Collecting data for Hersrch updates

[Sokwe]

Edited to include all previous suggestions

You seem to have added 'V' in front of the conduit names for the disallowed following conduits. This will only disallow a specific variant of a conduit (in this case, usually the standard variant). By entering the conduit name rather than the variant name, all variants of that conduit are forbidden. I did it this way to save myself the effort of entering each conduit variant name. For some reason, this is not done anywhere in the rest of the pairs list, even though the comments say that it works. Anyway, here is the "corrected" pairs data for L106P6 (using conduit names instead of variant names in most cases):

Code: Select all

# VL106P6-Bx167: 38b2o$22b2o14bo$11b2o9b2o12bobo$11b2o23b2o2$2b2o$2b2o$38b2obo$38bob2o2$2b2o39b2o$2b2o38bobo$38b2o2bo$38bobob2o$8bo31bo$39b2o4bo2bo$7bo37bo$5bo6b2o30bo4bo$12b2o29bobo2b2o$4bo36b2obo$15b2o$3o12b2o3bo$2bo16bobo19bo2bo$2b2o16bobo20b2o$22bo14b2o$22b2o13bo$35b3o7$44b2o$25bo18bobo$25bo20bo$25b3o18b2o$27bo11$25bo2bo$25bo$24bo4bo$23bobo2b2o$21b2obo$35b2o$35b2o2b2o$21bo2bo8b2obo2b2o$23b2o8b3o3$34b2o$34b2o!
VL106P6-VBX125
VL106P6-Bx222
VL106P6-F117
VL106P6-F171
VL106P6-F349
VL106P6-Fx77
VL106P6-Fx78
VL106P6-Fx158
VL106P6-Fx176
VL106P6-Fx349
VL106P6-L112
VL106P6-L156
# VL106P6-L198: 59b2o$43b2o14bo$43b2o12bobo$57b2o4$59b2obo$59bob2o$o$3o61b2o$2bo15b2o43bobo$2bo15b2o39b2o2bo$59bobob2o$61bo$30bo29b2o4bo2bo$29bobo34bo$30b2o33bo4bo$21b2o41bobo2b2o$22bo39b2obo$21bo$21b2o$62bo2bo$64b2o$58b2o$58bo$56b3o7$65b2o$46bo18bobo$46bobo18bo$46b3o18b2o$48bo11$46bo2bo$46bo$45bo4bo$44bobo2b2o$42b2obo$56b2o$56b2o2b2o$42bo2bo8b2obo2b2o$44b2o8b3o3$55b2o$55b2o!
VL106P6-VLX65P3
VL106P6-VLX65P3X1
VL106P6-VLX65P3X2
VL106P6-VLX65P3
VL106P6-VLX65P3X1
VL106P6-VLX65P3X2
VL106P6-VLX65P8
VL106P6-VLX65P8X1
VL106P6-VLX65P8X2
VL106P6-VLX65P4
VL106P6-VLX65P4X1
VL106P6-VLX65P4X2
VL106P6-VLX65P5
VL106P6-VLX65P5X1
VL106P6-VLX65P5X2
VL106P6-VLX65P7
VL106P6-VLX65P7X1
VL106P6-VLX65P7X2
# VL106P6-Lx73: 25b2o$24bo2bo$24bobobo$23b2obobo$25bobob2o$20b2o2b2obo2bo$26bobobob2o$16b2o5b2obob3o2bo$16bo7bo2bo3b2o$14b3o7bob5o$25bo4bo$26b4o$28bo$30bo$29b2o2$23b2o$4bo18bobo$4bobo18bo$4b3o18b2o$6bo11$4bo2bo$4bo$3bo4bo$2bobo2b2o$2obo$14b2o$14b2o2b2o$o2bo8b2obo2b2o$2b2o8b3o3$13b2o$13b2o!
VL106P6-Lx138
VL106P6-Lx142
# VL106P6-Lx159: 20b2o24b2o$20b2o8b2o14bo$10bo19b2o12bobo$10b3o31b2o$13bo$12b2o2$46b2obo$2bo43bob2o$2bo$3o48b2o$o49bobo$46b2o2bo$46bobob2o$48bo$2o45b2o4bo2bo$2o3bo47bo$4bobo4b2o39bo4bo$5bobo2bobo38bobo2b2o$7bo2bo38b2obo$7b4o9bo$19bobo$9b2o9b2o27bo2bo$9b2o40b2o$45b2o$45bo$43b3o7$52b2o$33bo18bobo$33bobo18bo$33b3o18b2o$35bo11$33bo2bo$33bo$32bo4bo$31bobo2b2o$29b2obo$43b2o$43b2o2b2o$29bo2bo8b2obo2b2o$31b2o8b3o3$42b2o$42b2o!
VL106P6-Lx212
VL106P6-VLX318P15
VL106P6-R190
VL106P6-VR318P15

I have also compiled data for Lx159:

Code: Select all

Lx159 Lx 159 15 -44

  # Standard variant, with eater
  VLX159 1 130 0 GRU data/vlx159.rle 1 6 data/ulx159.rle 2 3

  # Standard variant, without eater (no glider)
  VLX159X 1 130 0 GRX data/vlx159x.rle 1 6 data/ulx159x.rle 2 3

vlx159.rle:

Code: Select all

10b2o3b2o$10b2o4bo$16b3o2$9bo$8bobo$9bo$6b3o$6bo$3b2obo$3b2ob3o14b2o$
9bo13bo$8b2o11bobo$21b2o6$5bo$4bobo18b2o$4b2o19b2o9$24b2o$24b2o12$o$o
22b2o$3o20bo$2bo10b2o3b2o4b3o$11bo2bo4bo6bo$11b3o4bo$18b2o$11b2obo$11b
ob2o!

ulx159.rle:

Code: Select all

16b3o$17bo$17b3o2$9b4ob7o$9bob10o$10bob8o$14b6o$15b6o$14b7o$14b6o$10bo
3b6o$9b2o3b6ob2o$9b11ob3o$11b11o$10b11o$9b12o$9b15o$9b15o$7bob16obo$7b
21o$8b20o$10b15obo$10b14o$9b13o$10b12o$11b10o$10b12o$9b14o$9b17o$10b
17o$10b17o$9b17o$6b16o$5b16o$5b16o$4b16o$3b20o$3b20o$4b20o$3b22o$b24o$
24o$ob13ob10o$obob4ob7o2b2o4b2o$o3b4o2b3obo4bo4bo$3ob4o4b4o3b2o$7o4b2o
2bo$12b3o!

vlx159x.rle:

Code: Select all

15b2o$16bo$12bo3b3o2$14bo2$16bo3$6bo$6b3o14b2o$9bo13bo$8b2o11bobo$21b
2o6$5bo$4bobo18b2o$4b2o19b2o9$24b2o$24b2o12$o$o22b2o$3o20bo$2bo10b2o3b
2o4b3o$11bo2bo4bo6bo$11b3o4bo$18b2o$11b2obo$11bob2o!

ulx159x.rle:

Code: Select all

16b3o$17bo$17b3o2$15b6o$15b6o$15b5o$14b6o$15b6o$14b7o$14b6o$10bo3b6o$
9b2o3b6ob2o$9b11ob3o$11b11o$10b11o$9b12o$9b15o$9b15o$7bob16obo$7b21o$
8b20o$10b15obo$10b14o$9b13o$10b12o$11b10o$10b12o$9b14o$9b17o$10b17o$
10b17o$9b17o$6b16o$5b16o$5b16o$4b16o$3b20o$3b20o$4b20o$3b22o$b24o$24o$
ob13ob10o$obob4ob7o2b2o4b2o$o3b4o2b3obo4bo4bo$3ob4o4b4o3b2o$7o4b2o2bo$
12b3o!

pairs data:

Code: Select all

VL214P8-VLX159
VL214P8-VLX159X

VLX159-VL106P6
VLX159-VLX65P3X1
VLX159-VLX65P8X1
VLX159-VLX65P4X1
VLX159-VLX65P5X1
VLX159-VLX65P7X1
VLX159-VR64E3
VLX159-RX140
VLX159-RX140X
VLX159X-VLX84P15

I suspect that F216P6 and Fx177P6 cannot follow VLX159X, but I wasn't sure, so I did not exclude them in the pairs data.

[dvgrn]

You seem to have added 'V' in front of the conduit names for the disallowed following conduits. This will only disallow a specific variant of a conduit (in this case, usually the standard variant). By entering the conduit name rather than the variant name, all variants of that conduit are forbidden. I did it this way to save myself the effort of entering each conduit variant name. For some reason, this is not done anywhere in the rest of the pairs list, even though the comments say that it works.

Thanks for all this, and the Lx159. I'll probably roll it all into a new build over the weekend and do some more testing.

I switched to the standard variant names because when I tried the V-less conduit names in the forbidden-pairs list, Hersrch started throwing errors. I'll give it another try when I get a chance, and maybe look in the code to see how (or if!) the component-name support is actually implemented.

In other news, I just finally noticed last night that the new Bx198 + F166 (i.e., Lx134 + R64 + F166 -- where did that first get posted, by the way?) allows for a new Spartan Herschel splitter, _and_ one of the branches is an adjustable transceiver. Repeat time is 316 ticks, not much higher than the Bx198 (295 ticks) and well below the fastest stable G-to-H converter.

Has anybody put one of these Bx198 transceivers together already, and I'm just way behind the times? We've never had a really good Spartan transmitter, just lots of receivers. The transmitters have always been non-Spartan, or had various technical limitations that made them hard to use.

This one is still kind of big, and the input comes in from a weird direction. But it's not _too_ cramped -- and it's a Spartan Herschel splitter, too, and how much compactness can you really expect? So this is big news... at least for me!

I haven't quite figured out how to get a signal into Guam's Spartan Herschel-less B-heptomino conduit yet -- that's the unfinished part on the right of the pattern -- but once the signal is in there, there are certainly lots of intersecting gliders to work with to split off more Herschel signals!

Code: Select all

#C Bx198+F166 Spartan Herschel splitter and transceiver,
#C hooked up to a standard Spartan glider-to-Herschel converter,
#C with other experimental conduits hung off the right side.
#C B-heptomino conduit input doesn't recover correctly yet.
x = 413, y = 246, rule = LifeHistory
201.2A52.2A54.2A$200.A2.A50.A2.A52.A2.A$201.2A52.2A54.2A16$212.2A30.
2A54.2A$212.A.A28.A.A53.A.A$214.A28.A55.A$214.2A26.2A54.2A7$204.2A46.
2A54.2A$195.2A7.2A46.2A7.2A45.2A7.2A$196.A64.A55.A$196.A.A60.A.A53.A.
A$197.2A60.2A54.2A7$48.A151.A56.A55.A$48.3A148.A.A54.A.A53.A.A$51.A
147.2A56.2A54.2A$50.2A5.2A$57.2A3$59.2A165.2A2.2A54.2A$59.2A165.A4.A
55.A$72.2A150.A.A4.A.A53.A.A$72.A151.2A6.2A54.2A$70.A.A$23.A46.2A$21.
3A$20.A$20.2A$81.2A140.2A8.2A54.2A$81.A141.2A8.2A54.2A$79.A.A137.2A
16.2A54.2A$79.2A138.2A16.2A54.2A$205.2A44.2A54.2A$61.2A142.2A44.2A54.
2A$30.2A29.2A$30.2A36.2A$3.A63.A2.A$2.A.A63.2A$2.A.A37.A211.A55.A$3.A
37.A.A149.2A58.A.A7.2A44.A.A7.2A$42.2A149.2A58.A.A7.2A44.A.A7.2A$33.
2A180.2A24.2A11.A42.2A11.A$34.A180.A26.A55.A$31.3A182.3A20.3A53.3A$
31.A186.A20.A55.A$68.2A143.2A28.2A54.2A$68.2A143.2A28.2A54.2A11$75.2A
275.A$75.2A85.A187.3A$162.3A184.A$7.2A60.2A86.A7.A8.A38.A16.A38.A16.A
38.A16.A6.2A$7.2A40.2A19.A86.3A4.2A7.A.A20.2A15.3A13.A.A20.2A15.3A13.
A.A20.2A15.3A13.A.A20.2A$49.A.A18.A.A87.A13.A20.A.A18.A13.A20.A.A18.A
13.A20.A.A18.A13.A20.A.A$51.A19.2A75.2A9.2A35.A7.2A9.2A35.A7.2A9.2A
35.A7.2A9.2A35.A$.2A48.2A96.A55.A55.A55.A$.2A146.A.A53.A.A53.A.A53.A.
A$5.2A143.2A54.2A54.2A54.2A$5.2A2$26.2A$26.2A26.2A$2A52.2A40.2A$2A94.
A$97.3A49.2A34.2A18.2A34.2A18.2A34.2A18.2A34.2A$99.A21.2A26.2A34.2A
11.2A5.2A34.2A11.2A5.2A34.2A11.2A5.2A34.2A11.2A$121.2A75.2A54.2A54.2A
54.2A$88.2A288.2A$88.2A288.A.A$64.2A314.A$65.A314.2A$49.2A11.3A74.2A$
49.2A11.A75.A2.A28.2A54.2A54.2A54.2A71.A$139.A.A15.2A11.2A5.2A34.2A
11.2A5.2A34.2A11.2A5.2A34.2A11.2A5.2A63.A.A$140.A16.2A18.2A34.2A18.2A
34.2A18.2A34.2A18.2A63.A.A$35.2A374.A$36.A346.2A$33.3A92.2A253.2A$33.
A94.2A$6.2A13.2A22.2A$5.A.A13.A.A21.A$5.A17.A22.3A129.2A54.2A54.2A54.
2A$4.2A17.2A23.A128.A.A53.A.A53.A.A53.A.A$177.A55.A55.A55.A$168.A7.2A
9.2A35.A7.2A9.2A35.A7.2A9.2A35.A7.2A9.2A36.2A$115.2A50.A.A18.A13.A20.
A.A18.A13.A20.A.A18.A13.A20.A.A18.A13.A23.A$20.A94.2A51.2A15.3A13.A.A
20.2A15.3A13.A.A20.2A15.3A13.A.A4.2A14.2A15.3A13.A.A4.2A13.3A$19.A.A
97.2A64.A16.A38.A16.A38.A16.A5.A32.A16.A5.A14.A$20.2A96.A.A200.3A53.
3A$118.A204.A55.A$26.2A89.2A56.A$26.A103.2A43.3A$24.A.A103.2A46.A$24.
2A151.2A2$44.A87.2A$34.2A6.3A87.2A$34.2A5.A135.A$41.2A133.A.A$4.2A
171.2A10.2A$4.2A183.2A$40.2A199.2A54.2A$35.2A3.2A199.2A54.2A$35.2A
209.A55.A$7.2A162.2A71.3A53.3A$7.2A162.2A70.A55.A$163.A18.2A59.2A54.
2A$54.A108.3A10.2A4.A38.2A54.2A$45.A6.3A11.A99.A9.A6.3A35.2A54.2A$43.
3A5.A14.3A96.A.A9.3A5.A$42.A8.2A16.A14.A81.A12.A$16.2A24.2A24.2A12.3A
$16.A64.A117.2A$17.3A61.2A84.2A30.A$19.A147.2A28.A.A33.2A54.2A$155.A
37.2A2.2A34.2A54.2A$80.2A47.A9.A15.3A35.2A108.2A$61.2A17.2A47.3A5.3A
18.A144.2A$61.2A69.A3.A20.2A11.2A135.2A$131.2A3.2A32.2A135.2A4$46.2A$
46.A17.2A$47.3A15.A186.2A54.2A$49.A12.3A187.A.A53.A.A$62.A82.2A52.2A
53.A55.A$67.2A76.2A34.2A16.A.A52.2A54.2A$68.A64.2A45.A.A18.A35.2A$65.
3A64.A2.A44.A20.2A35.A$65.A61.2A4.2A44.2A4.2A48.3A$126.A.A55.A.A48.A$
126.A57.A98.2A$125.2A56.2A7.2A89.A.A$83.2A50.2A33.2A20.2A90.A$83.A.A
49.A34.A$58.2A25.A50.3A32.3A$59.A25.2A51.A34.A$59.A.A15.2A64.2A$60.2A
15.2A65.A$141.3A$141.A60.2A77.2A$202.A77.A.A$200.A.A77.A$200.2A77.2A
7.2A$288.2A3$182.A$182.3A$185.A$78.2A104.2A$78.A.A217.2A$80.A217.A$
80.2A214.A.A$296.2A6$204.2A$204.2A$68.2A$68.2A$56.2A$55.A.A$55.A$54.
2A2$79.2A$58.2A19.A109.2A94.2A$59.A17.A.A109.2A93.A2.A$59.A.A15.2A
206.2A$60.2A4.A$65.A.A$65.A.A$66.A10.2A$77.A.A$79.A$79.2A$64.2A$65.A$
62.3A$62.A22$99.3A$99.A$100.A!

[Sokwe]

I switched to the standard variant names because when I tried the V-less conduit names in the forbidden-pairs list, Hersrch started throwing errors. I'll give it another try when I get a chance, and maybe look in the code to see how (or if!) the component-name support is actually implemented.

It certainly works for me, but since there seems to be an issue, here is that data using variant names:

Code: Select all

# VL106P6-Bx167: 38b2o$22b2o14bo$11b2o9b2o12bobo$11b2o23b2o2$2b2o$2b2o$38b2obo$38bob2o2$2b2o39b2o$2b2o38bobo$38b2o2bo$38bobob2o$8bo31bo$39b2o4bo2bo$7bo37bo$5bo6b2o30bo4bo$12b2o29bobo2b2o$4bo36b2obo$15b2o$3o12b2o3bo$2bo16bobo19bo2bo$2b2o16bobo20b2o$22bo14b2o$22b2o13bo$35b3o7$44b2o$25bo18bobo$25bo20bo$25b3o18b2o$27bo11$25bo2bo$25bo$24bo4bo$23bobo2b2o$21b2obo$35b2o$35b2o2b2o$21bo2bo8b2obo2b2o$23b2o8b3o3$34b2o$34b2o!
VL106P6-VBX125
VL106P6-VBX222
VL106P6-VBX222W1
VL106P6-VBX222W2
VL106P6-VF117
VL106P6-VF171
VL106P6-VF171P2
VL106P6-VF171P2X
VL106P6-VF171P3
VL106P6-VF171P3X
VL106P6-VF171P5
VL106P6-VF171P5X
VL106P6-VF171P7
VL106P6-VF171P7X
VL106P6-VF171P15
VL106P6-VF349P15
VL106P6-VF349P15X
VL106P6-VFX77
VL106P6-VFX77S
VL106P6-VFX77SW
VL106P6-VFX78P8
VL106P6-VFX78P8B
VL106P6-VFX158
VL106P6-VFX176
VL106P6-VFX176X
VL106P6-VFX349P15
VL106P6-VFX349P15X
VL106P6-VL112
VL106P6-VL156
# VL106P6-L198: 59b2o$43b2o14bo$43b2o12bobo$57b2o4$59b2obo$59bob2o$o$3o61b2o$2bo15b2o43bobo$2bo15b2o39b2o2bo$59bobob2o$61bo$30bo29b2o4bo2bo$29bobo34bo$30b2o33bo4bo$21b2o41bobo2b2o$22bo39b2obo$21bo$21b2o$62bo2bo$64b2o$58b2o$58bo$56b3o7$65b2o$46bo18bobo$46bobo18bo$46b3o18b2o$48bo11$46bo2bo$46bo$45bo4bo$44bobo2b2o$42b2obo$56b2o$56b2o2b2o$42bo2bo8b2obo2b2o$44b2o8b3o3$55b2o$55b2o!
# VL106P6-L255: 30b2o24b2o$30b2o8b2o14bo$20bo19b2o12bobo$20b3o31b2o$16bo6bo$16b3o3b2o$11bo7bo$9b3o6b2o36b2obo$8bo47bob2o$8b2o$61b2o$60bobo$56b2o2bo$56bobob2o$o57bo$3o54b2o4bo2bo$2bo15b2o43bo$2bo15b2o42bo4bo$61bobo2b2o$59b2obo$30bo$29bobo$30b2o27bo2bo$21b2o38b2o$22bo32b2o$21bo33bo$21b2o30b3o7$62b2o$43bo18bobo$43bobo18bo$43b3o18b2o$45bo11$43bo2bo$43bo$42bo4bo$41bobo2b2o$39b2obo$53b2o$53b2o2b2o$39bo2bo8b2obo2b2o$41b2o8b3o3$52b2o$52b2o!
VL106P6-VLX65P3
VL106P6-VLX65P3X1
VL106P6-VLX65P3X2
VL106P6-VLX65P3
VL106P6-VLX65P3X1
VL106P6-VLX65P3X2
VL106P6-VLX65P8
VL106P6-VLX65P8X1
VL106P6-VLX65P8X2
VL106P6-VLX65P4
VL106P6-VLX65P4X1
VL106P6-VLX65P4X2
VL106P6-VLX65P5
VL106P6-VLX65P5X1
VL106P6-VLX65P5X2
VL106P6-VLX65P7
VL106P6-VLX65P7X1
VL106P6-VLX65P7X2
# VL106P6-Lx73: 25b2o$24bo2bo$24bobobo$23b2obobo$25bobob2o$20b2o2b2obo2bo$26bobobob2o$16b2o5b2obob3o2bo$16bo7bo2bo3b2o$14b3o7bob5o$25bo4bo$26b4o$28bo$30bo$29b2o2$23b2o$4bo18bobo$4bobo18bo$4b3o18b2o$6bo11$4bo2bo$4bo$3bo4bo$2bobo2b2o$2obo$14b2o$14b2o2b2o$o2bo8b2obo2b2o$2b2o8b3o3$13b2o$13b2o!
VL106P6-VLX138
VL106P6-VLX138X
VL106P6-VLX142
# VL106P6-Lx159: 20b2o24b2o$20b2o8b2o14bo$10bo19b2o12bobo$10b3o31b2o$13bo$12b2o2$46b2obo$2bo43bob2o$2bo$3o48b2o$o49bobo$46b2o2bo$46bobob2o$48bo$2o45b2o4bo2bo$2o3bo47bo$4bobo4b2o39bo4bo$5bobo2bobo38bobo2b2o$7bo2bo38b2obo$7b4o9bo$19bobo$9b2o9b2o27bo2bo$9b2o40b2o$45b2o$45bo$43b3o7$52b2o$33bo18bobo$33bobo18bo$33b3o18b2o$35bo11$33bo2bo$33bo$32bo4bo$31bobo2b2o$29b2obo$43b2o$43b2o2b2o$29bo2bo8b2obo2b2o$31b2o8b3o3$42b2o$42b2o!
VL106P6-VLX212
VL106P6-VLX212X
VL106P6-VLX318P15
VL106P6-VR190
VL106P6-VR190X
VL106P6-VR318P15

Also, here's the data for L255:

Code: Select all

L255 L 255 10 -54

  # Standard variant, with eater
  VL255 1 130 0 GRU data/vl255.rle 1 6 data/ul255.rle 2 3

  # Standard variant, without eater (no glider)
  VL255X 1 130 0 GRX data/vl255x.rle 1 6 data/ul255x.rle 2 3

vl255.rle:

Code: Select all

11b2o$11bo$9b3o6$17b2o$17bobo$19bo$19b2o5$21b2o$21bo$9b2o8bobo$9b2o8b
2o$23b2o$2obo19bo$ob2o17bobo$21b2o6$5bo$4bobo18b2o$4b2o19b2o9$24b2o$
24b2o12$o$o22b2o$3o20bo$2bo10b2o3b2o4b3o$11bo2bo4bo6bo$11b3o4bo$18b2o$
11b2obo$11bob2o!

ul255.rle:

Code: Select all

11b3o$12bo$10b3o2$9b7o$9b8o$10b8o$10b8o$9b6o2bo$9b7o$10b6o$10b6o$10b6o
$9b8o$8b8o$8b9o$8b9o$7b10o$7b10o2b2o$6b12ob3o$6b14o$4b16o$4bo2b13ob2o$
8b12ob3o$9b13o$9b12o$8b13o$9b15o$9b15o$7bob16obo$7b21o$8b20o$10b15obo$
10b14o$9b13o$10b12o$11b10o$10b12o$9b14o$9b17o$10b17o$10b17o$9b17o$6b
16o$5b16o$5b16o$4b16o$3b20o$3b20o$4b20o$3b22o$b24o$24o$ob13ob10o$obob
4ob7o2b2o4b2o$o3b4o2b3obo4bo4bo$3ob4o4b4o3b2o$7o4b2o2bo$12b3o!

vl255x.rle:

Code: Select all

11b2o$11bo$9b3o3bo2$13bo2$11bo10$21b2o$21bo$9b2o8bobo$9b2o8b2o$23b2o$
2obo19bo$ob2o17bobo$21b2o6$5bo$4bobo18b2o$4b2o19b2o9$24b2o$24b2o12$o$o
22b2o$3o20bo$2bo10b2o3b2o4b3o$11bo2bo4bo6bo$11b3o4bo$18b2o$11b2obo$11b
ob2o!

ul255x.rle:

Code: Select all

11b3o$12bo$10b3o2$9b6o$9b6o$10b5o$10b6o$9b6o$9b7o$10b6o$10b6o$10b6o$9b
8o$8b8o$8b9o$8b9o$7b10o$7b10o2b2o$6b12ob3o$6b14o$4b16o$4bo2b13ob2o$8b
12ob3o$9b13o$9b12o$8b13o$9b15o$9b15o$7bob16obo$7b21o$8b20o$10b15obo$
10b14o$9b13o$10b12o$11b10o$10b12o$9b14o$9b17o$10b17o$10b17o$9b17o$6b
16o$5b16o$5b16o$4b16o$3b20o$3b20o$4b20o$3b22o$b24o$24o$ob13ob10o$obob
4ob7o2b2o4b2o$o3b4o2b3obo4bo4bo$3ob4o4b4o3b2o$7o4b2o2bo$12b3o!

pairs data:

Code: Select all

VL214P8-VL255
VL214P8-VL255X

Edit: I accidentally used a full Herschel in vlx159x.rle instead of a ghost Herschel (see earlier post). Use this instead for vlx159x.rle:

Code: Select all

15b2o$16bo$12bo3b3o2$14bo2$16bo3$6bo$6b3o14b2o$9bo13bo$8b2o11bobo$21b
2o6$5bo$4bobo18b2o$4b2o19b2o9$24b2o$24b2o12$o$o22b2o$3o20bo$2bo10b2o3b
2o4b3o$11bo2bo4bo6bo$11b3o4bo$18b2o$11b2obo$11bob2o!

Also, F216P6 and Fx177P6 can follow VLX159X, like so:

Code: Select all

x = 44, y = 95, rule = B3/S23
22b2o$23bo$23b3o6$16b2o$15bobo$15bo$14b2o7$24b2o$24b2o2$31bob2o$31b2ob
o$8b2o$7bo2bo$6bobobo$6bobob2o$4b2obobo3bo21b2o$4bo2bo2b2obo21bo$b2obo
bobo4bo23bo$bo2b2o2bob2o24b4o$2b2o3bo2bo24bo4bo$4b2ob2obo23bobobobo$4b
o4bo24bobo4b2o$5b4o24b2o4b2o2bo$32bo2bobo2bob2o$7b2o22b3o3bo2bo$6bobo
23bo2bobob2o$6b2o25b2obobo$34bobobo$4b4o26bo2bo$3bo4bo26b2o$3b2ob2obo$
b2o3bo2bo6b2o$o2b2o2bob2o6bo$2obobobo4bo4b3o$3bo2bo2b2obo$3b2obobo3bo$
5bobob2o$5bobobo$4b2obo2bo$6bob2o$6bo$7b3o14b2o$10bo13bo$9b2o11bobo$
22b2o6$6bo$5bobo18b2o$5b2o19b2o9$25b2o$25b2o12$bo$bobo20b2o$b3o20bo$3b
o10b2o3b2o4b3o$12bo2bo4bo6bo$12b3o4bo$19b2o$12b2obo$12bob2o!

[dvgrn]

I haven't quite figured out how to get a signal into Guam's Spartan Herschel-less B-heptomino conduit yet...

Aha, never mind -- I see the new receiver-R-R-B conduit will do the trick, as it does in Guam's slower stable Herschel splitter. I'm still catching up on some of the new discoveries from back in January... This doesn't quite get me anything that I can add to Hersrch's Spartan toolkit, though -- or the Spartan++ toolkit, since Guam has made a convincing case for including tub-with-tail eaters for Gemini-spaceship-style glider-pair constructions.

Might there be a workable glider-constructible version of the Herschel-less B conduit, or have the relevant searches already been run? Even just playing around by hand, I can find various catalysts that will throw a little junk in the way of that awkward first output glider -- but so far I need an extra glider from elsewhere to make it totally clean:

Code: Select all

x = 83, y = 126, rule = B3/S23
39b2o$39b2o3$37b2o$37b2o$24b2o$25bo$25bobo$26b2o$22b2o51b2o$22b2o50bob
o$75bo6$31b2o2b2o$30bobo2b2o$32bo2$64b2o$46b2o16b2o11b2o$45bobo29b2o$
45bo$44b2o4$28b2o$28b2o26b2o$56b2o7$57b2o$56bobo$56bo$55b2o9b2o$67bo
13bo$64b3o4b2o7bobo$64bo7bo8bo$69b3o$69bo13$39b2o$39b2o3$37b2o$37b2o$
24b2o$25bo$25bobo$26b2o$22b2o51b2o$22b2o50bobo$75bo6$31b2o2b2o$30bobo
2b2o$32bo2$64b2o$64b2o11b2o$77b2o6$28b2o$28b2o26b2o$56b2o7$57b2o$56bob
o$56bo$55b2o9b2o$67bo13bo$64b3o4b2o7bobo$64bo7bo8bo$69b3o$69bo16$b2o$o
bo$2bo!

No doubt another addition to the toolkit should be dead spark coils, assuming they're about as easy to construct as tub-with-tail eaters (and there are only two orientations). Spark coils are only handy for one catalysis that I know of, but the G5 Herschel transmitter is so useful that it's almost certainly worth including it. Here's Guam's Herschel-merge circuit as a Spartan++ pattern, for example:

Code: Select all

x = 187, y = 185, rule = B3/S23
38b2o$37bobo$37b2o3$32bo$31bobo$31bobo$32bo$41b2o$41b2o13$18b2o35b2o$
18b2o35bo$56b3o$58bo6$23b2o$22bobo$22bo$21b2o2$41bo$40bobo$41b2o14$17b
2o$17b2o$21b2o$21b2o$35b2o$35b2o5$38bo$37bobo7b2o53b2o$37bobo7b2o52bob
o$25b2o11bo62b2o$26bo$23b3o$23bo72bo$27b2o66bobo$27b2o40b2o24bobo$69b
2o25bo$105b2o$105b2o2$78b2o$78bo$76bobo$21bo54b2o$13bo6bobo$13b3o5bo$
9b2o5bo$10bo4b2o$10bobo$11b2o2$55b2o25b2o35b2o$55b2o25b2o35bo$7b2o25b
2o84b3o$8bo25b2o86bo$8bobo$9b2o3$22b2o$22bobo6b2o54b2o$24bo6bo54bobo$
24b2o6b3o51bo$34bo50b2o2$105bo$104bobo$105b2o$10b3o7b2o$11bo8bobo$11b
3o8bo$22b2o6$172b2o$2b2o167bo2bo$bobo168b2o$bo$2o79b2o$81b2o$85b2o$85b
2o$99b2o$99b2o5$102bo67b2o$101bobo7b2o58bo$101bobo7b2o55b3o$89b2o11bo
65bo$90bo$87b3o$87bo$91b2o$91b2o4$127b2o$127b2o5b2o$134b2o39b2o$166b2o
7b2o$167bo$132b2o33bobo$132b2o34b2o$138b2o44b2o$138b2o44bo$182bobo$
182b2o2$118bo$118bobo$118b3o$120bo64bo$184bobo$185bo2$142b2o$111b2o29b
o$112bo27bobo$109b3o28b2o$109bo73b2o$165b2o16b2o$165b2o4$176b2o$176b2o
$124b2o3b2o$124bobobobo$126bobo50b2o$124bobobobo4b2o43bo$124b2o3b2o5bo
43bobo$135bo45b2o$135b2o3$173b2o$173b2o$155b2o$156bo$153b3o$153bo!

Last and least, a very late thought about a comment from back in January:

I think {the Herschel-less B conduit} is the first repeatable stable conduit that does not require a Herschel at some point.

What about a chain of F166's or Lx200's? You could say that those use Herschels, just three ticks back... but it seems to me then you could also say a B-heptomino is just a Herschel 20 ticks back, or an R-pentomino is a Herschel 48 ticks back. And then there's David Bell's swimmer track, which is a stable conduit for switch engines. It would definitely be good to have better S->H and H->S technology, though...!

[Guam]

G4 Herschel transmitter has repeat time 319, but I fount two spartan Herschel transmitters:
G2 (repeat time is 232):

Code: Select all

x = 97, y = 114, rule = B3/S23
76b2o$76b2o2$64b2o$64b2o19b2o$85b2o5$58b2o$57bo2bo$58b2o8$81b2o$71b2o
8bo$72bo6b3o$69b3o$69bo14$79b2o$79b2o2$73bo$71b3o$70bo$70b2o$90b2o$90b
o$88bobo$88b2o3$74b2o$73bobo$73bo$72b2o7$77b2o$78bo$75b3o$75bo14$85b2o
$85b2o2$76b2obo$76bob2o$94b2o$61b2o31bo$14bo46bo30bobo$6bo6bobo43bobo
30b2o$6b3o5bo44b2o$2b2o5bo67bo$3bo4b2o67b3o$3bobo62bo11bo$4b2o62b3o8b
2o14bo$71bo22bobo$46bo23b2o23bo$44b3o$2o25b2o15bo$bo25b2o15bo$bobo$2b
2o2$90b2o$15b2o73bobo$15bobo6b2o32b2o32bo$17bo6bo20b2o11b2o23b2o7b2o$
17b2o6bo20bo36bo$24b2o17b3o38b3o$43bo42bo4$3b3o7b2o$4bo8bobo$4b3o8bo$
15b2o!

G6 (repeat time is 242):

Code: Select all

x = 100, y = 72, rule = B3/S23
83b2o$83bo$81bobo$81b2o7$86b2o$86b2o2$98b2o$98bo$96bobo$96b2o$59b2o$
58bobo$58bo$57b2o3$67b2o$67b2o2$82b2o$72b2o8bo$73bo6b3o$70b3o$70bo4$
88b2o$88bobo$90bo$90b2o5$61b2o$14bo46bo$6bo6bobo43bobo18b2o$6b3o5bo44b
2o10b2o7b2o$2b2o5bo62bo$3bo4b2o62bobo$3bobo61b2o4b2o$4b2o62bo20b2o$68b
obo18bo$46bo22b2o16bobo$44b3o40b2o$2o25b2o15bo$bo25b2o15bo$bobo$2b2o3$
15b2o$15bobo6b2o32b2o$17bo6bo20b2o11b2o$17b2o6bo20bo$24b2o17b3o35b2o$
43bo37b2o2b2o$85bobo$87bo$87b2o$3b3o7b2o$4bo8bobo$4b3o8bo$15b2o!

[dvgrn]

G4 Herschel transmitter has repeat time 319, but I fount two spartan Herschel transmitters...

Interesting! If I remember correctly, though, that wonderful L156 variant (at the beginning of both of these transmitters) was the most troublesome of all Spartan conduits to construct with gliders -- the eaters are really close together at the corner. Regular L156 catalysts are much better spaced. The L156b variant was deliberately left out of the prototype construction arm, on the grounds that it was more trouble than it was worth, and so it didn't show up anywhere in Andrew Wade's Gemini spaceship either.

But here again, Life technology may have improved by now! This is a case like tub-with-tail eaters or dead spark coils where this old superstition should quite possibly be dropped and forgotten, as long as that tightly packed L156b constellation can be easily constructed in any orientation. My memory isn't clear on this, but Adam Goucher wrote a program to check for trivial Gemini glider-pair constructibility of all orientations of an input pattern, and I seem to recall there was some trouble with the L156b...?

There's a relatively small G5 Spartan++ Herschel splitter/transmitter -- needs that pesky tub-with-tail eater somewhere along the output lane. But the recovery time is way up at 489 ticks, slightly higher than even an old standard G-to-H converter. Could be very useful for cases where 500-tick recovery is good enough, though:

Code: Select all

x = 75, y = 99, rule = B3/S23
60b2o$60bo$58bobo$58b2o4$40bo$40b3o$43bo6b3o$42b2o8bo$51b3o6$37b2o$38b
o$38bobo$39b2o3$53b2o$53bobo$55bo$55b2o3$37b2o$36bobo$36bo$35b2o7b2o$
44b2o5$69bo$67b3o$66bo$65bobo$66bo3$36b2o26b2o$37bo26b2o$37bobo$38b2o
2$66b2o$66bo$64bobo$60b2o2b2o$60b2o$19b2o$20bo$20bobo$21b2o36b2o$59b2o
4b2o$65b2o4$69b2o$69bo$67bobo$67b2o10$35bo$35b3o35b2o$38bo34b2o$37b2o
3$68b2o$68b2o$72b2o$72b2o2$28b2o$28b2o36b2o$bo49b2o13b2o$obo48bobo$obo
50bo$bo39b2o10b2o$41bo$31b2o9b3o$32bo11bo$29b3o$29bo!

The only reason I'm still paying attention to the strict old definition of "Spartan" is that I'm still pretty interested in building a Geminoid spaceship with just a single construction/destruction arm instead of three. It's much harder to build things like tub-with-tail eaters using only unidirectional slow salvos... actually, it's much harder to build _anything_ with unidirectional slow salvos, but it can definitely be done.

The big advantage is that the new Geminoid replicator unit can be *tiny*. I've built one with 150 still lifes, 886 ON cells in a 287x348 box, and can probably cut it down more if I keep trying. I'll probably be starting a new Geminoid spaceship thread at some point -- seems like there's a good series of optimization challenges with putting together a one-arm Geminoid spaceship or replicator, that various people on the forums might be interested in tackling.

[Sokwe]

What about a chain of F166's or Lx200's? You could say that those use Herschels, just three ticks back... but it seems to me then you could also say a B-heptomino is just a Herschel 20 ticks back, or an R-pentomino is a Herschel 48 ticks back. And then there's David Bell's swimmer track, which is a stable conduit for switch engines.

I may have been reaching a bit with my comment. I would say that usually, the three-tick Herschel predecessor can be considered a Herschel, as it cleanly evolves into one (unlike the R or B). The only thing that complicates this is Fx158, which has that out-of-nowhere Herschel that doesn't evolve through the normal (B-based) pre-Herschel stages. I just completely forgot about the swimmer track.

Here is the data for one of the first conduits found by Guam, L309:

Code: Select all

L309 L 309 43 -47

  # Standard variant
  VL309 1 134 0 GR data/vl309.rle 3 7 data/ul309.rle 2 5
  
  # Variant with first block-suppressing eater missing
  VL309M 1 134 0 GR data/vl309m.rle 4 7 data/ul309.rle 2 5

vl309.rle:

Code: Select all

46b2o$46bo$44b3o16$44b2o$44b2o2$35b2obo$35bob2o$53b2o$53bo$51bobo$51b
2o$18b2o$19bo$19bobo$14b2o4b2o$6bo7b2o38bo$6b3o44bobo$9bo44bo$8b2o2$2o
$bo$bobo$2b2o44bo6b2o$47bobo5b2o$47bobo$48bo3$57bo$2bo53bobo$2bo54b2o$
2b3o$4bo19b2o$25bo23bo$14b2o6b3o22b3o$14b2o2b2o2bo2b3o18bo$18bobobobo
2bo18b2o$20bobob2o$20b2o!

ul309.rle:

Code: Select all

44b3o$45bo$43b3o2$42b6o$42b6o$43b5o$43b6o$42b6o$42b7o$43b6o$43b6o$43b
6o$42b8o$41b8o$41b9o$41b9o$40b10o$40b10o$39b11o$39b11o$37b13o$37bo2b8o
$41b7o$42b6o$42b6ob2o$41b7ob3o$28bo13b8o$27b3o12b7o$27b3o12b7o$13b2o6b
o4b5o10b7o4bo$12b4o5b3ob7o10b6obob3o$13b3o5b17o3b12obo$12bobo6b17o3b
10ob2o$11b6o4b30o$8b9o2b32o$9b42o$9b43o$2b2o6b42ob2o$b3ob3o2b46o$3b53o
$3b36o2b12obo$4b35o3b11o$3b36o4b11o$b38o6b8o$18o3b17o7b3ob5ob2o$ob15o
5b10o3bo10bob8o$obob4ob7o7b8o16b8o$o3b4o2b6o7b7o18b6o$3ob4o3b6o6b2o24b
3o$7o6b4o7bo23bo$12b5o$13b2ob3o$16b2o!

vl309m.rle:

Code: Select all

47b2o$47bo$45b3o16$45b2o$45b2o2$36b2obo$36bob2o$54b2o$54bo$52bobo$52b
2o$19b2o$20bo$20bobo$15b2o4b2o$7bo7b2o38bo$7b3o44bobo$10bo44bo$9b2o5$
49bo6b2o$48bobo5b2o$48bobo$49bo3$58bo$3bo53bobo$3bo54b2o$o2b3o$5bo19b
2o$26bo23bo$15b2o6b3o22b3o$15b2o2b2o2bo2b3o18bo$19bobobobo2bo18b2o$21b
obob2o$21b2o!

pairs data:

Code: Select all

*VL309M

VBX222-VL309

VBX222W2+VL309M

VFX77S-VL309

VFX77SW+VL309M

VL106P6-VL309

VL214P8-VL309

VLX73P8C-VL309

[dvgrn]

I switched to the standard variant names because when I tried the V-less conduit names in the forbidden-pairs list, Hersrch started throwing errors. I'll give it another try when I get a chance, and maybe look in the code to see how (or if!) the component-name support is actually implemented.

It certainly works for me, but since there seems to be an issue, here is that data using variant names...

Aha -- found the problem, I think, because it showed up again in the pairs data for the Lx159: "VLX159-RX140" and "VLX159-RX140X" throw errors because conduit names are case-sensitive. Hersrch likes "VLX159-Rx140" just fine -- no need for a separate entry for the X variant.

I've caught up with the backlog of Hersrch data again and posted the latest version (18 November 2012), adding Lx159(x), L161p8, Lx216p8(b), L255(x), and L309(m). The p8 ones are just 39-tick shorter variants of Lx200 and L255, using the same trick that shortens F117 into Fx78p8 and Fx153 into F114p8. I haven't checked all the new conduits thoroughly yet, but I _think_ I've gone through all the old ones, and Lx200 was the last one where the p8 reflector will fit.

I did some very basic tests, and all the new conduits do show up in search results. Let me know if any problems appear. I probably still won't have time to do a thorough review of forbidden pairs for a while yet.

Next I'd like to claim Lx198, Lx159p8 and Rx_162. Lx198 interacts with a 6-tick Herschel predecessor, but I think Hersrch should be able to handle it just fine. Lx159p8 is the standard p8 reduction of L198 (which is a different conduit from Lx198, I have to keep reminding myself...) Rx_162 is Guam's simple but magical cleanup of a spare beehive in an old conduit of Dietrich Leithner's (which was used as part of an early G-to-LWSS converter as well as a period doubler, by the way.) I'd rather not admit how long I spent trying to clean up that conduit, years ago, without ever thinking of adding a tub-with-tail eater right there...!

Any other new conduits are fair game -- I promise not to work on any of them until these three are done!

[Sokwe]

Aha -- found the problem, I think, because it showed up again in the pairs data for the Lx159...

Oops, sorry.

@Dave
There are some mistakes in the data for Lx216p8. Here is corrected data for this conduit:

Code: Select all

Lx216p8 Lx 216 10 -41

  # p8 slow variant (requires glider for reflector)
  VLX216P8 8 168 0 GRU data/vlx216p8.rle 6 6 data/ulx216p8.rle 8 3

  # p8 fast variant (requires glider for reflector)
  VLX216P8B 8 152 0 GRU data/vlx216p8b.rle 9 6 data/ulx216p8b.rle 9 3

vlx218p8.rle:

Code: Select all

15b2o$16bo$2b2o5b2o5b3o$3bo5b2o15b2o$3bobo8bo11bo$4b2o18bobo$8b2o2bo
11b2o$7bobo18b2o$8bo19bo$26bobo$3b3o20b2o$3b3o$3b3o$3o$3o$3o$10bo$9bob
o18b2o$9b2o19b2o9$29b2o$29b2o12$5bo$5bo22b2o$5b3o20bo$7bo10b2o3b2o4b3o
$16bo2bo4bo6bo$16b3o4bo$23b2o$16b2obo$16bob2o!

ulx218p8.rle:

Code: Select all

17b3o$18bo$11b2o5b3o$11b4o$7b2o3b10o$6b3ob12o3b2o$8b16ob3o$8b18o$6b2ob
17o$5b3o3b15ob2o$5b6ob14ob3o$4b6o3b15o$3bob4o5b13o$b4obo7b13o$6o9b15o$
5o11b14o$2b3o7b3ob15obo$2b2o7bob2ob18o$11b4ob18o$13b2ob15obo$16b14o$
15b13o$16b12o$17b10o$16b12o$15b14o$15b17o$16b17o$16b17o$15b17o$12b16o$
11b16o$11b16o$10b16o$9b20o$9b20o$10b20o$9b22o$7b24o$6b24o$6bob13ob10o$
6bobob4ob7o2b2o4b2o$6bo3b4o2b3obo4bo4bo$6b3ob4o4b4o3b2o$6b7o4b2o2bo$
18b3o!

vlx218p8b.rle:

Code: Select all

18b2o$19bo$19b3o$4b2o5b2o16b2o$5bo5b2o4bo11bo$5bobo19bobo$6b2o7bo11b2o
$10b2o19b2o$9bobo19bo$6bo3bo18bobo$6b2o21b2o$5b2obo$4bo2b3o$3bobobo$2b
obobo$3o2bo$bob2o8bo$2b2o8bobo18b2o$3bo8b2o19b2o9$32b2o$32b2o12$8bo$8b
o22b2o$8b3o20bo$10bo10b2o3b2o4b3o$19bo2bo4bo6bo$19b3o4bo$26b2o$19b2obo
$19bob2o!

ulx218p8b.rle:

Code: Select all

18b3o$19bo$12b2o5b3o$12b4o$13b10o$7b2o2b12o3b2o$6b3ob15ob3o$8b19o$8b
19o$6b2ob18ob2o$5b3o3b16ob3o$5b6ob17o$4b6o3b15o$3bob4o5b14o$b4obo8b16o
$6o10b15o$5o8b3ob15obo$2b3o7bob2ob18o$2b2o8b4ob18o$14b2ob15obo$17b14o$
16b13o$17b12o$18b10o$17b12o$16b14o$16b17o$17b17o$17b17o$16b17o$13b16o$
12b16o$12b16o$11b16o$10b20o$10b20o$11b20o$10b22o$8b24o$7b24o$7bob13ob
10o$7bobob4ob7o2b2o4b2o$7bo3b4o2b3obo4bo4bo$7b3ob4o4b4o3b2o$7b7o4b2o2b
o$19b3o!

There is a lot of pairs data that needs to be added for this conduit (few conduits can follow it), but I don't really want to deal with that right now. For now, just add this:

Code: Select all

VL214P8-VLX216P8
VL214P8-VLX216P8B

Also, the pairs data for R126 seems wrong. You've allowed it to be followed only by 'm' conduits, but R126 does not interact with the block of the output Herschel (you may have been thinking of a variant of Guam's B-->H converter based on this conduit, which does need that extra block). Normal F117-type conduits can follow R126 with no problems:

Code: Select all

x = 29, y = 64, rule = B3/S23
2b2o$3bo$3bobo$4b2o7$21b2o$20bo2bo$5bo15b2o$5bobo$5b3o$7bo$26b2o$26bob
o$27b2o$16bo$14b3o$4b3o6bo$4bo8b2o$3b2o6$18b2o$18bo$16bobo$16b2o3$2b2o
$bobo$bo$2o$20b2o$21bo$18b3o$18bo2$11b2o$11b2o16$10b3o$10bo$9b2o!

The biggest issue with this conduit is the first natural glider of the output Herschel, which travels back through the conduit, like in R64 but with less room. In many cases, a tub-with-tail eater can be placed so that R126 can follow conduits that end with the standard B-->H reaction, as shown by F177-R126-F117:

Code: Select all

x = 66, y = 64, rule = B3/S23
39b2o$40bo$27b2o11bobo$20b2o6bo12b2o$20bobo5bobo$10bo11bo6bobo$10b3o9b
2o6bo3b2o$13bo20b2o$2o10b2o$bo$bobo54b2o$2b2o53bo2bo$25b2o15bo15b2o$
25b2o15bo$42b3o$44bo$63b2o$63bobo$2bo61b2o$2bobo48bo$2b3o46b3o$4bo11b
2o23b3o6bo$16bo24bo8b2o$17b3o20b2o$19bo5$55b2o$55bo$53bobo$53b2o3$39b
2o$38bobo$38bo$37b2o$57b2o$58bo$55b3o$55bo2$48b2o$48b2o16$47b3o$47bo$
46b2o!

With Guam's new method of constructing tub-with-tail eaters, this might be constructable. The only connection that I am aware of that falls under the traditional definition of spartan is Fx77-R126-Lx200:

Code: Select all

x = 67, y = 57, rule = B3/S23
26bo$26b3o$13b2o3b2o9bo12b2o$13b2o3b2o8b2o12bobo$44bo$44b2o$6b2o$7bo$
7bobo$8b2o2$47b2o15bo$47b2o15bo$64b3o$66bo2$11b2o$12bo$9b3o28b2o$9bo
31bo$38b3o$38bo2$10b2o$11bo$11bobo$12b2o10b2o$24b2o2$24b2o$25bo$12b2o
11b3o$11bobo$11bo$o9b2o36b2o$3o44bobo$3bo43b2o$2b2o11b2o11bo$15b2o9b3o
$26bo$26bo15b2o$41bo2bo$42b2o2$bo$bobo$b3o$3bo2$25b2o$24bobo$24bo$23b
2o$16b2o$16bobo$18bo$18b2o!

In many other cases where Lx200 follows R126, a variant of the tub-with-tail eater can be used to clean up two gliders, shown here in Fx153-R126-Lx200:

Code: Select all

x = 91, y = 53, rule = B3/S23
50bo$50b3o$37b2o3b2o9bo14b2o$37b2o3b2o8b2o15bo$68bo$68b2o$30b2o$31bo$
31bobo$32b2o2$71b2o15bo$71b2o15bo$88b3o$90bo2$35b2o$36bo$33b3o28b2o$
33bo31bo$64bo$64b2o$32b2o$33bo2b2o$33bobobo$34bo$48b2o$36bo11b2o$35bob
o$25b2o9bo11b2o$2o24bo6b3o13bo$bo13b2o6b3o7bo15b3o$bobo11b2o6bo$2b2o$
72b2o$71bobo$71b2o$52bo$50b3o$33b2o15bo$2bo30b2o15bo15b2o$2bobo60bo2bo
$2b3o61b2o$4bo3$30b2o$30bo$11b2o3b2o13bo$12bo3bo13b2o17b2o$9b3o5b3o28b
obo$9bo9bo28bo$47b2o!

R126 can also follow F171 if one of the eaters is replaced with an eater 2:

Code: Select all

x = 54, y = 38, rule = B3/S23
11bo15b2o$10bobo15bo$10bobo15bobo$8b3ob2o15b2o$7bo$8b3ob2o$10bob2o$5bo
$5b3o$8bo$7b2o37b2o$45bo2bo$30bo15b2o$4b2o24bo$5bo24b3o$5bobo24bo$6b2o
43b2o$51bobo$52b2o$o$3o$3bo25b3o$2b2o25bo$28b2o6$bo$bobo$b3o$3bo2$10b
2o$11bo$8b3o$8bo!

There may be more tricky connections that I missed, but hopefully this gives an idea of the connective capabilities of R126.

[dvgrn]

There are some mistakes in the data for Lx216p8...

Oops. That conduit was a last-minute addition that I threw in as soon as I discovered it worked with at least one conduit (I tried the F116). Many thanks for digging up the fixes for me.

I'll start dealing with forbidden pairs more systematically as soon as I can find some more time -- maybe next weekend.

Also, the pairs data for R126 seems wrong. You've allowed it to be followed only by 'm' conduits, but R126 does not interact with the block of the output Herschel (you may have been thinking of a variant of Guam's B-->H converter based on this conduit, which does need that extra block).

Yes, I had added the B->G4+H to the sample pattern I posted a while back, and apparently that got me a little confused. I even noticed that that first eater wasn't disturbed in other cases, but still didn't manage to put two and two together.

Well, luckily it's an easy fix, or at least the first draft is... and I still think it was probably worthwhile adding all the 'm' conduits for other reasons.

Any opinion on whether it's worth adding separate R126 variants with tub-with-tail eaters and variations, plus an F171-with-eater2 variant, just so Hersrch can build clean candidate tracks instead of leaving it up to users to know how to fix things? The forbidden-pairs section is already starting to be kind of hard to keep track of and maintain... but it doesn't seem like it will be a big problem as far as Hersrch's search efficiency is concerned.

[Sokwe]

Any opinion on whether it's worth adding separate R126 variants with tub-with-tail eaters?

This is essentially what's done for R64 with the 'E' variants.

I'm not sure about adding a separate variant of F171, since its only purpose would be to precede R126. On the other hand, it shouldn't be assumed that users will know how to make these connections work. For now, perhaps you should continue to add the 'close' pairs as comments with an RLE of the connection.

Edit: I almost think we should add variants of the block+pre-Herschel based dependent conduits with the first glider eater removed, as there are now a number of conduits that rely on it being gone (e.g. Lx142, Bx167x, etc.).

Edit 2: I decided to compile the data for another of Guam's early conduits, F266:

Code: Select all

F266 F 266 53 -5

  # Standard variant
  VF266 1 214 0 GRX data/vf266.rle 3 9 data/uf266.rle 2 7

  # Variant with first block-suppressing eater missing
  VF266M 1 214 0 GRX data/vf266m.rle 4 9 data/uf266.rle 2 7

vf266.rle:

Code: Select all

3bo21bo$bo21bo$5bo21bo8b2o$3bo10b2o9bo10b2o$6bo7b2o13bo15b2o$6b3o18bo
17bo$9bo33bobo$8b2o33b2o2$2o$bo$bobo$2b2o2$55bo$55bo$55b3o$57bo2$2bo$
2bo$2b3o$4bo19b2o$25bo$14b2o6b3o$14b2o2b2o2bo$18bobobo7bo$20bob2o5bobo
9b2o$20bo9bo10bo$19b2o21b3o$44bo!

uf266.rle:

Code: Select all

28bo$27b3o$24b6o5b2o$13b2o10b6o3b4o$12b4o9b7o2b4o$13b3o6b12ob2obo$12bo
bo6b19ob2o$11b6o4b19ob3o$8b9o2b23o$9b32o$9b33o$2b2o6b33o$b3ob3o2b34ob
2o$3b46o2b2o$3b50obo$4b49obobo$3b50ob3o$b52o3bo$18o3b32o$ob15o5b30o$ob
ob4ob7o7b21o2b5o$o3b4o2b6o7b20o$3ob4o3b6o6b2o3b14o$7o6b4o7bo3b13o$12b
5o12b12o$13b2ob3o11b10o$16b2o11bo2bo4b5o$40b2o$40bo!

vf266m.rle:

Code: Select all

4bo21bo$2bo21bo$6bo21bo8b2o$4bo10b2o9bo10b2o$7bo7b2o13bo15b2o$7b3o18bo
17bo$10bo33bobo$9b2o33b2o7$56bo$56bo$56b3o$58bo2$3bo$3bo$o2b3o$5bo19b
2o$26bo$15b2o6b3o$15b2o2b2o2bo$19bobobo7bo$21bob2o5bobo9b2o$21bo9bo10b
o$20b2o21b3o$45bo!

pairs data:

Code: Select all

*VF266M

VBX222W2+VF266M

VF266-VF216P6
VF266-VF216P6X
VF266-VFX177P6
VF266-VFX177P6X
VF266-VLX84P15
VF266M-VF216P6
VF266M-VF216P6X
VF266M-VFX177P6
VF266M-VFX177P6X
VF266M-VLX84P15

VFX77S-VF266

VFX77SW+VF266M

VL106P6-VF266
VL106P6-VF266M

VL214P8-VF266
VL214P8-VF266M

I'll do Fx266 soon, as it is so similar.

Edit 3: Okay, here's the data for Fx266:

Code: Select all

Fx266 Fx 266 53 -9

  # Standard variant
  VFX266 1 214 0 GRX data/vfx266.rle 3 10 data/ufx266.rle 2 7

  # Variant with first block-suppressing eater missing
  VFX266M 1 214 0 GRX data/vfx266m.rle 4 10 data/ufx266.rle 2 7

vfx266.rle:

Code: Select all

44bo$42b3o$3bo26bo10bo$bo27bobo9b2o$5bo24bo$3bo10b2o6bo$6bo7b2o6b3o$6b
3o16bo$9bo14b2o$8b2o2$2o$bo$bobo53bo$2b2o51b3o$55bo$55bo5$2bo$2bo$2b3o
38b2o$4bo38bobo$45bo$14b2o29b2o$14b2o2b2o16b2o$18bobo15b2o$20bo2b2o$
20bobobo$19b2obo$22bo$22b2o!

ufx266.rle:

Code: Select all

40bo$40b2o$29bo2bo4b5o$13b2o15b10o$12b4o13b12o$13b3o8bo3b13o$8bo3bobo
8b2o3b14o$7b2o2b6o6b20o$7b10o6b21o2b5o$9b11o2b30o$9b44o$2b2o6b43o3bo$b
3ob3o2b43ob3o$3b50obobo$3b50obo$4b45o2b2o$3b41ob2o$b42o$18ob23o$ob15o
2b22o$obob4ob7o3b23o$o3b4o2b6o5b19ob3o$3ob4o3b6o4b19ob2o$7o6b4o5b12ob
2obo$12b5o8b7o2b4o$13b2ob3o5b7o3b4o$16b2o6b6o5b2o$22b3o2b3o$23b2o3bo!

vfx266m.rle:

Code: Select all

45bo$43b3o$4bo26bo10bo$2bo27bobo9b2o$6bo24bo$4bo10b2o6bo$7bo7b2o6b3o$
7b3o16bo$10bo14b2o$9b2o4$58bo$56b3o$56bo$56bo5$3bo$3bo$o2b3o38b2o$5bo
38bobo$46bo$15b2o29b2o$15b2o2b2o16b2o$19bobo15b2o$21bo2b2o$21bobobo$
20b2obo$23bo$23b2o!

pairs data:

Code: Select all

*VFX266M

VBX222W2+VFX266M

VFX77S-VFX266

VFX77SW+VFX266M

VFX266-VF216P6
VFX266-VF216P6X
VFX266-VFX177P6
VFX266-VFX177P6X
VFX266-VLX84P15
VFX266M-VF216P6
VFX266M-VF216P6X
VFX266M-VFX177P6
VFX266M-VFX177P6X
VFX266M-VLX84P15

VL106P6-VFX266
VL106P6-VFX266M

VL214P8-VFX266
VL214P8-VFX266M

[dvgrn]

I'm not sure about adding a separate variant of F171, since its only purpose would be to precede R126. On the other hand, it shouldn't be assumed that users will know how to make these connections work. For now, perhaps you should continue to add the 'close' pairs as comments with an RLE of the connection.

Yes, that sounds about right. Who knows -- by the time such low-priority tasks as this need to be done, maybe Karel will have come out with a whole new Java-based H-B-R-G-pi-2G-century-switchengine signal search utility.

I almost think we should add variants of the block+pre-Herschel based dependent conduits with the first glider eater removed, as there are now a number of conduits that rely on it being gone (e.g. Lx142, Bx167x, etc.).

Would a new F166m and Lx200m and so on actually solve any problems, though? The way I set it up, that first dependent-conduit eater automatically self-destructs whenever a conduit needs it to be gone. The row of dots marking the glider path serves a double purpose. So it shouldn't be any more of a problem than a ghost Herschel... I think for now I'm comfortable leaving my current kludgey solution the way it is.

... Thanks for the F266 / Fx266 data! But are you still avoiding component names in the pairs lists? That turned out to be a superstition based on a trivial case-sensitivity problem. It only shows up for reflected outputs like Fx119 -- you can't just edit VFX119B down to FX119, it has to be "Fx119".

With all the new conduits, the pairs list is getting really long, and it would seem to make sense to substitute a single component for a list of variants wherever possible. Eventually it will make the pairs list easier to maintain, I think.

Also, there are cases where we really _should_ be forbidding the component, not the variants. I did a few trial component-instead-of-variant substitutions recently, and noticed that there were places where the standard variant was forbidden but the dependent-conduit variant wasn't. For example, VBX222-VFX176 is currently listed but VBX222-VFX176X isn't mentioned. For connections like this, the presence or absence of an eater near the output shouldn't matter, so is there any reason why this shouldn't just say VBX222-Fx176?

I've started by replacing VFX158 with Fx158 (just as an experiment, since there's only one variant for that component anyway) and reduced some of the pairs listings for L214p8 and L106p6 while I was copying the F[x]266 data in. At some point I'll go through all the components with X-suffix variants, one at a time, and see where the pair listings could be cleaned up.

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

I'm now claiming Guam's/Extrementhusiast's R325 along with Lx198, Lx159p8, and Rx162. With any luck I'll have another Hersrch archive posted by the end of the weekend. Don't know if I'll have time for a lot of testing, but eventually I'll try a good sanity check -- run old original Hersrch on an arbitrary connection with a hundred different solutions, then make sure the latest Hersrch can still find every one of those -- along with a pile of new connections, no doubt. It will be interesting to see if the total number of candidates has just gone up 10% or something like that. I suspect it will be somewhere between a factor of two and an order of magnitude...!

[Sokwe]

The way I set it up, that first dependent-conduit eater automatically self-destructs whenever a conduit needs it to be gone. The row of dots marking the glider path serves a double purpose.

Oh, I see. Yes, that's a much better solution. Looking back, the pattern I provided for Bx167x also does this, but it was entirely unintentional.

[A]re you still avoiding component names in the pairs lists?... With all the new conduits, the pairs list is getting really long, and it would seem to make sense to substitute a single component for a list of variants wherever possible. Eventually it will make the pairs list easier to maintain, I think.

I was continuing to use the component names to keep it consistent with the rest of the list, but I'll use conduit names where appropriate from now on.

[T]here are cases where we really _should_ be forbidding the component, not the variants.

I think the component names should only be replaced by conduit names when there is no reasonable possibility that some modification could allow the two components to connect. I'm not sure exactly how to define this limit.


The next conduit I intend to submit is Bx285. There are two variants of this conduit, although I think the faster one (left) works in most cases where the slower one (right) works (the only exception I can see is when it is preceded by L214).

Code: Select all

x = 81, y = 73, rule = B3/S23
12b2o48b2o$12bobo47bobo$13b2o48b2o3$19bo49bo$18bobo47bobo$18bobo47bobo
$19bo49bo14$9bo49bo$9bo17b2o30bo17b2o$7b3o17bo29b3o17bo$7bo20b3o26bo
20b3o$30bo49bo3$74bob2o$58b2o14b2obo$23b2o32bobo$23bo33bo$21bobo32b2o$
21b2o3$60bo$59bobo$59b2o$5bo$4bobo18b2o$4b2o19b2o9$24b2o48b2o$24b2o48b
2o12$o49bo$obo20b2o25bobo20b2o$3o20bo26b3o20bo$2bo10b2o3b2o4b3o25bo10b
2o3b2o4b3o$11bo2bo4bo6bo34bo2bo4bo6bo$11b3o4bo42b3o4bo$18b2o48b2o$11b
2obo46b2obo$11bob2o46bob2o!

Should I add both variants?

Edit: Here are the reasonably-compact* stable conduits with repeat times below 300 ticks that have not been added nor 'claimed':

Code: Select all

x = 1311, y = 124, rule = B3/S23
1132bo$1132bo$1116b2o3b2o9b3o$1116b2o3b2o11bo8$1113b2o$1113b2o13b2o$
1128b2o$955b2o$956bo14b2o$956bobo12b2o9b2o$957b2o23b2o2$991b2o$991b2o$
953bob2o$953b2obo155b2o$1112b2o$41bo69bo79bo49bo49bo78bo79bo89bo69bo
79bo80bo84bo73bo60b2o8bo51b2o46bo69bo83bo$991b2o61bo14b2o$41bo69bo79bo
49bo49bo78bo79bo89bo69bo79bo80bo84bo73bo20b2o2b2o44bo52bobo12b2o9b2o
19bo69bo83bo$880b2o69bo2bobo98b2o23b2o$41bo69bo79bo49bo49bo78bo79bo89b
o69bo79bo80bo84bo24bo14b2o32bo21bobo46bo99bo69bo83bo$273b2o606bobo12b
2o9b2o42b2ob2o133b2o$41bo69bo79bo49bo31b2o16bo78bo79bo89bo69bo26b2o51b
o80bo84bo25b2o23b2o21bo70bo87b2o10bo54bo14bo83bo$638bo14b2o326b2o68bob
2o92b2o5b3o143b2o$41bo69bo79bo49bo49bo78bo79bo89bo69bo27bobo12b2o9b2o
24bo80bo84bo59b2o12bo50b2o18bo49b2obo46bo45b2o4bo17bo83bo28b2o14bo$
279b2o358b2o23b2o250b2o235b2o118b2o9b2o12bobo$41bo69bo79bo49bo37b2o10b
o78bo79bo89bo69bo79bo80bo84bo21bob2o48bo70bo87b2o10bo69bo83bo17b2o23b
2o$275b2o396b2o203b2obo110b3o94b2o$41bo69bo79bo49bo17bo15b2o14bo78bo
79bo89bo69bo62b2o15bo80bo84bo73bo61bo8bo47b2o2b2o46bo11b2o37b2o17bo83b
o8b2o$259bo375bob2o277b2o73b2o56bo2bobo59bo37b2o110b2o$41bo69bo79bo49b
o12b2ob3o31bo78bo79bo89bo21b2o46bo24b2obo51bo80bo84bo59b2o12bo70bo48bo
bo48bo9b3o4b2o3b2o46bo83bo44b2obo$254b2obo305bo14b2o296b2o2b2o167b2ob
2o57bo6bo4bo2bo173bob2o$41bo69bo79bo49bo38b2o9bo78bo79bo89bo22bobo12b
2o9b2o19bo62b2o15bo80bo32b2o50bo19bo2bobo48bo70bo99bo17bo4b3o44bo83bo$
280b2o59b2o221b2o23b2o82b2o130bo14b2o55bobo199b2o37b2o32b2o110b2o$41bo
69bo79bo49bo49bo50bo27bo79bo50bo38bo69bo22b2o2b2o51bo80bo33bobo12b2o9b
2o23bo19b2ob2o49bo70bo77b2o20bo21bob2o7b2o3bo12b2o17bo83bo8b2o$15b2o
325b3o156bo96b2o33bo2bobo167b2o23b2o290b2obo7bo3bobo159b2o2b2o$15bobo
23bo69bo79bo49bo49bo78bo79bo48b3o9b2o3b2o22bo57b2o10bo23bobo53bo80bo
84bo49b2o22bo70bo99bo33bo3b2o14b2o14bo83bo44bobo2bo$16b2o207b2o272bo
11b2o3b2o42bob2o69b2ob2o202b2o64b2o182b3o43bo18bobo144bobo$41bo69bo79b
o33b2o14bo49bo78bo79bo89bo19b2obo46bo79bo80bo68b2o14bo73bo70bo88bo10bo
32bob5o16bo13bo83bo45b2ob2o$663b2o137bob2o283b2o42bobo3bo17b2o$22bo18b
o69bo79bo49bo49bo78bo79bo89bo57b2o10bo52b2o25bo80bo30b2obo50bo60b3o10b
o70bo99bo31bo2bo4bo11b2o16bo83bo18b2o$21bobo207b2o365b2o317bo216b2o4b
2o11bobo118b2o$21bobo17bo69bo79bo39b2o8bo49bo78bo79bo89bo17b2o2b2o46bo
79bo80bo68b2o14bo59b2o12bo70bo99bo53bo15bo83bo$22bo204b2o88b2o239bo2bo
bo110b3o163b2o313b2o$41bo69bo79bo19bo15b2o12bo49bo26bo28b2o21bo79bo89b
o18bobo48bo63bo15bo80bo28b2o2b2o50bo73bo70bo99bo69bo83bo6b3o$211bo106b
obo26b2o170b2o37b2ob2o110b2o125bo2bobo458bo$41bo69bo79bo14b2ob3o29bo
49bo27b2o49bo79bo53b2o13b2o19bo69bo79bo80bo29bobo52bo73bo20bo49bo99bo
39bo29bo83bo8b2o$206b2obo294b2o82b2o210b2ob2o145bobo187bobo$41bo69bo
79bo40b2o7bo49bo78bo79bo89bo47b2o20bo79bo80bo84bo73bo19b2o49bo46b2o51b
o38b2o29bo83bo$232b2o596b2o215bobo$41bo69bo79bo49bo32b2o15bo78bo79bo
89bo69bo79bo80bo58b2o24bo73bo70bo45bo14b2o37bo69bo83bo$274bo324b3o444b
2o14bo$41bo69bo79bo49bo30bobo16bo78bo79bo89bo58bo10bo79bo80bo84bo17b2o
54bo70bo58bobo38bo69bo83bo$272b2o324b2o241b3o29bobo184b2o$41bo69bo79bo
49bo49bo78bo79bo89bo69bo21b2o56bo80bo69bo14bo16bo56bo70bo99bo69bo83bo$
631bobo206b2o30b2o$41bo69bo79bo49bo49bo78bo79bo69b2o18bo69bo20bo14b2o
42bo80bo84bo73bo70bo99bo69bo83bo$12bo507b2o108b2o14bo66b2o78b2o228b2o$
12bo17b2o9bo69bo79bo49bo49bo36b2o40bo79bo89bo69bo33bobo43bo22b2o56bo
13bo7b2o61bo73bo70bo13bo7b2o38bo37bo69bo83bo$10b3o17bo225bo70bobo314b
2o86b2o51b3o88bo138b3o44bobo226bob2o$10bo20b3o7bo69bo79bo49bo13bobo18b
2o13bo35bo42bo18bo60bo89bo69bo79bo41bo38bo16bo67bo18bobo52bo34b2o34bo
16bo44bo37bo53b2o14bo83bo35b2obo$33bo221b2o19b2o48b2o61b3o237bo100bobo
54b2o86b2o88b2o50b2o136b2o$41bo69bo79bo49bo49bo78bo21bo57bo33b2o54bo
69bo18b3o58bo35b2o2b2o39bo84bo73bo70bo99bo69bo83bo$29b2o198b2o148b2o
10b2o26bob2o36b2o24bo134bo11bo66b2o25b2o51b2o88b2o138b2o$27bo2bo10bo
69bo79bo35bo2bo10bo49bo78bo9bo38b2obo27bo9bo13b2o6b3o55bo30b2o37bo9b3o
8b2o14bo42bo9bo70bo8bo75bo13bo59bo70bo8bo90bo69bo83bo$11b2o14b2o182b2o
14b2o123b2o26bobo43b2o32bobo11b2o6bo88bo51bo22bobo51bobo77bobo87bobo
137bobo51b2obo$10bobo16b4o8bo69bo79bo18bobo16b4o8bo49bo60b2o16bo10b2o
36b3o4bo23bo10b2o77bo28bobo38bo11b2o23bo42bo10b2o68bo9b2o73bo14b2o57bo
70bo9b2o44bo6bob2o33bo69bo83bo$10bo18bo2bo120b2o55bo18bo2bo80b2o104bo
2bo4bo6bo134b2o160b2o323bobo$9b2o16bobo11bo69bo42bo36bo17b2o16bobo11bo
49bo21b2o55bo50b2o3b2o4b3o15bo89bo69bo79bo40b2o38bo39b2o43bo73bo70bo
54bobo42bo69bo83bo$27b2o39bo85bobo70b2o202bo379b2o244bo$41bo17bo8b3o5b
2o33bo37b2o4b2o34bo49bo33b2o14bo78bo60b2o17bo68b2o19bo69bo79bo80bo84bo
73bo70bo99bo69bo83bo$9bo49b3o9bo5bo72bo58bo65b2o242bo$9b3o29bo20bo7b2o
5b3o31bo32b2o3bo41bo17b3o29bo49bo78bo79bo58b2o3b2o4b3o17bo69bo79bo80bo
84bo34b2o37bo70bo64bo34bo69bo56bob2o23bo$o11bo36b2o10b2o56b2o24bo3b2o
49bo11bo86b2o80bo79bo45bo2bo4bo6bo28bo69bo20b2o57bo79bo24b2o63bo19b2o
48bo69bo53bobo63bo49bo46b2obo49bo$3o8b2o14bo13bo8bo60bo8bo13b2o6b3o46b
o8b3o8b2o14bo13bo49bo8bo52b2o15bo10bobo66bo10bobo43b3o4bo25bo10bobo56b
o10bobo18bobo45bo10bobo67bo9bobo22b2o2b2o44bo14bobo56bo10bobo20b2o35bo
9bobo52b2o33bo29bobo20b2o15bo9bo53b2o18bo25bo$3bo22bobo21bobo47b2o18bo
bo11b2o6bo60bo22bobo71bobo50bobo25b3o77b3o50b2o35b3o67b3o20bo56b3o77b
3o26bobo58b3o67b3o20bo46b3o117b3o20bo26b3ob2o41b3o4bo45b3ob2o$2b2o23bo
13bo9b2o46bobo9bo9b2o68bo10b2o23bo13bo49bo9b2o52bo14bo12bo11b2o53bo12b
o43b2obo29bo12bo56bo12bo11b2o7b2o44bo12bo67bo11bo28bo43bo16bo56bo12bo
10b2o3b2o4b3o33bo11bo19b2o66bo31bo10b2o3b2o4b3o13bo11bob2o41bo2bo4bo6b
o11bo27bob2o$99b2o71b2o181b2o38bo93b2o16bob2o124bo176b2o74b2o62bo2bo4b
o6bo66bo23bo83bo2bo4bo6bo72b2o3b2o4b3o$41bo69bo59bo2bo16bo49bo49bo78bo
25b3o51bo39bo49bo22b2o45bo25b3o51bo22b2o56bo21b2o61bo31b2o40bo21b3o4bo
41bo21b2o6b3o22b3o42bo40b3o4bo21bo68bo14bo$172b2o224bo88b3o73b2o2b2o
69bo74b2o2b2o74b2o2b2o160b2o62b2o2b2o2bo2b3o18bo93b2o89b2o$24b2o15bo
69bo79bo32b2o15bo49bo78bo79bo36bo52bo26bobo40bo79bo26bobo51bo25bobo56b
o73bo21b2obo45bo25bobobobo2bo18b2o44bo40b2obo25bo83bo$24bobo67b2o128bo
bo243b2o3b2o92bo149bo40b2o37bo152bob2o73bobob2o107bob2o163b2o$26bo14bo
51bo2bo14bo79bo34bo14bo49bo78bo79bo20bo3bo64bo28b2o39bo79bo28b2o40bo9b
o27b2o55bo73bo70bo27b2o70bo69bo83bo53b2o$bo24b2o23bo42b2o25bo55b2o22bo
24b2o23bo49bo130b2o34b3o5b3o282b3o419b2o$bobo18b2o17bo9bobo57bo9bobo
53bobo11bo9bobo18b2o17bo9bobo20b2o15bo9bobo66bo61b2o16bo17bo9bo61bo69b
o79bo80bo84bo22b2o49bo70bo99bo69bo10bobo70bo27b2o$b3o18bo28b3o67b3o54b
2o21b3o18bo28b3o20bo26b3o34b2o540bo301bo101bo$3bo11b2o6b3o15bo11bo11b
2o44bo11bo67bo11bo11b2o6b3o15bo11bo10b2o3b2o4b3o13bo11bo19b2o7b2o5bo
30bo79bo89bo69bo79bo59b2o19bo84bo20b3o50bo70bo99bo69bo9b2o25b2o45bo25b
3o25b2o$15bo9bo39bo5b2o142bo9bo36bo2bo4bo6bo46bo7b2o2b3o411b2o125bo
330b2o71bo27b2o$16b3o22bo24b3o2bo39bo36b2o5b3o33bo24b3o22bo20b3o4bo21b
o21b2o6b3o12bo33bo79bo89bo69bo79bo80bo84bo73bo70bo99bo69bo83bo39b2o$
18bo49bo3bo76bo5bo62bo50b2o42b2o2b2o2bo2b3o10b2o381b2ob2o569bobo$41bo
29b2o4b2o32bo34b3o5b2o35bo49bo20b2obo25bo25bobobobo2bo11bo31bo79bo89bo
69bo79bo30bobo47bo84bo73bo70bo99bo69bo22b2o59bo38bo$76bobo51b2o3b2o9bo
115bob2o53bobob2o11bo383bo2bobo469bo97b2o8b2o$41bo34bo34bo19bo3bo55bo
49bo49bo27b2o15b2o32bo79bo89bo69bo79bo29b2o2b2o45bo84bo73bo70bo99bo69b
o20b3o46b2o12bo47bo$75b2o51b3o5b3o277b2o342b2o430bo48b2o61b3o$41bo69bo
16bo9bo52bo49bo49bo78bo45b2o13b2o17bo89bo69bo79bo69b2o9bo84bo73bo70bo
99bo69bo32b2o49bo50bo$431b2o771bo$41bo69bo79bo49bo49bo78bo79bo89bo69bo
79bo31b2obo45bo84bo73bo70bo99bo69bo33b3o47bo$722bob2o481bo$760b2o$760b
2o2$726b2o23b2o$725bobo12b2o9b2o472b2o$411bo11b2o3b2o295bo14b2o483b2o
13b2o$411b3o9b2o3b2o294b2o514b2o$413bo$413bo6$1220bo11b2o3b2o$1220b3o
9b2o3b2o$1222bo$1222bo!

The last two are dependent conduits, but the last one (Rx328) is a little strange, as it loops back towards its input. It can work in a few sequences however, for example, Bx167-Rx328-L156:

Code: Select all

x = 89, y = 94, rule = B3/S23
78b2o$62b2o14bo$51b2o9b2o12bobo$51b2o23b2o2$42b2o$42b2o$78b2obo$78bob
2o2$42b2o$42b2o$78b2o2b2o$78bobo2bo$80bobo$79b2ob2o2$52b2o$52b2o2$30b
2o$31bo8b3o$31bobo8bo$32b2o8b2o5$21b2o$21bo6b2o$22bo6bo$21b2o6bobo$30b
2o3$43b2o$43bobo23bob2o$2bo15b2o25bo23b2obo$2bo15b2o25b2o$3o$o2$40b2o$
40bo$25b2o14b3o$26bo16bo$23b3o6bo$23bo7bobo$32bo2$59bo$59bo$43b2o3b2o
9b3ob2o$43b2o3b2o11bob2o5$87b2o$87b2o2$40b2o$40b2o$87b2o$87b2o$73b2o$
58b2o12bobo$58bo13bo$59b3o9b2o8b2o$61bo19bo$82b3o$84bo$39b2o$39b2o12$
64bo$40b2o20bobo$41bo20b3o$38b3o4b2o3b2o10bo$38bo6bo4bo2bo$46bo4b3o$
45b2o$50bob2o$50b2obo!

*You know, sort of...

[dvgrn]

The next conduit I intend to submit is Bx285. There are two variants of this conduit...
Should I add both variants?

Well, how about if the slower one is only needed after L214, just put the slow-variant RLE after the usual comment?

Code: Select all

# L214-Bx285: 12b2o$12bobo$13b2o3$19bo$18bobo$18bobo$19bo14$9bo$9bo17b2o$7b3o17bo$7bo20b3o$30bo3$24bob2o$8b2o14b2obo$7bobo$7bo$6b2o4$10bo$9bobo$9b2o12$24b2o$24b2o12$o$o22b2o$3o20bo$2bo11bo3b2o4b3o$11bo2bo4bo6bo$11b3o4bo$18b2o$11b2obo$11bob2o!

But if you find too many more such exceptions, then maybe it's time to define an official variant.

Here are the reasonably-compact* stable conduits with repeat times below 300 ticks that have not been added nor 'claimed':...

After a quick glance, I'd say these should probably all get added to Hersrch. Maybe just one or two more incremental updates after this weekend's, then -- unless some more new conduits show up! (I do still have my transceiver experiment to try, though...)

[Sokwe]

Here's the data for Bx285:

Code: Select all

Bx285 Bx 285 7 -42

  # Standard variant (following glider required)
  VBX285 1 229 0 GRU data/vbx285.rle 1 6 data/ubx285.rle 2 3

vbx285.rle:

Code: Select all

12b2o$12bobo$13b2o3$19bo$18bobo$18bobo$19bo14$9bo$9bo17b2o$7b3o17bo$7b
o20b3o$11bo18bo$13bo2$14bo$16bo$23b2o$23bo$21bobo$21b2o2$3b2o$4bo$b3o$
bo$5bo$4bobo18b2o$4b2o19b2o9$24b2o$24b2o12$o$o22b2o$3o20bo$2bo10b2o3b
2o4b3o$11bo2bo4bo6bo$11b3o4bo$18b2o$11b2obo$11bob2o!

ubx285.rle:

Code: Select all

13b2o4b2o$13bobo2b5o$14b10o$15bob7o$16bob6o$15b8o$14b9o$15b8o$15b7o$
15b6ob3o$15b13o$16bob10o$17b14o$17b14o$16b16o$16b15o$17b14o$16b14o$15b
13o$15b13o$15b14o$12b2ob13o$10bob14ob3o$8bobob14o2b2o$8b3ob11obo3bo$8b
o3b12o$12b11o$12b11o$11b9o$11b10o$11bo2b7o$14b6o$14b6o$14b6o$13b7o$14b
7o$14b7o$11b2ob7o$12b11o$13b7ob3o$15b10o$15b10o$14b11o$15b10o$14bob8o$
13b9o$11b11o$11b10o$10b12o$9b14o$9b17o$10b17o$10b17o$9b17o$6b16o$5b16o
$5b16o$4b16o$3b20o$3b20o$4b20o$3b22o$b24o$24o$ob13ob10o$obob4ob7o2b2o
4b2o$o3b4o2b3obo4bo4bo$3ob4o4b4o3b2o$7o4b2o2bo$12b3o!

Pairs data:

Code: Select all

Bx285-VBX167X
# Bx285-Bx198: 7b2o$7b2o3$b2o$b2o$5b2o$5b2o15bo$22bo$22b3o$24bo$2o$2o7$24bo4$20bo4$16bo11b2o$6b2o13b2o5bobo$5bobo13bobo5b2o$5bo17bo$4b2o17b2o$35bo$34bobo$34bobo$20bo14bo$19bobo$20b2o2$26b2o$26bo$24bobo$24b2o6$4b2o$4b2o19bo$25bo17b2o$23b3o17bo$23bo20b3o$7b2o37bo$7b2o2$40bob2o$24b2o14b2obo$23bobo$23bo$16b2o4b2o$16bo$17b3o$19bo$26bo$25bobo$25b2o12$40b2o$40b2o12$16bo$16bobo20b2o$16b3o20bo$18bo10b2o3b2o4b3o$27bo2bo4bo6bo$27b3o4bo$34b2o$27b2obo$27bob2o!
# Bx285-Bx260: 17b2o$16bobo$16b2o3$11bo$10bobo$10bobo$11bo14$21bo$2b2o17bo$3bo17b3o$3o20bo$o18bo$17bo2$16bo11b2o$6b2o6bo6b2o5bobo$5bobo13bobo5b2o$5bo17bo$4b2o17b2o$35bo$34bobo$34bobo$20bo14bo$19bobo$20b2o2$26b2o$26bo$24bobo$24b2o6$4b2o$4b2o19bo$25bo17b2o$23b3o17bo$23bo20b3o$7b2o37bo$7b2o2$40bob2o$24b2o14b2obo$23bobo$23bo$16b2o4b2o$16bo$17b3o$19bo$26bo$25bobo$25b2o12$40b2o$40b2o12$16bo$16bobo20b2o$16b3o20bo$18bo10b2o3b2o4b3o$27bo2bo4bo6bo$27b3o4bo$34b2o$27b2obo$27bob2o!
Bx285-Fx119
# Bx285-VLX73P8B: 29b2o$29bobo$30b2o$13b2o$13bo$11b3o22bo$35bobo$15bo19bobo$36bo$16bo$18bo2$19bo4$3b3o$3b3o$3b3o$3o$3o$3o$26bo$26bo17b2o$24b3o17bo$24bo20b3o$47bo5$40b2o$40bo$38bobo$38b2o5$17b2o$18bo3bo$15b3o3bobo18b2o$15bo5b2o19b2o9$41b2o$41b2o12$17bo$17bobo20b2o$17b3o20bo$19bo10b2o3b2o4b3o$28bo2bo4bo6bo$28b3o4bo$35b2o$28b2obo$28bob2o!
Bx285-VLX73P8C
Bx285-Lx216p8
Bx285-VR64E1
Bx285-VR64E3
Bx285-Rx140
Bx285-Rx202

# L106-Bx285: 25b2o$25bo$23bobo$23b2o$41b2o24b2o$41b2o8b2o14bo$31bo19b2o12bobo$31b3o31b2o$34bo$33b2o$6b2o$5bo2bo58b2obo$6b2o59bob2o2$72b2o$71bobo$b2o64b2o2bo$obo64bobob2o$2o67bo$68b2o4bo2bo$74bo$22b3o48bo4bo$24bo47bobo2b2o$24b2o44b2obo$41bo$40bobo$36b2o3b2o27bo2bo$36bobo33b2o$38bo27b2o$38b2o26bo$64b3o7$73b2o$54bo18bobo$54bobo18bo$54b3o18b2o$56bo11$54bo2bo$54bo$53bo4bo$52bobo2b2o$50b2obo$64b2o$64b2o2b2o$50bo2bo8b2obo2b2o$52b2o8b3o3$63b2o$63b2o!

# L214-Bx285: 25b2o$25bo$23bobo$23b2o4b2o$29bo37b2o$30bo20b2o14bo$29b2o20b2o12bobo$65b2o3$6b2o$5bo2bo58b2obo$6b2o59bob2o4$b2o64b2o2b2o$obo64bobo2bo$2o67bobo$38bo29b2ob2o$37bobo$22b3o5b2o6b2o$24bo5bo$24b2o5b3o$33bo4$47b2o17b2o$43b2obo2bob2o13bo$43b2o2bo4bo11b3o$48bo$49bobo12$36bo$36bo$36b3ob2o$38bob2o4$74b2o$74bobo$76bo$76b2o$38b2o$39bo$36b3o$36bo8$47b2ob6o$47b2ob6o$47b2o$47b2o5b2o$47b2o5b2o$47b2o5b2o$54b2o$47b6ob2o$47b6ob2o2$45bo!

There are currently three cases that require the other variant of the conduit, but two of these cases (Bx285+Bx198 and Bx285+Bx260) seem to not allow any currently-known conduits to follow them.

[dvgrn]

[T]here are cases where we really _should_ be forbidding the component, not the variants.

I think the component names should only be replaced by conduit names when there is no reasonable possibility that some modification could allow the two components to connect. I'm not sure exactly how to define this limit.

Sometimes vague limits are exactly what's needed...!

I can think of a type of overlap that's generally sufficient for claiming impossibility, but I'm not sure that it's a necessary condition. You might say there are two types of catalysts: primary catalysts affect the leading edge of an active pattern, where secondary catalysts only modify sparks or clean up extra junk still lifes.

The first eater in an Fx77 is a primary catalyst, for example -- if you take it out, the reaction is immediately changed radically and the output Herschel is never produced. The first eater in an Fx153 is a secondary catalyst -- it's just suppressing an extra block, so the Herschel still gets where it's going if you leave out that eater.

Seems like if a conduit's active area overlaps the reaction area of a *primary* catalyst in a connected conduit, then it's pretty safe to say that finding a modification of the connected conduit is wishful thinking. If there's no room for that primary catalyst, then any modification that fixes the overlap problem is just plain going to be a different conduit anyway.

Unfortunately or fortunately, that leaves a lot of wiggle room in the definition of the "active area" for the first conduit. Many conduits have a primary active area and a larger secondary active area. A good example is the Fx119x in Hersrch's Spartan data, which is pretty much all primary active area except near the first eater -- as opposed to the standard Spartan Fx119 in the same folder, which has a huge secondary active area.

So if two conduits' primary active areas overlap, it's pretty safe to say the combination is impossible. Primary active areas would include the active parts of the primary catalysts -- the heads of fishhook eaters and so forth... very much along the lines of Hersrch's overlap diagrams, so perhaps this is all an exercise in stating the obvious!

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

I've rolled in the Bx285 data, and am working on compression ratings and other statistics for the four conduits on my list. So the December build of Hersrch is coming along nicely; further bulletins as events warrant.

[dvgrn]

I'm now claiming Guam's/Extrementhusiast's R325 along with Lx198, Lx159p8, and Rx162.

Well, it turned out that Lx159p8 was thoroughly pointless. I had forgotten already that there's a perfectly good stable version of Lx159. And R325 turned out to be really R246, as far as I can see... I also added Extrementhusiast's several-times-revised Lx79p6, since it's now at least as compact and potentially useful as most of the other p6 circuitry that Hersrch already has registered.

So the full list of new additions is Bx285, F266(m), Fx266(m), Lx79p6, Lx198(b), R246, and Rx162. Of these, Rx162 appears to show up the most often in new searches. For p6 searches, Lx79p6 shows up all over the place, too. I've gotten all these conduits to appear in search results. They connect correctly to surrounding components and generally seem to be behaving themselves, so I figure it's time for a new release (4 December 2012). I've updated the link and conduit list at the top of this thread.

Still haven't found time or energy to tackle forbidden pairs yet. That's the next task on my list -- I'll be going through test.bat results with a fine-tooth comb, looking at all the conduits I've added so far, revising the pairs data and possibly adding some forbidden triples and quadruples and so forth, before I add any more new conduits. Maybe enough time will pass that it will be clear I should switch over to supporting Karel's new Java-based Hersrch effort instead...!

[dvgrn]

This is just a placeholder to remind any interested parties (or maybe just me) that there's some experimenting to be done to put together small to medium-size composite Snark-assisted Herschel conduits that can reach (X,Y,T) offsets that were never reachable before.

As a random sample, here's a glider-to-Herschel converter attached to a Snark-assisted Lx117+F117 composite:

Code: Select all

#C composite Lx234 = Lx117+F117+Snark
x = 72, y = 118, rule = LifeHistory
59.2A$58.A2.A$59.2A8$54.3D$55.D$55.3D15$25.A$25.3A28.2A$28.A27.2A$27.
2A$63.A.2A$63.2A.A$16.2A.2A$16.2A.A$19.A25.2A$19.3A4.2A18.A$17.2A3.A
3.2A18.A.A$16.A2.4A24.2A$16.2A.A15.2A$17.A2.3A12.A.A$17.A5.A13.A23.2A
$18.5A14.2A22.A.A$20.A42.A$63.2A6$48.3D$49.D8.2A$49.3D6.A$59.3A$61.A
6$41.2A13.2A$42.A13.A$39.3A12.A.A$39.A14.2A2$39.A$39.3A$30.A11.A$4.A
9.A15.3A8.2A14.A$4.3A5.3A18.A22.A.A$7.A3.A20.2A23.A$6.2A3.2A$62.A$60.
3A$59.A$59.2A$7.C$8.C22.D$6.3C22.D.D33.2A.2A$20.2A9.3D34.A.2A$20.2A
11.D11.2A21.A$8.2A35.A14.2A4.3A$7.A2.A35.3A11.2A3.A3.2A$2.2A4.2A38.A
16.4A2.A$.A.A47.2A15.A.2A$.A48.A.A12.3A2.A$2A48.A13.A5.A$10.2A37.2A
14.5A$10.A56.A$11.3A$13.A5$37.A$37.3A$40.A$39.2A7$49.2A$42.2A5.A.A$
42.2A7.A$51.2A2$38.A$37.A.A.2A$37.A.A.A.A$36.2A.A.A.A2.A$37.A2.2A.4A$
37.A4.A$38.3A.A2.2A$40.2A3.2A!

EDIT: Similarly, you can subtract 39 ticks from the Lx138 to get an incomplete L99 conduit, which can be completed in the same way to get an L99+F117 composite conduit (and various other conduits could be swapped in for the F117, of course):

Code: Select all

#C composite L216 = L99+F117+Snark
x = 53, y = 75, rule = B3/S23
8b2o$7bo2bo$8b2o8$12b3o$13bo$11b3o15$43bo$11b2o28b3o$11b2o27bo$40b2o$
2b2obo$2bob2o$48b2ob2o$49bob2o$22b2o25bo$22bo18b2o4b3o$20bobo18b2o3bo
3b2o$20b2o24b4o2bo$32b2o15bob2o$31bobo12b3o2bo$6b2o23bo13bo5bo$5bobo
22b2o14b5o$5bo42bo$4b2o6$19b2o$9b2o8bo$10bo6b3o$7b3o$7bo2$9bo$9b3o$o
11bo$3o8b2o$3bo$2b2o3$24b2o$24b2o3$bo$bobo20b2obo$b3o20b2ob3o$3bo11b2o
13bo$15bo8b2ob3o$16b3o4bo2b2o$18bo4b2o!

A number of Guam's discoveries, where a glider generated toward the end of a composite conduit cleans up a leftover B-heptomino block, might now have one or more Snark-assisted variants with different Herschel output locations. It might also be worth re-running old searches to find not-quite-perfect Herschel circuits that left a little junk behind, which could now be vanished away by a Snark or two.

And if anyone remembers any other newish Herschel conduits that ought to be added to Hersrch, this would be a good place to post them...!

[Extrementhusiast]

I remember making a p4 variant to my Lx79p6 conduit a while back (it was actually the first variant I found), but stupidly didn't save it. However, I do remember that it included a modified T-nosed p4, in a space which a regular T-nosed p4 couldn't fit. (Additionally, the T was pointing downwards, like the actual letter.)

For reference, the p6 variant:

Code: Select all

x = 28, y = 34, rule = LifeHistory
16.2A$15.A2.A$18.A$18.A$12.A3.A.A$11.A3.A.A$11.A4.A$12.4A7$23.D$.2A
20.D.D$2.A20.3D$2.A.A20.D$3.2A3$21.A3.2A$16.2A2.A.A3.A$16.A3.2A3.A$
17.3A4.A$19.A.4A.A$20.2A3.A.A$24.A2.A$25.2A2$4.3C7.2A$5.C8.A.A$5.3C8.
A$16.2A!

[simsim314]

I'm not sure about this, but here is another version of vl156 that shoots glider in different direction:

Code: Select all

x = 67, y = 72, rule = LifeHistory
15$41.3D$42.D$40.3D16$40.2A$31.2A7.2A$32.A$32.A.A$33.2A6$32.A$32.3A$
23.A11.A$23.3A8.2A14.A$26.A22.A.A$25.2A23.A4$47.2A$47.A.A$49.A$24.C
24.2A$24.C.C18.2A$24.3C18.A$26.C11.2A6.3A$38.A9.A$39.3A$41.A!

Here is what present in Hersch (without an eater that kills one glider):

Code: Select all

x = 49, y = 66, rule = LifeHistory
7$25.3D$26.D$24.3D16$24.2A$15.2A7.2A$16.A$16.A.A$17.2A$33.2A$33.A$31.
A.A$31.2A2$16.A$16.3A$7.A11.A$7.3A8.2A14.A$10.A22.A.A$9.2A23.A7$8.C$
8.C.C$8.3C$10.C11.2A$22.A$23.3A$25.A!

[dvgrn]

I'm not sure about this, but here is another version of vl156 that shoots glider in different direction...

Yes, that's one of the alternate conduits that you might call "well-known but not well-documented". There didn't seem to be a reason to include it in Hersrch, because it takes up basically the same area and Hersrch doesn't keep track of gliders.

I've often wished that you could search for glider outputs (placement and/or timing) with Hersrch. But there are now several dozen H-to-G converters with various geometries and limitations, and it's a non-trivial problem to include them in the search. The problem of designing a good generalized tool gets complicated very quickly -- see Karel Suhajda's notes on a Java version of Hersrch. Really it would be nice to support gliders as signals, and also pi, R, B, tandem gliders, and maybe other signal types not yet invented -- swimmers? *WSSes? loafers, even?

Anyway, just to make the record a little more complete, here's an old period-doubler version of the L156, with yet another glider output:

Code: Select all

#C Paul Callahan's period-doubling L156b, with a transparent beehive
#C   creating an edge-shooting glider output
x = 67, y = 65, rule = B3/S23
47b2o$46bo2bo$47b2o19$61b2o$61bobo$63bo$63b2o4$50b2o$50b2o$63b2o$41b2o
bo18bo$41bob2o16bobo$26bo34b2o$24b3o$o22bo$3o20b2o$3bo$2b2o38bo23bo$
42b3o19b3o$33bo11bo17bo$33b3o8b2o17b2o$36bo$11bo23b2o$9b3o48b2o$9bobo
47bo2bo$9bo50b2o4$34bo$34bobo$23b2o9b3o$10b2o11b2o11bo11b2o$11bo36bo$
8b3o38b3o$8bo42bo5$23b2o$22bobo$22bo$21b2o!

It's possible to bolt on another few conduits to get a slower non-period-doubling conduit which is still an edge-shooter -- see the pattern on the far right, below, where it forms the first stage of a three-way Herschel splitter. It would be easy to get rid of the tub-with-tail by going back to the other form of Herschel receiver... once you have that many gliders flying around, you can catch them and produce Herschels in a bazillion different ways -- all of them probably a little awkward and not quite what you want in any given situation (!).

Code: Select all

#C known uses forL156b conduit, from the left:
#C composite F561 conduit, two maybe-Spartan Herschel splitters,
#C  and a non-Spartan three-way Herschel splitter.  Recovery times
#C  are 251 for the first three, and 503 for the fourth.
x = 630, y = 160, rule = B3/S23
327b2o$326bo2bo$327b2o16$338b2o$338bobo$340bo$340b2o7$330b2o$330b2o$
188bo$186b3o132b2obo$169b2o14bo9b2o3b2o119bob2o182b2o$41bo127bo15b2o8b
2o3b2o138b2o166bo$39b3o128bo169b2o153bo11bo$38bo130b2o179b2o143b3o9b2o
$38b2o167b2o141bo147bo$139bo67bo140bobo134b2o10b2o39bo$138bobo64bobo
140b2o136bo50bobo$138bobo64b2o279bobo48bobo$139bo332b2o13b2o49bo$166b
2o304b2o36b2o$45b2o119b2o342b2o$45b2o$466b2o$466b2o$202b2o266b2o$202bo
141b2o124b2o$173b2o28b3o138bobo$156b2o15bo31bo140bo$156bo17b3o169b2o
153b2o$157b3o16bo288b2o34bo$159bo305b2o35b3o$504bo3$29b2o158b2o138b2o$
29b2o158b2o138b2o6$471b2o$52b2o158b2o138b2o116bobo$52b2o158b2o138b2o
116bo$90bo159bo139bo78b2o$88b3o157b3o137b3o$87bo159bo139bo$87b2o158b2o
138b2o$78b2o158b2o138b2o$78bobo37bo119bobo37bo99bobo37bo178b2o$79bo37b
obo119bo37bobo99bo37bobo129bo48bo$52b2o63bobo92b2o63bobo72b2o63bobo
129b3o5b2o38bo$51bo2bo63bo92bo2bo63bo72bo2bo63bo133bo4b2o38b2o$52b2o
36b2o120b2o36b2o100b2o36b2o159b2o$59b2o29b2o127b2o29b2o107b2o29b2o236b
o$27b2o30b2o126b2o30b2o106b2o30b2o116b2o148bobo$27b2o149b2o7b2o129b2o
7b2o148b2o73b2o73bobo$179bo139bo232b2o3b2o13b2o54bo$18b2obo157bobo137b
obo146b2obo34b2o49b2o13b2o26b2o$18bob2o158b2o138b2o146bob2o34b2o92b2o
2$100b2o158b2o138b2o140b2o$46b2o53bo104b2o53bo84b2o53bo141bo$46b2o50b
3o105b2o50b3o85b2o50b3o142bobo$50b2o46bo111b2o46bo91b2o46bo145bobo$19b
o29bobo127bo29bobo107bo29bobo117bo23bo30bo20bo3b2o$19b3o27bo129b3o27bo
109b3o27bo119b3o19b3o29bobo23b2o59b2o$10bo11bo25b2o120bo11bo25b2o100bo
11bo25b2o110bo11bo17bo15b2o15bo2bo84bo$10b3o8b2o14bo23b2o107b3o8b2o14b
o23b2o87b3o8b2o14bo23b2o97b3o8b2o17b2o13bo2bo15b2o69b2o11b3o$13bo22bob
o22b2o110bo22bobo22b2o90bo22bobo22b2o100bo42b2o87b2o11bo$12b2o23bo134b
2o23bo114b2o23bo124b2o49b2o$487b2o24b2o$63b2o158b2o138b2o121bo2bo91b2o
$63b2o158b2o138b2o122b2o93bo$34b2o158b2o138b2o243b3o$34bobo157bobo137b
obo231b2o9bo$36bo159bo139bo231b2o21b2o$11bo24b2o133bo24b2o113bo24b2o
123bo129bo$11bobo18b2o137bobo18b2o117bobo18b2o127bobo36b2o90b3o$11b3o
18bo138b3o18bo118b3o18bo128b3o36b2o92bo$13bo11b2o6b3o137bo11b2o6b3o
117bo11b2o6b3o127bo11b2o27b2o42b2o$25bo9bo149bo9bo129bo9bo139bo27bobo
41bobo11b2o$26b3o157b3o137b3o147b3o24bo43bo13b2o$28bo159bo139bo149bo
23b2o42b2o$515b2o$304b2o148b2o59b2o$305bo149bo$2b2o158b2o138b3o147b3o$
3bo159bo138bo149bo64b2o$3o157b3o354b2o$o159bo16$519b2o$519b2o$554b2o$
555bo$554bo$554b2o$545b2o$545bobo37bo$546bo37bobo$584bobo$585bo$557b2o
$526b2o29b2o$526b2o6$567b2o$513b2o53bo$513b2o50b3o$517b2o46bo$516bobo$
516bo$515b2o$528b2o$528b2o3$530b2o$530b2o!

[simsim314]

I've often wished that you could search for glider outputs (placement and/or timing) with Hersrch

I think we need a design tool, with GUI which capable of handling all kind of signal that it has in the dynamic library (input-output style). Something like "universal signal design tool", with search capabilities built in (like universal Hersch utility).

It's actually on my "todo list", but I want to commit to some other projects before.

[dvgrn]

As mentioned on the Demonoid thread just now, I'd like to get the Spartan data folder for Hersrch updated, finally, to include at least a representative subset of new conduit discoveries from the past couple of years.

This is going to include a fairly large number of new H-to-H connections that include BRx46B as a component -- e.g., if we just add up to two B-to-Bs, F166 transmutes into an Lx conduit or a longer F:
HFx107B + BFx59H = F166
F166+BRx46B = Lx212
Lx212+BRx46B = F258
etc.

Code: Select all

x = 196, y = 59, rule = LifeHistory
96.B$95.3D$95.BDB$94.2B3D$94.5B$94.6B$94.6B72.2A$94.5B74.A$93.6B56.2B
15.A$94.6B54.4B14.2A$93.7B54.4B15.B$93.6B51.B2.6B14.3B$93.6B50.2AB.2B
2A2B13.6B$93.6B50.2A3BA2BA3B.B8.10B$92.8B50.2B.2B2A7B3.2B2.11B3.2B2.
2B$93.8B52.18BD3B2A15BD$25.2A65.9B53.17B2D2B2A15BDBD$6.2A18.A65.9B53.
18B2D18B3DB$7.A17.A66.10B52.18BD21BD$7.A.AB14.2A65.5B2A3B51.18BD21B$
8.2AB.2A12.B65.5B2A4B48.B.13B2.B4.13B.B$10.2B2AB11.3B63.11B47.2AB.12B
7.7B.B$9.3B2.B2.2B6.6B61.4BD7BA.2A42.2A4BD9B$10.10B4.10B60.B3D4B2.2A.
A43.2B.B3D7B$9.13B2.11B3.2B2.2B32.3B15.D2B2D2B35.3B15.D2B2D6B$9.15BD
3B2A15BD30.5B13.5B37.5B13.10B$9.15B2D2B2A15BDBD27.6B13.5B36.6B13.6B$
8.17B2D18B3DB26.8B10.7B35.8B10.7B$8.17BD21BD27.8B11.6B35.8B11.6B$8.
16BD21B23.B4.9B11.5B30.B4.9B11.5B$8.17B.13B.B27.3B3.11B6.8B29.3B3.11B
6.8B$6.27B.B31.6B2.13B2.10B27.6B2.13B2.10B$4B2.22B32.4B2.7B.13B.12B
20.4B2.7B.13B.12B$28B32.26B2.13B19.26B2.13B$BD27B31.BD25B.12B20.BD25B
.12B$BDBD26B30.BDBD23B.13B19.BDBD23B.13B$B3DB2A23B30.B3DB2A34B19.B3DB
2A34B$3BDB2A21B32.3BDB2A22BD13B17.3BDB2A22BD13B$29B31.29B2D9BA4B15.
29B2D9BA4B$2.26B34.28B2D7BABA4B16.28B2D7BABA4B$4.B2.19B38.B2.23BD8BAB
A4B18.B2.23BD8BABA4B$6.19B41.23BD10BA4B21.23BD10BA4B$5.21B39.24B2.B.
2B3.5B22.24B2.B.2B3.5B$4.9B.12B38.25B.3B7.B23.25B.3B7.B$3.4B.5B.12B
37.4B.20B2.B2AB5.3B21.4B.20B2.B2AB5.3B$2.4B2.4B3.9B.B2A34.4B2.18B5.2A
6.B2AB19.4B2.18B5.2A6.B2AB$.4B13.6B.BA.A32.4B4.18B.B11.2A19.4B4.18B.B
11.2A$4B14.6B4.A31.4B6.18B2A30.4B6.18B2A$3B15.5B5.2A30.3B7.18B2A30.3B
7.18B2A$2B17.3B38.2B8.3B3.13B31.2B8.3B3.13B$B19.2B38.B10.B2.3B.11B31.
B10.B2.3B.11B$19.4B51.2A2.9B47.2A2.9B$18.B2AB53.A3.7B49.A3.7B$19.2A
51.3A5.4B48.3A5.4B$72.A8.5B46.A8.5B$84.2A58.2A$84.A59.A$85.3A57.3A$
87.A59.A!

The same can be done for HFx197B (as shown above), HR203B, HL141B, HF94B, and HFx58B, so that's a dozen new conduits right there. I'd also like to add a dozen or so adjustments of F262+4N:

Code: Select all

x = 60, y = 46, rule = LifeHistory
55.A$53.3A$36.A15.A$36.3A13.2A4.2A$39.A9.5B4.A$38.2A3.B4.4B3.BA.A$38.
8B.7B.B2A$40.16B$40.16B$39.17B$24.2B11.18B$24.3B3.2B2.20B$24.13BC15B$
24.13BCBC13B$24.13B3C13B$24.15BC12B$24.19B3.6B$24.7B.8B6.7B$24.B.B4.
8B9.5B$30.8B11.4B$29.8B11.B2A2B$28.8B13.2A$27.8B$6.B2AB16.8B$7.BA7.2A
7.8B$8.ABA4.B2AB5.8B6.A$9.2A.B3.3B4.8B5.3A$10.5B2.B4.8B5.A$11.9B.8B6.
2A$8.3B.16B8.B$8.19B9.3B$.B3.21B9.6B$2AB.22B.B3.B2.10B$2A43B3.2B2.2B$
.B.31BD3B2A15BD$4.7B2A21B2D2B2A15BDBD$5.B.4B2A9B2D11B2D18B3DB$9.14B2D
10BD21BD$8.A3.11BD10BD21B$6.3A5.12B2.3B5.13B.B$5.A8.13B3.3B3.7B.B$5.
2A9.5B.4B5.2A$16.B2A2B4.2B4.A$17.2A5.BA2B4.3A$24.A.A7.A$25.A!

I could add B60, but it doesn't seem to have any likely output connections besides another B60 -- no point in slowing Hersrch down with that.

I suppose Lx321 is a decent Spartan Herschel splitter now, though kind of big and slow to recover -- not sure if it's worth adding a string of adjusted variants for that one:

Code: Select all

x = 60, y = 68, rule = LifeHistory
41.B$40.3D$40.BDB$39.2B3D$39.5B$39.6B$39.6B$39.5B$38.6B$39.6B$38.7B$
38.6B$38.6B$38.6B$37.8B$38.8B$37.9B$37.9B$37.10B$37.5B2A3B$37.5B2A4B$
37.11B$37.4BD7BA.2A$39.B3D4B2.2A.A$39.D2B2D2B$39.5B$31.2A5.6B$18.B13.
A6.6B$18.2B12.A.AB2.8B$18.3B12.2AB2.9B$18.4B13.12B$18.5B12.12B$18.6B
11.10B.B2A$18.7B11.8B2.BA.A$18.8B9.8B6.A$19.8B9.8B5.2A$20.8B9.7B$21.
8B5.11B$22.10B.12B$23.22B.B$24.24B$17.A7.9B3D13B$17.3A6.9BD14B$20.A6.
8B3D11B2A$7.2A10.2A6.20B.B2A$8.A10.5B2.19B4.B$8.A.AB9.25B12.2A$9.2AB.
3B4.25B8.B3.A.A$11.7B.9BD16B4.6B.B2A$11.14B2.B2D16B2.8B.B$12.13B2.2B
2D17B.6B.2B$11.18BD29B$9.19BD23B2A6B$7.18B.25BA2BA5B$7.2BC15B2.23B.B
2A5B$6.3BCBC4B.7B3.24B.7B$7.2B3C4B2.7B2.24B4.3B$6.5BC4B4.2B2A4.23B$5.
10B3.4B2A4.22B$4.4B10.2A8.5B.16B$4.3B12.A6.5B3.16B$2.4B10.3A7.2AB5.
14B$2.2A12.A10.A6.14B$3.A20.3A8.4B4.5B$3A21.A11.4B3.3B$A34.4B$34.B2A
2B$35.2A!

What other denizens of Sparta have showed up recently that Hersrch doesn't know about? I've done a quick sweep through the elementary conduits collection, but it's easy to miss interesting new combinations.

[chris_c]

What other denizens of Sparta have showed up recently that Hersrch doesn't know about? I've done a quick sweep through the elementary conduits collection, but it's easy to miss interesting new combinations.

Here are a few, all based on the alternative form of HFx58B:

Code: Select all

#C Lx204, Lx142 and Rx164
x = 201, y = 58, rule = LifeHistory
120.13B$119.15B$119.16B$119.17B$119.16B$121.13B$120.4B.2B2A5B$119.4B
2.2B2A3B$118.4B2.8B59.A$26.A8.B70.A10.4B4.2B3D3B58.3A$17.A8.3A5.3D60.
A8.3A7.4B5.3BD4B44.A16.A$17.3A9.A4.BDB60.3A9.A5.4B7.2B3D2B44.3A13.2A$
20.A7.2A3.2B3D62.A7.2A4.4B8.7B47.A12.4B$7.2A10.2A7.10B49.2A10.2A7.9B
9.6B35.2A10.2A8.2A4.B2A$8.A10.5B6.9B49.A10.5B6.6B10.6B36.A10.5B4.B2AB
2.BA.A$8.A.AB9.4B4.10B19.2A28.A.AB9.4B4.6B11.5B37.A.AB9.4B3.B.B3.B2A$
9.2AB.3B4.18B15.B3.A.A29.2AB.3B4.15B10.6B38.2AB.3B4.13B.3B$11.7B.9BD
13B7.6B.B2A32.7B.9BD13B4.6B39.7B.9BD7B$11.17B2D13B.2B2.8B.B33.17B2D
13B2.7B39.17B2D6B$12.17B2D17B.6B.2B34.17B2D12B2.6B41.17B2D6B$11.18BD
29B32.18BD13B.7B40.18BD7B$9.19BD23B2A6B29.19BD23B37.19BD2BD5B$7.18B.B
.23BA2BA5B27.18B.B.24B35.18B.B.3B3D3B$7.2BC15B4.21B.B2A5B28.2BC15B3.
6B2.17B34.2BC15B2.4B.D4B.BA$6.3BCBC4B.7B6.21B.7B27.3BCBC4B.7B5.6B3.
15B3.B29.3BC.C4B.7B3.11BA.A$7.2B3C4B2.7B3.5B.B.15B4.3B29.2B3C4B2.7B4.
5B5.15B.B2A29.2B3C4B2.7B3.6B.B.2BA$6.5BC4B4.2B2A4.2A5.16B35.5BC4B4.2B
2A6.2B6.18B2A28.5BC4B4.2B2A3.7B4.B$5.10B3.4B2A5.A6.14B35.10B3.4B2A5.
2A7.12B2A3B.B28.10B3.4B2A4.7B$4.4B10.2A6.3A9.12B34.4B10.2A9.A9.11B2A
2B30.4B10.2A8.7B$4.3B12.A6.A10.13B34.3B12.A7.A.A9.10B2.2B31.3B12.A8.
8B$2.4B10.3A16.2AB.B2.6B34.4B10.3A8.2A12.B.6B3.2B28.4B10.3A8.7B.B2A$
2.2A12.A17.A.AB4.6B34.2A12.A28.4B2.B2AB27.2A12.A11.6B.BA.A$3.A30.A8.
5B35.A40.B2A2B3.2A29.A24.6B4.A$3A30.2A10.B34.3A42.2A33.3A24.2B2D2BD4.
2A$A79.A79.A19.A.2A2.4B3DB$180.2A.A7BD4B$185.11B$185.4B2A5B$186.3B2A
5B$186.10B$187.9B$187.9B$187.8B$188.8B$189.6B$189.6B$189.6B$188.7B$
188.6B$189.6B$189.5B$188.6B$188.6B$189.5B$189.3D2B$190.BDB$190.3D$
191.B!

[Sokwe]

Here's a minor updated to Hersrch. I added the following fast conduits (repeat time less than 77): Bx202, Fx198, L200, R49, R152, Rx164. I also added the slower conduit Fx191 (repeat time 94).

Hersrch-15Jul2022.zip

(267.13 KiB) Downloaded 101 times

The following variants of Rx65 still need to be added:

Code: Select all

x = 174, y = 65, rule = B3/S23
b2ob2o2b2o41b2ob2o2b2o41b2ob2o2b2o41b2ob2o2b2o$2bobo2bobo3b2o37bobo2bo
bo3b2o37bobo2bobo3b2o37bobo2bobo3b2o$bo3b2o7bo36bo3b2o7bo8b2o26bo3b2o
7bo36bo3b2o7bo$2b2obo8b3o35b2obo8b3o6bo28b2obo8b3o35b2obo8b3o$3bobo2bo
44bobo2bo15b3o26bobo2bo44bobo2bo12b2o$2bo2bobobo42bo2bobobo11b3o2bo25b
o2bobobo42bo2bobobo11b2o$o2b3ob2o41bo2b3ob2o11bo3b3o23bo2b3ob2o41bo2b
3ob2o$3o47b3o17bob2o26b3o19b2o26b3o$3bo49bo17bobob2o26bo17bo2bo28bo18b
2o$2b2o18bo29b2o18bo2bo26b2o16bo4bo26b2o16bob2o$21b2o51bo2bo41bo6bo43b
o$70bo4b3o41bo6bo43bo$19b3o2bobo43b5o45bo4bo44bo2bo$21bob4o44bo3bo45bo
2bo46b2o$25b2o45b2obo46b2o$73bob2o$bo49bo20bo28bo49bo$bo19b2o28bo19bo
29bo19b2o28bo19b2o$b3o12b2obo2bo28b3o12b2obo2b3o26b3o12b2obo2bo28b3o
12b2obo2bo$3bo11bobob3o31bo11bobob3o2bo28bo11bobob3o31bo11bobob3o$15bo
bo47bobo4bo42bobo47bobo$13b3ob5o41b3ob5o41b3ob5o41b3ob5o$12bo3bo4bo40b
o3bo45bo3bo4bo40bo3bo4bo$12b2o2bobo43b2o2bob2o42b2o2bobo43b2o2bobo$17b
2o48bobo47b2o48b2o17$13b2o48b2o48b2o48b2o$14bo49bo8b2o39bo49bo$14b3o
47b3o6bo40b3o47b3o$74b3o94b2o$11bo49bo9b3o2bo34bo49bo9b2o$o12bo36bo12b
o6bo3b3o23bo12bo36bo12bo$3o6bo40b3o6bo10bob2o26b3o6bo12b2o26b3o6bo$3bo
7bo41bo7bo9bobob2o26bo7bo9bo2bo28bo7bo10b2o$2b2o3bo14bo29b2o3bo14bo2bo
26b2o3bo12bo4bo26b2o3bo12bob2o$9bo11b2o36bo14bo2bo31bo9bo6bo32bo10bo$
70bo4b3o41bo6bo43bo$19b3o2bobo43b5o45bo4bo44bo2bo$21bob4o44bo3bo45bo2b
o46b2o$25b2o45b2obo46b2o$73bob2o$bo49bo20bo28bo49bo$bo19b2o28bo19bo29b
o19b2o28bo19b2o$b3o12b2obo2bo28b3o12b2obo2b3o26b3o12b2obo2bo28b3o12b2o
bo2bo$3bo11bobob3o31bo11bobob3o2bo28bo11bobob3o31bo11bobob3o$15bobo47b
obo4bo42bobo47bobo$13b3ob5o41b3ob5o41b3ob5o41b3ob5o$12bo3bo4bo40bo3bo
45bo3bo4bo40bo3bo4bo$12b2o2bobo43b2o2bob2o42b2o2bobo43b2o2bobo$17b2o
48bobo47b2o48b2o!

I think the p5 and p8 variants should be direct replacements to existing conduits, but the p3 and p4 are not, as there are some cases where the existing p3 and p4 variants are better (in either bounding box or compatibility with other conduits).