(23,5)c/79 spaceship components

[codeholic]

This topic is a follow-up to http://conwaylife.com/forums/viewtopic. ... 8450#p8450

I found a couple of 3/2 track multiplicators. The distance between tracks is adjustable.

Code: Select all

x = 297, y = 481, rule = B3/S23
95bo$94b2o$94bobo22$53bo$52b2o$52bobo6$116b2o$116bobo$117bo2$111b2o$
110bobo$112bo3$2o$obo$o12$121b2o$121bobo$122bo10$77b2o$77bobo$77bo9$
126b2o$96b2o28bobo$97b2o28bo$96bo3$25b2o$24b2o$26bo$93bo$92bo2bo$92b2o
bo$92b3o$92bobo$91bo2bo$92bobo$94b2o$95bo$91b2o$91b2o$91b2o$92bob2o$
92b3o$93bo37b2o$131bobo$132bo7$126bo$125bobo$125bobo$126bo6$109bo$108b
2o$108bobo3$117bo18b2o$116bobo17bobo$116bobo18bo$49b3o62bo$49bo63bobo$
50bo65bo$115bo$113b2o2$131bo$130bobo$130bobo$131bo11$141b2o$141bobo$
142bo7$136bo$135bobo$135bobo$136bo10$75bo$74b2o$74bobo$139b3o$139bo$
140bo$145bo$143bo2bo$142bo4bo$141bo5bo$144b3o$142bo$135b2o5b2o$134bobo
11b2o$136bo10bo29$99b2o$99bobo$99bo10$120b2o49b2o$121b2o47b2o$120bo51b
o8$114bo$113bobo$112b2ob2ob2o$113bob2ob2o$113bo2$116bo$115b2o3$115bo$
115bo3bo$115bo2bo$114b2ob2o$115bobo$116bo15$131bo$130b2o$130bobo$195b
3o$195bo$196bo38$155b2o$155bobo$155bo65bo$220b2o$220bobo39$180b2o$179b
2o$181bo63b2o$245bobo$245bo39$204b3o$204bo$205bo64b2o$269b2o$271bo38$
230bo$229b2o$229bobo$294b3o$294bo$295bo38$254b2o$254bobo$254bo!

Code: Select all

x = 255, y = 431, rule = B3/S23
66b2o$66bobo$66bo24$42bo$41b2o$41bobo2$86b2o$86bobo$87bo4$79b2o$80b2o$
79bo15$91b2o$91bobo$92bo7$86bo$85bobo$85bobo$86bo2$b2o$2o$2bo63b2o$66b
obo$66bo5$76b2o18b2o$76b2o18bobo$75b2o20bo$64b3o7b2o$66bo7bobo$65bo10b
o$74bobo$74b2o2$91bo$90bobo$90bobo$91bo11$101b2o$101bobo$102bo7$96bo$
95bobo$25b3o67bobo$25bo70bo$26bo9$50bo47bo$50b2o45b2o$49bobo45bobo4$
106b2o$106b2o3$42bo54b2o5b2o$41bobo54b2o2b3o$97bo4bo$104b2o2$45bo2$42b
obo$42bobo8$44b3o$44bo$43b3o13$57b2o$56b2o$58bo10$82b3o43b3o$84bo43bo$
83bo45bo23$76bobo$75bo2bo$75bo2bo$74bo3bo$75bobo$76bo10$88b2o$88bobo$
88bo65bo$153b2o$153bobo39$113b2o$112b2o$114bo63b2o$178bobo$178bo39$
137b3o$137bo$138bo64b2o$202b2o$204bo38$163bo$162b2o$162bobo$227b3o$
227bo$228bo38$187b2o$187bobo$187bo65bo$252b2o$252bobo39$212b2o$211b2o$
213bo!

[codeholic]

A forward rake.

Code: Select all

x = 228, y = 395, rule = B3/S23
20b3o$20bo$21bo7$63bo$62bobo$63b2o17$35b3o$35b3o$33b2o$33bo2b3o$68bo$
33bo33bobo$33bo3b2o29b2o$34bo2bo$34bobo5$b2o$2o44bo$2bo42b2o$45bobo11$
73bo$72bobo$73b2o21$78bo$77bobo$78b2o4$25b3o$25bo44b2o$26bo43bobo$70bo
14$83bo$82bobo$83b2o4$85b3o$85bobo2$84bo4bo$91bo$87bob2o$86bo$87bo$87b
o3$86bo2b2o2$86bo3bo$90bo$87bobo2$87b3o2$51bo$50b2o$50bobo9$102bo$75b
2o24b2o$74bobo24bobo$76bo8$67bo$67bo14$70bo$69b2o$71bo$68b3o$69bo12$
81b3o$81bo44b2o$82bo43bobo$126bo39$107bo$106b2o$106bobo42b2o$150b2o$
152bo39$131b2o$131bobo$131bo43b3o$175bo$176bo39$156b2o$155b2o44bo$157b
o42b2o$200bobo40$180b3o$180bo44b2o$181bo43bobo$225bo39$206bo$205b2o$
205bobo!

[codeholic]

Another forward rake. (Because of the slope there are two types of forward rakes.)

Code: Select all

x = 194, y = 408, rule = B3/S23
44b2o$44b2o5$48b2o$48b2o6$34b3o$36bo$35bo8$49b2o$49b2o5$53b2o$53b2o16$
54b2o$54b2o5$58b2o$58b2o2$20bo$20b2o$19bobo11$61b2o$61b2o6$63b2o$63b2o
$46b2o$45bobo$47bo8$66b2o$66b2o$65b2o$66bobo$66b2o$67bo6$5b2o61b2o$4bo
bo61b2o$6bo3$39bo$5bo32b3o$2bobobo30bo$2b2obo30b2ob2o$37b2ob2o$37b2ob
2o$37bo2bo$37b3o$b3o34b2o$2ob2o$2obobo67b2o$bo3bo67b2o$bo2bo$2b3o2$48b
3o$48bo$49bo$73b2o$73b2o$57b2o$58bo2$55bobo2$53b3o$54bo2bo$55b2o5$80b
2o$80b2o3$19b3o48b2o$19bo51b2o$20bo49bo3$78b2o$68b2ob2o5b2o$67bo3bo$
70bo$67bo4$67b3o$67bo2b2o$68b3o$69bo3$85b2o$85b2o8$83b2o$83b2o8$84bo$
86bo$86bo$87b2o$86bobo$45bo41b3o$44b2o43bo$44bobo40b2o$87b2o5$93b2o$
92b2o$94bo2$75b2o$74bobo$76bo25$66bo2bo$64b4o$70bo$64b2o2b2o2$68b2o$
69b2o$65bo2b2o5$117b3o$117bo$75b3o40bo$75bo$76bo38$143bo$142b2o$101bo
40bobo$100b2o$100bobo39$167b2o$167bobo$125b2o40bo$125bobo$125bo39$192b
2o$191b2o$150b2o41bo$149b2o$151bo41$174b3o$174bo$175bo!

[Freywa]

Ivan Fomichev! The big problem is not how to construct additional tracks. The problem is the helix itself. Because this is an oblique reaction, spaceships would be very awkward. We'd need something else as a burner - maybe a Herschel? I have no idea.

[codeholic]

Whether it is awkward or not, there is no other known technology.

E. g. the following LWSS salvo (given it is infinite) can be considered as advancing (23,5)c/79 to the north-east. A burning helix of this velocity consisting of standard spaceships, if it exists, should be easily constructible by the same script that found the burning helix for the caterpillar.

Code: Select all

x = 44, y = 137, rule = B3/S23
3o$o2bo$o$o$bobo12$6bo$5b3o$4b2obo$4b3o$5b2o13$10b3o$9bo2bo$12bo$12bo$
9bobo12$16bo$15b3o$15bob2o$16b3o$16b2o13$20b3o$20bo2bo$20bo$20bo$21bob
o12$26bo$25b3o$24b2obo$24b3o$25b2o13$30b3o$29bo2bo$32bo$32bo$29bobo12$
36bo$35b3o$35bob2o$36b3o$36b2o13$40b3o$40bo2bo$40bo$40bo$41bobo!

[Freywa]

Whether it is awkward or not, there is no other known technology.

It doesn't work, and it's easy to prove it doesn't work.

Suppose the spaceship we are talking about has a "classical" helix, i.e. one like that used in the Caterpillar. Then, over one period (79 generations), the Herschel/glider track moves up 23 cells but the spaceships in the helix move up 39. Thus the helix moves closer to the track, eventually careening with it and destroying the spaceship.

But there is a solution: have the tracks "cycle" the helix by continually adding and removing spaceships from it, in such a way that the helix retains its position relative to the tracks. By judicious choice of helix shape, it should be possible to achieve the (23,5)c/79 velocity. The image linked here summarises my ideas and presents a vector-based approach to solving the helix problem. I myself call this kind of cycling helix a bus and the entire spaceship the bus service.

[codeholic]

It doesn't work, and it's easy to prove it doesn't work.

Suppose the spaceship we are talking about has a "classical" helix, i.e. one like that used in the Caterpillar. Then, over one period (79 generations), the Herschel/glider track moves up 23 cells but the spaceships in the helix move up 39. Thus the helix moves closer to the track, eventually careening with it and destroying the spaceship.

It proves nothing to me. My working assumption is, that after the first tracks start a little bit behind the burning front, they should spawn new tracks, whose beginnings loosely follow the slope of the spaceship stream. There is no place where tracks and the spaceship stream intersect.

But there is a solution: have the tracks "cycle" the helix by continually adding and removing spaceships from it, in such a way that the helix retains its position relative to the tracks. By judicious choice of helix shape, it should be possible to achieve the (23,5)c/79 velocity. The image linked here summarises my ideas and presents a vector-based approach to solving the helix problem. I myself call this kind of cycling helix a bus and the entire spaceship the bus service.

I think you don't quite understand, how the caterpillar works. It doesn't make sense to create a spaceship in the stream in the first place, if you destroy it later. The goal of the spaceship stream is to form a burning helix in the front (as well as other components, such as glider multipliers and adjustors), not just to be there for its own sake. I recommend you reading the article about the caterpillar by Gabriel Nivasch.

[DivusIulius]

I recommend you reading the article about the caterpillar by Gabriel Nivasch.

An outstanding reference, thanks!

[Extrementhusiast]

So we would need some sort of pushing reaction that would move at (23,5)c/79. (The reaction must also emit at least one glider.)

In other words: "Find a finite object that can annihilate all members of the given stream."

In other other words: "Find another (23,5)c/79 reaction supported from the back."

(Actually, a collision that generates a backward glider might be sufficient.)

[Freywa]

Ah, yes! Indeed the spaceships do move across the tracks without drifting away or crashing. The next question would be this: how would we make the spaceship train itself? We'd probably use the classic 3-glider syntheses, as unlike the caterpillar we're not limited to having the helix on the sides. Maybe a little "pocket" for this task, where the necessary components can be made freely without interfering with the tracks.

[Freywa]

The second problem to solve for this spaceship (after the helix) is how to clean the gliders at the back. Here's a preliminary construction.

Code: Select all

x = 459, y = 712, rule = B3/S23
417bo$415b2o$416b2o39$456bobo$456b2o$391bo65bo$391bobo$391b2o38$432bo$
431bo$431b3o$366bobo$366b2o$367bo38$408bo$406b2o$342bo64b2o$341bo$341b
3o39$382bo$382bobo$318bo63b2o$316b2o$317b2o39$357bobo$357b2o$292bo65bo
$292bobo$292b2o38$333bo$332bo$332b3o$267bobo$267b2o$268bo38$309bo$307b
2o$243bo64b2o$242bo$242b3o39$283bo$283bobo$219bo63b2o$217b2o$218b2o39$
258bobo$258b2o$193bo65bo$193bobo$193b2o38$234bo$233bo$233b3o$168bobo$
168b2o$169bo38$210bo$208b2o$144bo64b2o$143bo$143b3o39$184bo$184bobo$
120bo63b2o$118b2o$119b2o39$159bobo$159b2o$94bo65bo$94bobo$94b2o38$135b
o$134bo$134b3o$69bobo$69b2o$70bo38$111bo$109b2o$45bo64b2o$44bo$44b3o
39$85bo$85bobo$21bo63b2o$19b2o$20b2o5$14b2o$14b2o$16bo$14b3o$13bobo$
14b2o$13bo$13bo$11bo8$3o$3o$o2bo$b3o$b2o!

And since I got tired of pasting, here's a 5-glider track for reference.

Code: Select all

x = 102, y = 174, rule = B3/S23
101bo$99b2o$100b2o41$75bo$75bobo$75b2o41$50bobo$50b2o$51bo40$26bo$25bo
$25b3o41$2bo$2o$b2o!

[Extrementhusiast]

It can be done much more simply:

Code: Select all

x = 444, y = 704, rule = B3/S23
443bo$441b2o$442b2o2$406bo$404b2o$405b2o37$417bo$417bobo$417b2o2$380bo
$380bobo$380b2o37$392bobo$392b2o$393bo2$355bobo$355b2o$356bo36$368bo$
367bo$367b3o2$331bo$330bo$330b3o37$344bo$342b2o$343b2o2$307bo$305b2o$
306b2o37$318bo$318bobo$318b2o2$281bo$281bobo$281b2o37$293bobo$293b2o$
294bo2$256bobo$256b2o$257bo36$269bo$268bo$268b3o2$232bo$231bo$231b3o
37$245bo$243b2o$244b2o2$208bo$206b2o$207b2o37$219bo$219bobo$219b2o2$
182bo$182bobo$182b2o37$194bobo$194b2o$195bo2$157bobo$157b2o$158bo36$
170bo$169bo$169b3o2$133bo$132bo$132b3o37$146bo$144b2o$145b2o2$109bo$
107b2o$108b2o37$120bo$120bobo$120b2o2$83bo$83bobo$83b2o37$95bobo$95b2o
$96bo2$58bobo$58b2o$59bo36$71bo$70bo$70b3o2$34bo$33bo$33b3o37$47bo$45b
2o$46b2o2$10bo$8b2o$9b2o5$3b2o$3b2o$5bo$3b3o$2bobo$3b2o$2bo$2bo$o!

It works with most vanishing reactions.

EDIT: This could work for a back end:

Code: Select all

x = 937, y = 1248, rule = B3/S23
934bobo$934b2o$935bo2$897bobo$897b2o$898bo5$701bo$699b2o$700b2o29$910b
o$909bo$909b3o2$873bo$872bo$872b3o6$675bo$675bobo$675b2o29$886bo$884b
2o$885b2o2$849bo$847b2o$848b2o6$650bobo$650b2o$651bo29$860bo$860bobo$
860b2o2$823bo$823bobo$823b2o5$626bo$625bo$625b3o30$835bobo$835b2o$836b
o2$798bobo$798b2o$799bo5$602bo$600b2o$601b2o29$811bo$810bo$810b3o2$
774bo$773bo$773b3o6$576bo$576bobo$576b2o29$787bo$785b2o$786b2o2$750bo$
748b2o$749b2o6$551bobo$551b2o$552bo29$761bo$761bobo$761b2o2$724bo$724b
obo$724b2o5$527bo$526bo$526b3o30$736bobo$736b2o$737bo2$699bobo$699b2o$
700bo5$503bo$501b2o$502b2o29$712bo$711bo$711b3o2$675bo$674bo$674b3o6$
477bo$477bobo$477b2o29$688bo$686b2o$687b2o2$651bo$649b2o$650b2o6$452bo
bo$452b2o$453bo29$662bo$662bobo$662b2o2$625bo$625bobo$625b2o5$428bo$
427bo$427b3o30$637bobo$637b2o$638bo2$600bobo$600b2o$601bo5$404bo$402b
2o$403b2o29$613bo$612bo$612b3o2$576bo$575bo$575b3o6$378bo$378bobo$378b
2o29$589bo$587b2o$588b2o2$552bo$550b2o$551b2o6$353bobo$353b2o$354bo29$
563bo$563bobo$563b2o2$526bo$526bobo$526b2o5$329bo$328bo$328b3o30$538bo
bo$538b2o$539bo2$501bobo$501b2o$502bo5$305bo$303b2o$304b2o29$514bo$
513bo$513b3o2$477bo$476bo$476b3o6$279bo$279bobo$279b2o29$490bo$488b2o$
489b2o2$453bo$451b2o$452b2o6$254bobo$254b2o$255bo29$464bo$464bobo$464b
2o2$427bo$427bobo$427b2o5$230bo$229bo$229b3o30$439bobo$439b2o$440bo2$
402bobo$402b2o$403bo5$206bo$204b2o$205b2o29$415bo$414bo$414b3o2$378bo$
377bo$377b3o6$180bo$180bobo$180b2o29$391bo$389b2o$390b2o2$354bo$352b2o
$353b2o6$155bobo$155b2o$156bo29$365bo$365bobo$365b2o2$328bo$328bobo$
328b2o5$131bo$130bo187bo$130b3o183b6o$316b2o2b2o2$320bo$316b2o2bo$316b
2o3bo$320bo$120b2o196bobo$119b3o197bo$119bo2bo$119b2ob2o$119b2ob2o$
118b2ob2o$119bo$120b3o$121bo15$340bobo$327bo12b2o$328bo12bo$326b3o6$
129bo$127bobo$128b2o2$298bo$296b2o$297b2o6$99bobo$99b2o$100bo29$142bob
o$143b2o$143bo2$272bo$272bobo$272b2o5$75bo$74bo$74b3o29$158bo$159bo$
157b3o3$247bobo$247b2o$248bo5$51bo$49b2o$50b2o29$172bo$173b2o$172b2o2$
223bo$222bo$222b3o6$25bo$25bobo$25b2o29$188bo$186bobo$187b2o2$199bo$
197b2o$198b2o6$obo$2o$bo2$175bo$174b2o$174bobo2$149b3o$149bo$150bo$
125b2o$124b2o$126bo$100b2o$100bobo$100bo$76bo$75b2o$75bobo2$50b3o$50bo
$51bo$26b2o$25b2o$27bo$b2o$bobo$bo!

As for the front end, we could take the same approach that was used to build the Caterpillar, except for (23,5)c/79. (As for the period, I'm thinking p316.)

For reference:

Code: Select all

x = 334, y = 216, rule = B3/S23
271b3o$270bo2bo$273bo$269bo3bo$269bo3bo$273bo$160b3o107bobo$159bo2bo$
162bo$158bo3bo$162bo$159bobo2$b3o$o2bo$3bo$3bo273bo$obo273b3o$276bob2o
$277b3o$277b3o$277b3o$166bo110b2o$165b3o$165bob2o$166b3o$166b3o$166b2o
2$7bo$6b3o$6bob2o$7b3o$7b2o272b3o$281bo2bo$281bo$281bo3bo$281bo3bo$
281bo$170b3o109bobo$170bo2bo$170bo$170bo3bo$170bo$171bobo2$11b3o$11bo
2bo$11bo$11bo275bo$12bobo271b3o$285b2obo$285b3o$285b3o$285b3o$176bo
109b2o$175b3o$174b2obo$174b3o$174b3o$175b2o2$17bo$16b3o$15b2obo$15b3o$
16b2o273b3o$290bo2bo$293bo$289bo3bo$289bo3bo$293bo$180b3o107bobo$179bo
2bo$182bo$178bo3bo$182bo$179bobo2$21b3o$20bo2bo$23bo$23bo273bo$20bobo
273b3o$296bob2o$297b3o$297b3o$297b3o$186bo110b2o$185b3o$185bob2o$186b
3o$186b3o$186b2o2$27bo$26b3o$26bob2o$27b3o$27b2o272b3o$301bo2bo$301bo$
301bo3bo$301bo3bo$301bo$190b3o109bobo$190bo2bo$190bo$190bo3bo$190bo$
191bobo2$31b3o$31bo2bo$31bo$31bo275bo$32bobo271b3o$305b2obo$305b3o$
305b3o$305b3o$196bo109b2o$195b3o$194b2obo$194b3o$194b3o$195b2o2$37bo$
36b3o$35b2obo$35b3o$36b2o273b3o$310bo2bo$313bo$309bo3bo$309bo3bo$313bo
$200b3o107bobo$199bo2bo$202bo$198bo3bo$202bo$199bobo2$41b3o$40bo2bo$
43bo$43bo273bo$40bobo273b3o$316bob2o$317b3o$317b3o$317b3o$206bo110b2o$
205b3o$205bob2o$206b3o$206b3o$206b2o2$47bo$46b3o$46bob2o$47b3o$47b2o
272b3o$321bo2bo$321bo$321bo3bo$321bo3bo$321bo$210b3o109bobo$210bo2bo$
210bo$210bo3bo$210bo$211bobo2$51b3o$51bo2bo$51bo$51bo275bo$52bobo271b
3o$325b2obo$325b3o$325b3o$325b3o$216bo109b2o$215b3o$214b2obo$214b3o$
214b3o$215b2o2$57bo$56b3o$55b2obo$55b3o$56b2o273b3o$330bo2bo$333bo$
329bo3bo$329bo3bo$333bo$220b3o107bobo$219bo2bo$222bo$218bo3bo$222bo$
219bobo2$61b3o$60bo2bo$63bo$63bo$60bobo!

[codeholic]

For building the helix in the caterpillar they used a table of glider collisions with up to 2 standard spaceships. I wonder, if the helix could be made smaller if they used glider collisions with more spaceships, say, up to 3.

[codeholic]

There are two reactions that suppress a sideways glider by a sideways glider from a parallel track, that might be useful for track period adjustments.

Code: Select all

x = 30, y = 8, rule = B3/S23
10b2o16b2o$10b2o16b2o$20bo$b2o5b2o10b2o4b2o$obo5b2o9bobo4b2o$2bo7bo17b
o$8b2o16b2o$8b2o16b2o!

[oblique]

Part of the problem with this ship-to-be is, that it is not moving in a "normal" direction.

With the original carterpillar you could just send an LWSS (for example) alongside the ship and never had to fear that it will crash into it.

This monster here ist not even moving in the direction implied by the trail it's riding on!

I've tried to visualize this in the following patern:

Code: Select all

#CXRLE Pos=-736,-431
x = 1328, y = 770, rule = B3/S23
2o$obo$o8$528bo$527b3o$526b2obo$526b3o$527b2o11$43b2o$42b2o$44bo23$85b
3o$85bo$86bo3$550b3o$549bo2bo$552bo$552bo$549bobo15$129bo$128b2o$128bo
bo23$171b2o401bo$171bobo399b3o$171bo401bob2o$574b3o$574b2o21$214b2o$
213b2o$215bo18$596b3o$596bo2bo$596bo$596bo$597bobo$256b3o$256bo$257bo
22$300bo$299b2o$299bobo13$620bo$619b3o$618b2obo$618b3o$619b2o6$342b2o$
342bobo370b3o$342bo374bo$716bo22$385b2o332bo$384b2o333b2o$386bo331bobo
8$642b3o$641bo2bo$644bo$644bo$641bobo11$427b3o292b2o$427bo293bobo$428b
o294bo22$471bo$470b2o253b2o$470bobo253b2o$725bo2$666bo$665b3o$665bob2o
$666b3o$666b2o16$513b2o$513bobo212b3o$513bo216bo$729bo22$556b2o130b3o
41bo$555b2o131bo2bo40b2o$557bo130bo42bobo$688bo$689bobo21$598b3o134b2o
$598bo135bobo$599bo136bo17$712bo$711b3o$710b2obo$710b3o$711b2o$642bo$
641b2o95b2o$641bobo95b2o$738bo9$902b2o$901b4o$901b2ob2o$154b2o373b2o
372b2o$153b2ob2o371b2o$154b4o726bo2bo$155b2o731bo$884bo3bo$216b4o332b
2o331b4o$216bo3bo331b2o$216bo$217bo2bo649b2o$280b2o586b2ob2o$279b4o
292b2o107b2o182b4o$278b2ob2o292b2o107bobo54b3o125b2o$279b2o403bo58bo$
742bo109b4o$342bo2bo505bo3bo$341bo256b2o255bo$341bo3bo252b2o251bo2bo$
341b4o491b2o$835b4o$835b2ob2o$404b2o215b2o214b2o$403b2ob2o213b2o$404b
4o410bo2bo$405b2o415bo$818bo3bo$466b4o174b2o88b3o82b4o$466bo3bo173b2o
87bo2bo$466bo269bo$467bo2bo265bo67b2o$530b2o201bobo66b2ob2o$529b4o134b
2o133b4o$528b2ob2o134b2o134b2o$529b2o$786b4o$592bo2bo189bo3bo$591bo98b
2o35b2o16bo43bo$591bo3bo94b2o34b2o17b2o38bo2bo$591b4o133bo15bobo23b2o$
769b4o$769b2ob2o$654b2o57b2o56b2o$653b2ob2o55b2o$654b4o94bo2bo$655b2o
99bo$752bo3bo$716b4o16b2o15b4o$716bo3bo15b2o$716bo$717bo2bo5$726bobo
15bo$726b2o17b2o$727bo16b2o5$769b3o$737bobo29bo$736bo33bo$736bo$736bo
2bo$736b3o2$684bo$683bo58bo$683b3o54bobo$741b2o12$642bo$640b2o95bobo$
641b2o95b2o73bo$738bo73b2o$812bobo10$714b2o$598bo115b3o18bo$598bobo
112bob2o19bo$598b2o113b3o18b3o$714bo9$855b2o$855bobo$855bo$555bobo173b
o$555b2o175b2o$556bo174b2o12$513bo$512bo216bo$512b3o212bobo$728b2o$
689bobo$692bo$692bo$689bo2bo$690b3o205b2o$897b2o$899bo5$471bo$469b2o
253bobo$470b2o253b2o$725bo12$427bo294bo$427bobo293bo$427b2o292b3o$940b
3o$940bo$941bo4$667b2o$666b3o$666b2obo$667b3o$668bo2$384bobo331bo$384b
2o333b2o$385bo332b2o10$984bo$983b2o$342bo640bobo$341bo374bo$341b3o370b
obo$715b2o11$645bobo$644bo$644bo$644bo2bo$644b3o5$1026b2o$1026bobo$
1026bo23$1069b2o$622b2o444b2o$622b3o445bo$621bob2o$621b3o$622bo20$
1111b3o$1111bo$1112bo8$597bobo$600bo$600bo$597bo2bo$598b3o10$1155bo$
1154b2o$1154bobo19$575b2o$574b3o$574b2obo$575b3o$576bo620b2o$1197bobo$
1197bo23$1240b2o$1239b2o$1241bo3$553bobo$552bo$552bo$552bo2bo$552b3o
16$1282b3o$1282bo$1283bo13$530b2o$530b3o$529bob2o$529b3o$530bo5$1326bo
$1325b2o$1325bobo!

I used a small .py skript to create a block plus gliders and LWSS in all directions then iterate 79 generations and move the universe by (-23,-5)

Then I added some additional gliders to emphasize the trail.

[oblique]

As for the helix:

The point is, that the helix itself is not a "spaceship" it's more a carefully designed crawler-like reaction.

The carterpillar uses a series of *WSS convay vs. glider collisions which push the glider around in such a way that the resulting speed is that of the main ship body.

We would need to design our own convoy pattern.

On the plus side: because all the rakes creating the ships in this *WSS-streams are moving with the ship we would not have to worry about more then one iteration of the pattern. (Well. Probably we will have to filter out some of the gliders ... I don't think we will be able to find a p79 helix).

Speaking of findig ... I did not find the original pattern list nor the database or the scripts mentioned by Gabriel Nivasch on his articel about the Carterpillar. I fear they didn't publish it.

I used gencol to collided any combination of two *WSS and a glider. The result is a really *huge* matrix of litterally thousand of collisions - and yes I did filter for gilder-only results.

I will try to think of a way to catalogize them in a way that I could write a searching program. ...

As for the topic: "is it possible to reach the front without crashing the SS-stream into the glidertrails":

The constructions used by the original carterpillar need one or two (for the LWSS) glider salvos with gliders from two directions.

On one side of the ship we are able to collide the backrake of the base-crawler with the corresponding forward rake (the first one posted by codeholic viewtopic.php?f=2&t=1142#p8459)

If you run the pattern posted by him you will see such a collision after the trail is used up (and thus some gliders of the lower backrake are escaping).

The other forward rake can only be collided with inner parts of the ship itself. (Because the other backrake is in fact continuing the trail).

Hopefully we will be able to adjust the timing accordingly, but I am pretty sure that this can be done.

Thus we should be able to create an forward *WSS stream as far away from the trails as we need. (The "as far away part" is easy btw.: just move the forward rakes farter away from the backward ones).

If I did my math right (following in the next post for you to rip apart) the *WSS ships should be fast enough to out run the main ship.

Another if: if I get the article (http://www.gabrielnivasch.org/fun/life/caterpillar) right, all we need should be *one* glider any n*79 generations in front of the ship. We should be able to duplicate and rephase gliders as needed. (Just noticing: probably we will not be able to use the "delay reaction" http://www.gabrielnivasch.org/fun/life/ ... edirects=0, I fear).

[oblique]

Now for the speed calculations ...

Since we are colliding some forward and some backward rakes to form our ships, the important part are the backward ones: they have to be created closer to the front of the ship and they are travelling in the "wrong" direction.

Let's for the moment just look at one of the LWSS streams and only at the closest-to-the-front back rake of the construction of this stream (if we can do it once, we probably can do it again!)

Infact I will look at just *one* glider and *one* LWSS ...

OK. Let's say start by producing our glider directly at the closest-to-the-front rake.

Lets say the LWSS is created at a given number of t generations after the start.
For example t = 4*79 = 316

The glider has moved to (-79, 79) and the rake has advanced to (92, 20)
Let's say the LWSS is created at (-79, 79) [probably not excactly, but a constant displacement won't matter to much here]

The LWSS is moving stricly orthogonal; namely with (1,0)c/2

Therefore: it's a race between our rake at (92,20) moving (23,5)c/79 and the LWSS starting at (-79,79)

The distance at this point in time is thus (-171, 59).

The rake will need approx 12*79 (=948) ticks to cross the lane in which the LWSS is moving.
At this point in time the rake will be at (92+12*23, 20+12*5) and the LWSS at (-79+474, 79+0)

Since 92*12*23 = 368 and -79+474 = 395 the LWSS will have been the first.

Above that: we can allways move the rake further apart to give the LWSS yet even more time ...

Since we are trying to create a space ship all distances within our ship are (modulo some noise) fixed. Even If the rake we are considering is 50 million cells behind the edge of the ship and 20 millions below the the first trail to create ... those are just constants. We can allways move the back rakes further away ...

[oblique]

so close ...

I tried to reproduce effect of the basic reaction by using *WSS instead of a glider.

It WORKED!

If you run this:

Code: Select all

x = 15, y = 22, rule = B3/S23
3o$bo$b3o2$12bo$11b3o$11bob2o$12b3o$12b2o7$6b3o$6bo2bo$6bo$6bo3bo$6bo
3bo$6bo$7bobo!

for exactly 79 generation it reproduces the herschel in exactly the right spot.

But: the reaction is way to slow.

We could only use it for every other cycle:

Code: Select all

x = 24, y = 55, rule = B3/S23
3o$bo$b3o2$12bo$11b3o$11bob2o$12b3o$12b2o7$6b3o$6bo2bo$6bo$6bo3bo$6bo
3bo$6bo$7bobo6b3o$16bo2bo$16bo$16bo$17bobo6$12bo$11b3o$10b2obo$10b3o$
10b3o$10b3o$11b2o9bo$21b3o$20b2obo$20b3o$21b2o7$16b3o$15bo2bo$18bo$14b
o3bo$14bo3bo$18bo$15bobo!

(generate this for about 45-50 gens and you see the second flotilla crash into the unsettled first reaction)

Here is a not-quite-working example for using a glider for the second cyle ... It would work if we could get rid of the extra glider in generation 138

Code: Select all

x = 41, y = 113, rule = B3/S23
39bo$38bo$38b3o56$3o$bo$b3o2$12bo$11b3o$11bob2o$12b3o$12b2o7$6b3o$6bo
2bo$6bo$6bo3bo$6bo3bo$6bo$7bobo16$22bo$21b3o$20b2obo$20b3o$21b2o7$16b
3o$15bo2bo$18bo$14bo3bo$14bo3bo$18bo$15bobo!

Maybe someone finds a way to clean up the mess in pattern 2?

[oblique]

For building the helix in the caterpillar they used a table of glider collisions with up to 2 standard spaceships. I wonder, if the helix could be made smaller if they used glider collisions with more spaceships, say, up to 3.

This happened when I used gencols for crashing an additional *WSS into a collection of glider + 2*WSS patterns.

Code: Select all

time gencols -pat g+flotilla/g+2ss-13-30.list obj/allss_n.list -nph 4 -tc 5 14 -gen 50 -filt g >g+3ss.list
real    10131m4.710s
user    10099m32.195s
sys     3m4.812s

(Yep. That's just over a complete week of computing time ... compared to about 2-3 mins for the glider plus 2 ships version)
The resulting file contains about 250 000 patterns (including some duplicates )

By the way: I put the waiting to some good use ...
Does anybody care for a program that analyses such collisions and gerenerates SQL-Statements to pump then in a mysql-Database?

I could post the resulting sql, too ... but that would be around 7 MB (gzip compressed, uncompressed more like 115 Megs).

[triller]

I, too, have been looking at this leading edge of the helix (though with much less success than you've achieved).
Congrats.

It struck me while studying the progression of your pattern that devolving it some number of generations might prove useful. That point being the full formation of the b-hep. Your triple barrage evolves the b-hep @ gen 59, even though the second salvo has caused the recursive reaction to GTS. The SW glider fully forms @ gen 21, the first LWSS begins to react after gen 26 and the HWSS after 27 (this last without consideration of the block that would be formed by an evolving b-hep.

Considering the reaction from the b-hep, the block fully forms @ gen 15, the Herschel and block pair @ 20, and the glider @ 41. It then remains to re-create the b-hep at gen 79 using the slag at or near the time of glider formation PLUS whatever can be formed heading north and using the block. Of course any reaction using the block would have to allow the glider to escape.

EDIT added pic

evolve-b.png (8.28 KiB) Viewed 61911 times

I'll continue my efforts from this end.
triller

[codeholic]

so close ...

I tried to reproduce effect of the basic reaction by using *WSS instead of a glider.

It WORKED!

Great idea! How did you find this collision?

[oblique]

so close ...

I tried to reproduce effect of the basic reaction by using *WSS instead of a glider.

It WORKED!

Great idea! How did you find this collision?

Well ... I figured it might be worth it.
If we *could* create a trail by just 1 simple p79 reaction the rest of the ship might be several times easier to build.
And since the base reaction is just a herschel + one single glider ...

I used gencols to generate collisions between a herschel an up to 2 space ships.

Then I used a modification of my own pattern-cataloguing program to find one that regeneration the herschel at (23,5) @ t=79

BTW: for one arcane reason I seem unable to find even the base reaction with gencols itself.

I used something like

gencols -pat herschel.life obj/glider.list -nph 4 -tc 5 80 -gen 80 -nosynch -test1 herschel79.life -filt ap

(I tried all kinds of variantions with or without the -nosynch and with any kind of filter ...)

herschel.life contains:

Code: Select all

#P
..*
***
*.*
*

herschel79.life contains:

Code: Select all

#P
.
.
.
.
.
.........................*
.......................***
.......................*.*
.......................*

The method should be simliar to what the bhep-halfpuffer example included with gencols does.

[oblique]

I found a helix ....

Its p316, alas

Code: Select all

#CXRLE Pos=-154,-5
x = 161, y = 406, rule = B3/S23
3o$o2bo$o$o3bo$o3bo$o5b2o$bobo2bobo$6bo4$7bo$6b3o$5b2obo$5b3o$5b3o$5b
3o$6b2o26$5bo$4b3o44b3o$3b2obo43bo2bo4b3o$3b3o47bo3bo2bo$3b3o43bo3bo6b
o$3b3o43bo3bo2bo3bo$4b2o8bo38bo2bo3bo$13b3o34bobo7bo$13bob2o40bobo$14b
3o$14b2o32bo$47b3o$47bob2o$48b3o$48b2o6$33b3o$33bo2bo$33bo$20b3o10bo$
20bo2bo10bobo$20bo$20bo3bo16b3o$20bo3bo9bo6bo2bo$20bo12b3o5bo$21bobo8b
2obo5bo3bo$32b3o6bo3bo$33b2o6bo$42bobo2$27bo$26b3o$25b2obo$25b3o$25b3o
$25b3o$26b2o26$25bo$24b3o44b3o$23b2obo43bo2bo4b3o$23b3o47bo3bo2bo$23b
3o43bo3bo6bo$23b3o43bo3bo2bo3bo$24b2o8bo38bo2bo3bo$33b3o34bobo7bo$33bo
b2o40bobo$34b3o$34b2o32bo$67b3o$67bob2o$68b3o$68b2o6$53b3o$53bo2bo$53b
o$40b3o10bo$40bo2bo10bobo$40bo$40bo3bo16b3o$40bo3bo9bo6bo2bo$40bo12b3o
5bo$41bobo8b2obo5bo3bo$52b3o6bo3bo$53b2o6bo$62bobo2$47bo$46b3o$45b2obo
$45b3o$45b3o$45b3o$46b2o26$45bo$44b3o44b3o$43b2obo43bo2bo4b3o$43b3o47b
o3bo2bo$43b3o43bo3bo6bo$43b3o43bo3bo2bo3bo$44b2o8bo38bo2bo3bo$53b3o34b
obo7bo$53bob2o40bobo$54b3o$54b2o32bo$87b3o$87bob2o$88b3o$88b2o6$73b3o$
73bo2bo$73bo$60b3o10bo$60bo2bo10bobo$60bo$60bo3bo16b3o$60bo3bo9bo6bo2b
o$60bo12b3o5bo$61bobo8b2obo5bo3bo$72b3o6bo3bo$73b2o6bo$82bobo2$67bo$
66b3o$65b2obo$65b3o$65b3o$65b3o$66b2o26$65bo$64b3o44b3o$63b2obo43bo2bo
4b3o$63b3o47bo3bo2bo$63b3o43bo3bo6bo$63b3o43bo3bo2bo3bo$64b2o8bo38bo2b
o3bo$73b3o34bobo7bo$73bob2o40bobo$74b3o$74b2o32bo$107b3o$107bob2o$108b
3o$108b2o6$93b3o$93bo2bo$93bo$80b3o10bo$80bo2bo10bobo$80bo$80bo3bo16b
3o$80bo3bo9bo6bo2bo$80bo12b3o5bo$81bobo8b2obo5bo3bo$92b3o6bo3bo$93b2o
6bo$102bobo2$87bo$86b3o$85b2obo$85b3o$85b3o$85b3o$86b2o26$85bo$84b3o
44b3o$83b2obo43bo2bo4b3o$83b3o47bo3bo2bo$83b3o43bo3bo6bo$83b3o43bo3bo
2bo3bo$84b2o8bo38bo2bo3bo$93b3o34bobo7bo$93bob2o40bobo$94b3o$94b2o32bo
$127b3o$127bob2o$128b3o$128b2o6$113b3o$113bo2bo$113bo$100b3o10bo$100bo
2bo10bobo$100bo$100bo3bo16b3o$100bo3bo9bo6bo2bo$100bo12b3o5bo$101bobo
8b2obo5bo3bo$112b3o6bo3bo$113b2o6bo$122bobo2$107bo$106b3o$105b2obo$
105b3o$105b3o$105b3o$106b2o26$105bo$104b3o44b3o$103b2obo43bo2bo4b3o$
103b3o47bo3bo2bo$103b3o43bo3bo6bo$103b3o43bo3bo2bo3bo$104b2o8bo38bo2bo
3bo$113b3o34bobo7bo$113bob2o40bobo$114b3o$114b2o32bo$147b3o$147bob2o$
148b3o$148b2o6$133b3o$133bo2bo$133bo$133bo$134bobo2$141b3o$134bo6bo2bo
$133b3o5bo$132b2obo5bo3bo$132b3o6bo3bo$133b2o6bo$142bobo!

The next (fairly easy) step is to use the reactions of the original caterpillar (or something like them) to fill in the gaps and to build at least one other trails (codeholic found ways to create more the trails if you have at least two of them).

The next BIG problem will be how create p316 rakes ...

(Using theese to create the spaceships should be not-to-hard)

[Extrementhusiast]

Not bad! What's interesting is that the HWSS-LWSS salvo that collides with the block can be delayed for however long is necessary.

Now for the fanout devices.

[codeholic]

I found a helix ....

Congrats! Very well done! It seems that you found a way to make use of g+3ss collisions. Just out of curiosity, have you tried to find a shorter helix for the caterpillar using the extended collision database?

The next (fairly easy) step is to use the reactions of the original caterpillar (or something like them) to fill in the gaps and to build at least one other trails (codeholic found ways to create more the trails if you have at least two of them).

It is quite probable, that two initial tracks won't suffice, because there has to be a mechanism for adjusting track periods and it cannot be done as easy as in the caterpillar because of sideways gliders that Herschels emit compared to clean pies travelling along blinker tracks.

The next BIG problem will be how create p316 rakes ...

When building the caterpillar they minimized the number of long-period rakes by using kickback reactions. E. g. the following is LWSS synthesis taken from robocat.zip

Code: Select all

x = 452, y = 790, rule = B3/S23
319bo$320bo$318b3o8$73bo$74b2o$73b2o$8bo$9b2o$8b2o6$138bobo$139b2o$
139bo5$223bo$221bobo$222b2o67$53bo$53b2o$52bobo22$15b2o$16b2o$15bo9$
120b2o$121b2o$120bo21$83bo$83b2o$82bobo$448b3o$448bo2bo$448bo$448bo$
449bobo4$188bo$188b2o$187bobo2$387bo$385bobo$386b2o8$140bobo$141b2o$
141bo$75bobo$76b2o$76bo5$150b2o54bo$151b2o54b2o239b3o$150bo55b2o239bo
2bo$450bo$450bo$447bobo2$290bo$291bo$289b3o2$255b2o$256b2o$255bo21$
218bo229b3o$218b2o228bo2bo$217bobo228bo$448bo$449bobo7$323bo$323b2o$
322bobo20$448b3o$447bo2bo$285b2o163bo$286b2o162bo$285bo161bobo9$390b2o
$391b2o$390bo18$448b3o$448bo2bo$448bo$448bo$449bobo12$420b3o$420bo$
421bo15$448b3o$447bo2bo$450bo$450bo$447bobo13$353b2o$353bobo$353bo7$
208bo$209b2o$208b2o$143bo$144b2o$143b2o2$448b3o$448bo2bo$448bo$448bo$
273bobo173bobo$274b2o$274bo4$281bobo$281b2o$282bo7$285b3o$285bo$286bo
12$448b3o$447bo2bo$450bo$450bo$447bobo29$448b3o$448bo2bo$448bo$448bo$
449bobo29$448b3o$447bo2bo$450bo$450bo$447bobo13$b3o$3bo$2bo14$448b3o$
448bo2bo$448bo$448bo$449bobo14$69b2o$68bobo$70bo11$275bobo$276b2o$276b
o171b3o$210bobo234bo2bo$211b2o237bo$211bo238bo$447bobo4$341bo$342b2o$
341b2o3$343bo$341bobo$342b2o5$136b3o$138bo$137bo11$448b3o$448bo2bo$
448bo$448bo$449bobo17$204b2o$b2o200bobo$obo202bo$2bo9$448b3o$447bo2bo$
450bo$450bo$447bobo19$271b3o$68b3o202bo$70bo201bo$69bo7$448b3o$448bo2b
o$448bo$448bo$449bobo21$136b2o$135bobo$137bo6$448b3o$447bo2bo$450bo$
450bo$447bobo23$203b3o$205bo$204bo4$448b3o$297bo150bo2bo$297bobo148bo$
297b2o149bo$449bobo$278bo$279b2o$278b2o6$408bobo$409b2o$409bo2$410bo$
411bo$409b3o9$271b2o$270bobo$272bo3$448b3o$447bo2bo$450bo$450bo$447bob
o20$344b3o$346bo$345bo4$338b3o$340bo$339bo$448b3o$448bo2bo$448bo$448bo
$449bobo21$412b2o$411bobo$413bo4$406b2o$405bobo$407bo40b3o$447bo2bo$
450bo$450bo$447bobo!