gn collection, which is now updated with simsim314's additions as well. Does anyone know offhand the answer to my question (in the above linked post) about the broken G18, by the way?

I've added the Q-to-W to my current build of the ECC (I think I've typed "elementary conduits collection" enough times now...) Will post another update to the ECC some week soon.

This H-to-G1 looks eerily familiar... is there a pi-to-century that looks just like it, or something? Or maybe I'm just getting paranoid, due to recent incontrovertible evidence of my failing memory...!

[calcyman]

but there's a workable algebra of elementary conduits

To a first approximation it's a category, where the objects are things like H and R, and the morphisms are conduits. There is a functor from this category to the monoid of natural numbers (where a conduit is mapped to the delay) and another functor to the automorphism group of Z^2 (which encodes how the object is rotated, reflected and/or translated).

The reason I write 'to a first approximation' is that this ignores geometric constraints (such as the impossibility of F117-R64-R64-R64-R64-F117).

What letter are we using for the primed Herschel? I believe that 'I' is unused, and it's logical by virtue of being the successor of 'H' in the Roman alphabet.

[dvgrn]

What letter are we using for the primed Herschel? I believe that 'I' is unused, and it's logical by virtue of being the successor of 'H' in the Roman alphabet.

Let's just call it H' for now. Capital i looks like lowercase L, and there's no need to assign a single-letter abbreviation until a target is proven to be a real branch point for actual known conduits. Otherwise we could use up the alphabet really quick with H'', P''', C''', and lots of silly "conduits" that don't need any catalysts.

On a mostly-unrelated question, Karel Suhajda's Hersrch spends a lot of its time building an Aho-Corasick tree before it runs a search. I've looked that up in Wikipedia, but I don't have a really good sense of how efficient it might be compared to other methods.

It looks as if Hersrch's search method is probably going take too long, as soon as we add things like tandem gliders and syringes, and transitions through vacuum like G/G2/G4/G5/G6->[+1,+1]. Any thoughts on how to actually find workable connections between target1@F[0](0,0) and target2@arbitrary_orientation[T][X,Y]?

For example, how much of the search could usefully be done just once, slowly and painfully, and saved to a database?

[simeks]

Does anyone know offhand the answer to my question (in the above linked post) about the broken G18, by the way

It was posted here (as a reduction of this)

This H-to-G21 could be added too.

[simeks]

A new H-to-G with timing NW6T106:

Code: Select all

x = 27, y = 34, rule = LifeHistory
11.2C$12.C$10.C$10.5C$14.C$8.4C$8.C2.C4$25.2C$25.C$23.C.C$23.2C8$E$E.
E$3E$2.E16.C$18.C.C$19.C4$15.2C$15.C.C$17.C$17.2C!


[thunk]

Simeks' SE5T12 should probably be included. It turns out to be very useful.

[biggiemac]

Some X-to-Y conduits release a glider in addition to their output. If the glider is a different timing than any of the X-to-G listed then I feel like such conduits should also have a place in the X-to-G table. I just rediscovered an R-to-G that I thought must be obvious, only to not see any such G output in the R-to-G table. I eventually saw it in the R-to-B table, while the conduit I found had used an eater to prevent the B from being formed.

[dvgrn]

Some X-to-Y conduits release a glider in addition to their output. If the glider is a different timing than any of the X-to-G listed then I feel like such conduits should also have a place in the X-to-G table. I just rediscovered an R-to-G that I thought must be obvious, only to not see any such G output in the R-to-G table. I eventually saw it in the R-to-B table, while the conduit I found had used an eater to prevent the B from being formed.

Yes, I'd like to mark optional eaters with a different color, in general, and/or somehow indicate the trail or signature of the optional output object.

I've seen a few more conduits going by that should probably be marked in yellow. Not stopping to sort that out right now -- it gets a little tricky, because some combinations have to be connected with custom catalysts that would be hard to find if you were starting with just the elementary versions.

Eventually some ElemSrch utility will have a good collection of all known ways to custom-weld every possible pair of conduits together -- but in the meantime it's better to keep a record of that kind of thing in the main collection, I think.

--------------------------------

I've finally worked my way through the P's and Q's. They were just as annoying as I thought they would be, because of their symmetry. When they convert to asymmetric objects, there's always either a mirror-image L and Rx, or Lx and R, conduit pair.

The current naming convention didn't seem to have a really good way of handling that. I didn't really like the looks of "PLxR196H", meaning "either PLx196H or PR196H"... so I added each conduit separately.

Then I got to the pi-to-glider converters, and wasn't sure I wanted to add fifteen more converters just to deal with the mirror-image problem. What would people rather see -- more trivially equivalent converters, or uglier names? PNW6T134 could theoretically become PNWSW6T134, I suppose. But I think I'll probably just end up adding the fifteen extra P-to-G variants (?).

I'd like to avoid characters in names that are forbidden in any common operating systems, so things like "PNW/SW6T134" or "P[NW|SW]6T134" are no good. At some point I think it will make sense to change all the filenames to match the names given in the #N comment lines... that way conduits will automatically sort themselves by output direction, and it will be a little easier to find the right file to edit.

Here's the lane diagram for pi-to-glider conduits. Mostly from sheer inertia I kept the pi centers where I wanted them:

Code: Select all

x = 140, y = 83, rule = LifeHistory
45.2D3.2D41.2D3.2D$44.D2.D.D2.D39.D2.D.D2.D$46.D2.D2.D35.3D3.D2.D2.D$
45.D3.D2.D40.D3.D2.D$44.4D2.2D40.4D2.2D2$40.D3.2D53.D3.2D$39.2D2.D2.D
9.3A23.3A13.2D2.D2.D$3D.6D.3D26.D2.D2.D9.A3B21.3BA9.3D2.D2.D2.D21.3D.
9D$.2D2.D2.D3.D27.D2.D2.D10.A2BA19.A2BA15.D2.D2.D22.2D2.D2.D3.D$.2D2.
D2.D3.D26.3D2.2D12.4B17.4B15.3D2.2D23.2D2.D2.D.D$.D.D.D2.D.D.D46.A2BA
15.A2BA46.D.D.D2.3D$.D.D.D2.D.D.D25.2D11.3A6.4B13.4B6.3A11.2D25.D.D.D
2.D.D$.D.D.D2.D.D.D24.D2.D10.A3B6.A2BA11.A2BA6.3BA10.D2.D24.D.D.D2.D$
.D2.2D2.D.D.D24.D2.D11.A2BA6.4B9.4B6.A2BA11.D2.D24.D2.2D2.D3.D$3D.2D
3.D.D25.D2.D12.4B6.A2BA7.A2BA6.4B12.D2.D23.3D.2D.6D$38.2D14.A2BA6.4B
5.4B6.A2BA14.2D$46.3A6.4B6.A2BA3.A2BA6.4B6.3A$28.D3.2D12.A3B6.A2BA6.
3B2.4B6.A2BA6.3BA12.D3.2D$27.2D2.D2.D12.A2BA6.4B6.AB3.BA6.4B6.A2BA12.
2D2.D2.D$23.3D2.D2.D2.D13.4B6.A2BA17.A2BA6.4B14.D2.D2.D$28.D2.D2.D14.
A2BA6.4B15.4B6.A2BA15.D2.D2.D$27.3D2.2D7.3A6.4B6.A2BA13.A2BA6.4B6.3A
6.3D2.2D$41.A3B6.A2BA6.4B11.4B6.A2BA6.3BA$21.2D3.2D14.A2BA6.4B6.AB13.
BA6.4B6.A2BA14.2D3.2D$20.D2.D.D2.D14.4B6.A2BA27.A2BA6.4B14.D2.D.D2.D$
16.3D3.D2.D2.D15.A2BA6.4B25.4B6.A2BA17.D2.D2.D$21.D3.D2.D7.3A6.4B6.A
2BA23.A2BA6.4B6.3A8.D3.D2.D$20.4D2.2D8.A3B6.A2BA6.4B21.4B6.A2BA6.3BA
7.4D2.2D$37.A2BA6.4B6.A2BA19.A2BA6.4B6.A2BA$38.4B6.A2BA6.4B17.4B6.A2B
A6.4B$39.A2BA6.4B6.A2BA15.A2BA6.4B6.A2BA$40.4B6.A2BA6.4B13.4B6.A2BA6.
4B$41.A2BA6.4B6.A2BA11.A2BA6.4B6.A2BA$42.4B6.AB8.4B9.4B8.BA6.4B$43.A
2BA16.A2BA7.A2BA16.A2BA$44.4B16.4B5.4B16.4B$45.A2BA16.AB7.BA16.A2BA$
46.4B41.4B$47.AB43.BA$69.3D20.B$70.CD$69.3D$47.AB43.BA$46.4B41.4B$45.
A2BA16.AB7.BA16.A2BA$44.4B16.4B5.4B16.4B$43.A2BA16.A2BA7.A2BA16.A2BA$
42.4B6.AB8.4B9.4B8.BA6.4B$41.A2BA6.4B6.A2BA11.A2BA6.4B6.A2BA$40.4B6.A
2BA6.4B13.4B6.A2BA6.4B$39.A2BA6.4B6.A2BA15.A2BA6.4B6.A2BA$38.4B6.A2BA
6.4B17.4B6.A2BA6.4B$37.A2BA6.4B6.A2BA19.A2BA6.4B6.A2BA$21.2D3.2D8.A3B
6.A2BA6.4B21.4B6.A2BA6.3BA8.2D3.2D$20.D2.D.D2.D7.3A6.4B6.A2BA23.A2BA
6.4B6.3A7.D2.D.D2.D$16.3D3.D2.D2.D15.A2BA6.4B25.4B6.A2BA17.D2.D2.D$
21.D3.D2.D14.4B6.A2BA27.A2BA6.4B15.D3.D2.D$20.4D2.2D14.A2BA6.4B6.AB
13.BA6.4B6.A2BA13.4D2.2D$41.A3B6.A2BA6.4B11.4B6.A2BA6.3BA$28.D3.2D7.
3A6.4B6.A2BA13.A2BA6.4B6.3A7.D3.2D$27.2D2.D2.D14.A2BA6.4B15.4B6.A2BA
14.2D2.D2.D$23.3D2.D2.D2.D13.4B6.A2BA17.A2BA6.4B14.D2.D2.D$28.D2.D2.D
12.A2BA6.4B6.AB3.BA6.4B6.A2BA13.D2.D2.D$27.3D2.2D12.A3B6.A2BA6.4B.4B
6.A2BA6.3BA11.3D2.2D$46.3A6.4B6.A2BA3.A2BA6.4B6.3A$38.2D14.A2BA6.4B5.
4B6.A2BA14.2D$2.4D.3D.3D23.D2.D12.4B6.A2BA7.A2BA6.4B12.D2.D25.4D.6D$.
D3.D2.D3.D24.D2.D11.A2BA6.4B9.4B6.A2BA11.D2.D24.D3.D2.D3.D$.D6.D3.D
24.D2.D10.A3B6.A2BA11.A2BA6.3BA10.D2.D24.D6.D.D$2.3D3.D.D.D25.2D11.3A
6.4B13.4B6.3A11.2D26.3D3.3D$5.D2.D.D.D46.A2BA15.A2BA50.D2.D.D$5.D2.D.
D.D27.D3.2D12.4B17.4B16.D3.2D27.D2.D$.D3.D2.D.D.D26.2D2.D2.D10.A2BA
19.A2BA14.2D2.D2.D22.D3.D2.D3.D$.4D4.D.D28.D2.D2.D9.A3B21.3BA9.3D2.D
2.D2.D22.4D2.6D$40.D2.D2.D9.3A23.3A14.D2.D2.D$39.3D2.2D52.3D2.2D2$45.
2D3.2D41.2D3.2D$44.D2.D.D2.D39.D2.D.D2.D$46.D2.D2.D35.3D3.D2.D2.D$45.
D3.D2.D40.D3.D2.D$44.4D2.2D40.4D2.2D!

[Sphenocorona]

I have a different suggestion for conventions with symmetry-breaking conduits: Instead of naming it "PLxR196H" (ugly) or having both "PR196H" and "PLx196H" listed separately (space consuming and visibly redundant), we can do something else:

Let's refer to both of them as "PRv196H". I used "v" here as an example, it could be some other letter.

By the same logic, this "QL226H"/"QRx226H" would simply just be called "QLv226H" regardless of mirroring:

Code: Select all

x = 38, y = 58, rule = LifeHistory
34.4B$10.2A3.A17.4B$11.A2.A.A15.4B$9.A4.A.A14.4B$9.5A.A14.4B$13.A15.
4B$9.2AB2.A13.4B$8.B2A3B.A11.4B$8.5B.2A10.4B$9.3B13.4B$5.8B11.4B$4.9B
10.4B$3.10B9.4B$3.10B.3B4.4B$3.14B3.4B$3.15B.4B$3.19B$3.3BD14B5.B$3.
3B2D16B.B2A$.6B2D17B2A$7B3D17B$.6B2D18B$3.3B2D17B$3.3BD17B$3.20B$3.
20B$3.20B$3.19B$4.17B$4.18B$6.15B$6.11B3.B2A$7.10B3.BA.A$6.12B5.A$6.
12B5.2A$6.13B$6.12B$7.10B6.2B$8.9B5.4B$7.9B5.5B$8.8B5.6B$7.10B.9B$7.
21B$7.17B2A2B$3.A.2AB.14BABA2B$.3AB2AB2.13B2A3B$A4.B4.18B$.3A.2A4.17B
$3.A.A6.15B$3.A.A7.2B2.2B3.6B$4.A7.BA8.5B$12.A.A6.6B$13.2A6.6B$22.5B$
22.3D2B$23.BDB$23.3D$24.B!

(By the way, that one wasn't quite shown right in the October 1st version of the collection, I don't know if it's fixed in the new version).

QF116B would have to then be QFv116B, and all symmetry breaking P->X and Q->X conduits would have a "v" in their names when referenced as a pair. Referring to an individual isomer would still be done as QF116B or QFx116B, as in that case the mirroring matters again.

For glider outputs, we could have the south lane forms used in the dual listing, as the FNG is a south glider.

[dvgrn]

I have a different suggestion for conventions with symmetry-breaking conduits: Instead of naming it "PLxR196H" (ugly) or having both "PR196H" and "PLx196H" listed separately (space consuming and visibly redundant), we can do something else:

Let's refer to both of them as "PRv196H". I used "v" here as an example, it could be some other letter.

I kind of like this idea. It would certainly be possible to get used to it. A new letter would keep things short but signal a mirror-reflection option.

I'm not absolutely convinced yet, though. I've posted an October 25 update so people can get a better look at the current option, which is indeed space-consuming and visibly redundant. Pros and cons so far:

1) "Rv" doesn't immediately tell you that it means R or Lx, as opposed to Rx or L. That ends up being just another arbitrary convention to learn (though it's certainly not difficult).

2) If you're hunting through the collection for, let's say, Lx conduits that fit a particular purpose, it's really convenient to have them all right there and visible next to each other. This is even more true for X-to-G converters. With the Redundancy Option, once the files are renamed to match the #N names, everything will sort into the right order automatically, and you really get to see everything that's available.

2.5) Okay, it's true, it will be a minor pain again to get the yellow composite conduits to sort to the end.

3) For any "v" conduit, an ElemSrch utility will probably just have to have some extra parsing code, and will convert Lv and Rv and Fv conduits into mirror-image pairs of conduits in its internal library. It doesn't seem worth doing the extra handling of mirror-image cases in the middle of a search, when there are relatively few conduit types that need that kind of special handling.

3.5) Putting that another way: there really aren't very many converters with mirror-symmetric inputs, so a minor amount of duplication might be the simplest and least confusing option.

9) What happens if a rotationally symmetric target signal shows up? LOMs might actually allow L or R or Lx or Rx isomers, so I guess we could use "Rz" and maybe "Rxz" to avoid those duplications if the problem comes up. I'm not really very worried about this last point.

(By the way, [QL226H] wasn't quite shown right in the October 1st version of the collection, I don't know if it's fixed in the new version).

It's fixed now -- thanks for pointing it out. I also found that my C-to-G recognizer script was getting some lanes and timing a little bit wrong -- have redone those names. Quite likely I've made some new mistakes elsewhere.

[gmc_nxtman]

Not sure why this H-to-G4 isn't in the collection:

Code: Select all

x = 32, y = 22, rule = LifeHistory
27.A$25.3A$8.A15.A$8.3A13.2A4.2A$11.A9.5B4.A$10.2A3.B4.4B3.BA.A$10.8B
.7B.B2A$12.16B$12.16B$11.17B$9.18B$7.19B$7.2BE15B$6.3BEBE13B$7.2B3E
13B$6.5BE12B$5.10B3.6B$4.8B6.7B$3.8B9.5B$2.8B11.4B$.8B11.B2A2B$8B13.
2A!


By the way, here are a simple seed and synthesis or the H' object, in which the H' is generated "out of nowhere" both times:

Code: Select all

x = 8, y = 7, rule = B3/S23
o$obo$2o2$bo4b2o$obo3b2o$2o!


Code: Select all

x = 16, y = 9, rule = B3/S23
13bo$o3bo8bobo$b2obobo6b2o$2o2b2o3$9b2o$9bobo$9bo!


Also, is there any way to restore the beehive here?

Code: Select all

x = 8, y = 8, rule = B3/S23
bo$obo2bo$b2o2bobo$5b2o2$2b2o$bo2bo$2b2o!


[thunk]

I was bored today, so I made an unofficial update of the collection, addressing all the TODO things dvgrn has listed after last entry. That is, if I didn't bork anything, which is likely to have happened.

My naming scheme for the Gn->G, H, R is a bit idiosyncratic--I tried to apply the standard conventions, but they needed some modification. Also, please excuse my mediocre typesetting skills.

[dvgrn]

My naming scheme for the Gn->G, H, R is a bit idiosyncratic--I tried to apply the standard conventions, but they needed some modification. Also, please excuse my mediocre typesetting skills.

Do you also have a ZIP archive of all the individual conduit files, so that the conduit-compiler script can rebuild the whole stamp-collection pattern when more conduits are added? Or did you do the editing by hand, and that's why the typesetting skills were needed?

The names look reasonable on first glance at least. I definitely didn't have any better ideas for how to hang names on the G(n)-to-X conduits. It will take a while to go through everything again thoroughly and see what exactly has been changed. Anyway, thanks for the update!

[thunk]

Do you also have a ZIP archive of all the individual conduit files, so that the conduit-compiler script can rebuild the whole stamp-collection pattern when more conduits are added? Or did you do the editing by hand, and that's why the typesetting skills were needed?

The names look reasonable on first glance at least. I definitely didn't have any better ideas for how to hang names on the G(n)-to-X conduits. It will take a while to go through everything again thoroughly and see what exactly has been changed. Anyway, thanks for the update!

Sorry, I did it by hand--so I don't have an archive of the files.

A note on naming: o and e at the end of reflectors/converters mean "odd lanes" and "even lanes". In the first release, they originally referred to color-changing and color-preserving, respectively, but the new system worked better with the XWSS->G converters. I just posted the mk3 release which has these designations fixed.
Canonical configurations of the L and M are as follows:

Code: Select all

x = 15, y = 5, rule = LifeHistory
11.D$D2.D5.D3.D$4.D9.D$D3.C4.D4.C$.4D5.5D!


(edited to clarify new release)

[thunk]

Conduit files!

I actually used the conduit name as the file name, so some scripts might have to be rewritten.

Also I'm starting to doubt the choice of "o" and "e" for odd and even-lane spaceship reflectors, as it looks like it's going to cause parsing problems. If anyone has any better ideas for naming the G(n)-->things, please speak up.

EDIT: Added about 30 conduits to the folder that were missing from the collection. Some of those are of dubious utility, but they can always be taken out later if dvgrn doesn't want to clutter things unnecessarily.

EDIT 2: Incorporated the added conduits (and other things) into the old naming scheme, so they appear automatically as part of the collection generation.

[triller]

What is the cheapest sacrificial glider to Herschel reaction? .rle please!

[thunk]

What is the cheapest sacrificial glider to Herschel reaction? .rle please!

Sorry, I don't have the answer to that offhand. But glider+block makes a pi (as in the center of the syringe). And a glider+ sideways boat makes a B heptomino (again, forgot the exact one).

Also, a conduit collection update!
Added a few more recent discoveries and a few small H-Gs from the H-G collection (no point in adding all the elementary H-Gs there, they already have their own thing).

EDIT: The "PF40B" in the original is a broken version of the PFx99H (which looks composite at first glance). The former has been replaced with the latter.
The prefix HFx77H has been removed from Kazyan's H->M, as it is unnecessary. The designation has been changed accordingly.

[dvgrn]

What is the cheapest sacrificial glider to Herschel reaction? .rle please!

Sorry, I don't have the answer to that offhand. But glider+block makes a pi (as in the center of the syringe). And a glider+ sideways boat makes a B heptomino...

Glider+block to pi to Herschel is fairly cheap, depending on what exactly you need. The cheapest one is even free, no sacrifice needed, if you have time and space for a Silver G-to-H -- or it just needs a glider from somewhere to cleanup an extra beehive, instead of a Herschel to rebuild something.

Another recent option is glider+beehive to R to Herschel:

Code: Select all

x = 67, y = 75, rule = B3/S23
43bo$41b3o$40bo$40b2o$31b2o$31bobo$32bo3$43b2o$43b2o14$13b2o$13b2o$34b
2o$33bo2bo$34b2obo$37bo$37b2o3$62b2o$62bo$7b2o55bo$8bo35b2o14b5o$8bobo
34bo13bo$9b2o16b2o16bobo12b3o$27b2o17b2o15bo$60b4o$55b2o3bo3b2o$55b2o
4b3o2bo$27b2o34bob2o$27b2o34bo$9bo52b2o$9bobo$9b3o$11bo42b2o$54bo$27bo
27b3o$26bobo28bo$27bo4bo$2b2o26b3o$3bo25bo$3o26b2o$o16b2o$16bobo$16bo$
15b2o3$19b2o$18bobo5b2o$18bo7b2o$17b2o2$31bo$27b2obobo$26bobobobo$23bo
2bobobobob2o$23b4ob2o2bo2bo$27bo4b2o$25bobo$25b2o!

There are lots of other stop-and-restart Herschel circuits available nowadays, if that's the kind of thing you need. They really ought to be collected into their own thread, probably -- haven't gotten around to doing that yet.

If you have more specific requirements for recovery time, available space, or reconstruction options, just post them and I can probably dig up something to match.