Half-bakery reaction with glider

For discussion of specific patterns or specific families of patterns, both newly-discovered and well-known.
User avatar
dvgrn
Moderator
Posts: 10610
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Half-bakery reaction with glider

Post by dvgrn » July 6th, 2014, 7:10 pm

chris_c wrote:I think any two 90-turners should be able to make any desired 180-turner (of one particular color) . Increasing the gap between the 90-turners requires an extra HB for one of the outbound gliders, which gives an odd delay. In this example input gliders of delay 100 on the same lane are converted into gliders of delay 101 on the same lane...
Oh, it's definitely true that any two 90-turners will make a 180-turner -- but they won't always make an 180-turner with a cheap monochromatic recipe. Your first 90-turner was "anti", or color-changing, and the right one was "same", or color-preserving... and it just so happened that one of your 90-edge-shooters needed a "same" input and one needed an "anti" input, so it worked out nicely.

If you try using the alternate edge-shooter with either of the 90-turners in the reference pattern, you'll find that you end up having to build the edge-shooter with the wrong color gliders. (Unless I got confused and put everything together backward, which is always quite possible. Please check!)

P.S. Can someone also give an independent opinion on the feasibility of the MWSS+G trick for the A1 glider? It still looks to me as if that will be much cheaper than a superslow elbow.

User avatar
codeholic
Moderator
Posts: 1147
Joined: September 13th, 2011, 8:23 am
Location: Hamburg, Germany

Re: Half-bakery reaction with glider

Post by codeholic » July 6th, 2014, 7:23 pm

dvgrn wrote:Can someone also give an independent opinion on the feasibility of the MWSS+G trick for the A1 glider? It still looks to me as if that will be much cheaper than a superslow elbow.
I modified the script so that it could look for *WSS recipes, but it seems it's not optimised for performance enough (maybe yet). It may be easier, though, to find a recipe that emits 2 NW gliders and a LWSS recipe, that could be used in combination to produce a SW glider. I'm investigating these options.
Ivan Fomichev

chris_c
Posts: 966
Joined: June 28th, 2014, 7:15 am

Re: Half-bakery reaction with glider

Post by chris_c » July 7th, 2014, 5:36 am

chris_c wrote:Increasing the gap between the 90-turners requires an extra HB for one of the outbound gliders, which gives an odd delay
dvgrn wrote:but they won't always make an 180-turner with a cheap monochromatic recipe
I am talking about shifiting the distance between the 90-turners by 1 step diagonally so the color does not change. Here is a more complete example of what I mean. Everything is monochromatic but in the left hand pattern the 90-edge shooter is (1,1) closer to the 90-turner and an extra HB has been added. The "trigger" glider in both patterns is the 9th glider that goes into the turner. In the initial pattern they are separated by 300 lanes and in the same phase. After 1500 generations they have been converted to SW gliders, they are still 300 lanes apart, but they have opposite phase.

Code: Select all

x = 600, y = 396, rule = B3/S23
206b2o298b2o$206b2o298b2o13$196b3o297b3o$198bo299bo$197bo299bo4$208b2o
298b2o$208b2o298b2o7$184b3o297b3o$186bo299bo$185bo299bo22$180b3o297b3o
$182bo299bo$181bo299bo17$163bo299bo$163b2o298b2o$162bobo297bobo12$295b
2o$295b2o299b2o$596b2o$118b3o297b3o$120bo299bo$119bo299bo15$297b2o$
297b2o299b2o$115bo299bo182b2o$115b2o298b2o$114bobo297bobo19$102b3o297b
3o$104bo299bo$103bo299bo12$281b2o$280bo2bo298b2o$280bobo298bo2bo$278b
2obo299bobo$238b3o36bo2bo298b2obo$85bo154bo36bobo105bo153b3o36bo2bo$
85b2o152bo38bo106b2o154bo36bobo$84bobo297bobo153bo38bo7$269b2o$268bo2b
o$268bobo$266b2obo$226b3o36bo2bo$228bo36bobo259b3o$227bo38bo262bo$528b
o5$70b3o297b3o$72bo299bo$71bo299bo10$200b3o23b2o$202bo22bobo273b3o23b
2o$201bo25bo275bo22bobo$502bo25bo2$27b3o297b3o$29bo299bo$28bo299bo3$
198b3o$200bo298b3o$199bo301bo$500bo4$17b3o297b3o$19bo299bo$18bo299bo4$
195b2o$194bobo299b2o$196bo298bobo$497bo9$199b2o$198bobo299b2o$200bo
298bobo$501bo8$3o297b3o$2bo299bo$bo299bo14$166b2o$165bobo299b2o$167bo
298bobo$468bo14$157b2o$156bobo299b2o$158bo298bobo$459bo106$32b3o$34bo
298b3o$33bo301bo$334bo9$22b3o$24bo298b3o$23bo301bo$324bo9$12b3o$14bo
298b3o$13bo301bo$314bo!

User avatar
dvgrn
Moderator
Posts: 10610
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Half-bakery reaction with glider

Post by dvgrn » July 7th, 2014, 7:31 am

chris_c wrote:
chris_c wrote:Increasing the gap between the 90-turners requires an extra HB for one of the outbound gliders, which gives an odd delay
dvgrn wrote:but they won't always make an 180-turner with a cheap monochromatic recipe
I am talking about shifiting the distance between the 90-turners by 1 step diagonally so the color does not change.
Sure, that makes sense. Sorry, I was probably reading too much into your original statement that "any two 90-turners should be able to make any desired 180-turner". There are pairings of a 90-turner with a 90-edge-shooter that can't be used, because the two pieces would have to be constructed with different glider colors.

Once you have a matching turner-plus-edge-shooter pair, you certainly can solve any possible phasing problems. Some output phases will need slightly wider shooting ranges, up to +14hd or +7fd, I guess (half-diagonal/full diagonal).

I think I'll try doing everything with 180-turners first, now that we have such a big collection of them. It would be nice to be able to fit all slow-salvo constructions into a single 39-lane-plus shooting range -- i.e., to never (or seldom) need a double reset to shoot a glider at each target.

39 lanes is quite a bit too narrow for eight simultaneous targets, I suspect. That's assuming two targets for glider(s) and a *WSS for A1, plus one target each for the six gliders in the SE. But we can stretch the range a bit, by never pairing a far-NW glider in the first recipe in a scan with a far-SE glider in the last recipe. There's lots of flexibility as far as skipping gliders or rearranging them between scans.

So I guess I'll have a quick look what the narrowest compatible set of 180-turners will be. Maybe it will be clear that we'll end up needing a double-width shooting range anyway, and then the 90-turners might come in very handy indeed.
codeholic wrote:I modified the script so that it could look for *WSS recipes, but it seems it's not optimised for performance enough (maybe yet). It may be easier, though, to find a recipe that emits 2 NW gliders and a LWSS recipe, that could be used in combination to produce a SW glider. I'm investigating these options.
It would be pretty easy to use chris_c's (-3,-6) mechanism to manually string together a workable MWSS seed. -- blocks are easy to build even with monochromatic salvos:

Code: Select all

#C Blockic forward MWSS seed
x = 53, y = 33, rule = B3/S23
50b2o$50bobo$51bo2$45b2o$44bobo$46bo8$9b2o4b2o$9b2o4b2o5$16b2o$16b2o9$
2o$b2o$o!
Here's the source of the MWSS seed. I guess it would be nice to have a two-NW-glider turner here, too -- no particular synchronization needed, just anything that can shoot down the extra block before it gets in the way of a future MWSS+G glider.

User avatar
codeholic
Moderator
Posts: 1147
Joined: September 13th, 2011, 8:23 am
Location: Hamburg, Germany

Re: Half-bakery reaction with glider

Post by codeholic » July 7th, 2014, 8:47 am

dvgrn wrote:Here's the source of the MWSS seed.
Well, I believe the origin of this seed is http://conwaylife.com/forums/viewtopic.php?p=8849#p8849 :)

I'll use the blockic seed as a last resort. I've been hoping to find a "natural" monochromatic recipe, because that would probably need much less scans than indirect seed construction.
Ivan Fomichev

chris_c
Posts: 966
Joined: June 28th, 2014, 7:15 am

Re: Half-bakery reaction with glider

Post by chris_c » July 7th, 2014, 11:40 am

dvgrn wrote:Once you have a matching turner-plus-edge-shooter pair, you certainly can solve any possible phasing problems. Some output phases will need slightly wider shooting ranges, up to +14hd or +7fd, I guess (half-diagonal/full diagonal).

I think I'll try doing everything with 180-turners first, now that we have such a big collection of them. It would be nice to be able to fit all slow-salvo constructions into a single 39-lane-plus shooting range -- i.e., to never (or seldom) need a double reset to shoot a glider at each target.
Yes, two 90-turners are going to be wider I suppose, but how many of the 180-turners will turn out to require 3 resets or less when the need to move by (-3,-6) is taken into account? I've not looked properly at the other 180-turners to be honest.

I like the *WSS's ideas you guys have been coming up with. As an alternative I have been re-investigating NW salvos that give a (3,6) junk push and a SW glider. After modifying codeholics search scripts I found the following seven glider salvo that does the trick

Code: Select all

x = 138, y = 116, rule = B3/S23
2o$2o9$14b3o$14bo$15bo18$26b3o$26bo$27bo8$45b3o$45bo$46bo8$63b3o$63bo$
64bo12$68b3o$68bo$69bo47$105b3o27b3o$105bo29bo$106bo29bo!
That's quite a nice improvement on the 11 gliders in my very first post but the *WSS method looks like it could still end up the winner. The first 5 gliders in the salvo were found by computer search and the two clean up gliders were added by me. I do not intend to let the search move on to 6 glider salvos because searching through all 5 glider salvos is taking a long time and I estimate the probability that a 6 glider salvo gives a (3,6) block push and a SW glider _and_ requires no clean up is going to be small.

User avatar
codeholic
Moderator
Posts: 1147
Joined: September 13th, 2011, 8:23 am
Location: Hamburg, Germany

Re: Half-bakery reaction with glider

Post by codeholic » July 7th, 2014, 6:08 pm

A variant of construction of the blockic seed took 6 (?) scans and 17 gliders from a loaf without cleanup and restoring the target at (-3, -6).

Code: Select all

x = 459, y = 477, rule = B3/S23
456bo$455bobo$455bo2bo$456b2o22$437b3o$439bo$438bo48$391b3o$393bo$392b
o23$374b3o$376bo$375bo22$356bo$356b2o$355bobo23$335bo$335b2o$334bobo
23$258bo$258b2o$257bobo24$244b3o$246bo$245bo23$247b3o$249bo$248bo22$
229bo$229b2o$228bobo48$173bo$173b2o$172bobo24$129b3o$131bo$130bo23$
112b3o$114bo$113bo23$91b3o$93bo$92bo23$70b3o$72bo$71bo22$44bo$44b2o$
43bobo24$3o$2bo$bo22$2bo$2b2o$bobo!
Ivan Fomichev

User avatar
dvgrn
Moderator
Posts: 10610
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Half-bakery reaction with glider

Post by dvgrn » July 7th, 2014, 6:25 pm

codeholic wrote:
dvgrn wrote:Here's the source of the MWSS seed.
Well, I believe the origin of this seed is http://conwaylife.com/forums/viewtopic.php?p=8849#p8849 :)
Hmm, well, maybe I was a little lazy about looking up the original source...

I had a quick look at (-3,-6) block moves from Glue, by the way, but haven't found anything better than what chris_c has already put together. Here's a four-scan recipe, for example:

Code: Select all

x = 163, y = 162, rule = LifeHistory
161.2E$161.2E14$141.2A$140.A.A$142.A12$131.2A$130.A.A$132.A18$121.A$
121.2A$120.A.A3$108.2A$107.A.A$109.A25$81.A$81.2A$80.A.A35$36.A$36.2A
$35.A.A23$13.2A$12.A.A$14.A14$.A$.2A$A.A!
I should probably do the rest of the coding to sort out how many scans each of the known monochromatic recipes actually costs. There's a pile of probably-useless monochromatic recipes for constellations in the linked post now, along with the original table of block moves, in case that's useful to anyone.
codeholic wrote:I'll use the blockic seed as a last resort. I've been hoping to find a "natural" monochromatic recipe, because that would probably need much less scans than indirect seed construction...

A variant of construction of the blockic seed took 6 (?) scans and 17 gliders from a loaf without cleanup and restoring the target at (-3, -6).
That sounds about right. So far I had managed a 16-glider construction starting from a block, also not counting the (-3,-6) target push:

Code: Select all

#C non-HBK-compatible MWSS seed recipe
x = 2223, y = 2235, rule = B3/S23
2o$2o19$15b3o$15bo$16bo148$165b3o$165bo$166bo148$314b2o$313b2o$315bo
148$468b2o$467b2o$469bo148$602b2o$601b2o$603bo148$757b3o$757bo$758bo
148$922b2o$921b2o$923bo117$1014b3o$1014bo$1015bo148$1164b3o$1164bo$
1165bo148$1324b3o$1324bo$1325bo148$1461b2o$1460b2o$1462bo148$1613b2o$
1612b2o$1614bo148$1764b3o$1764bo$1765bo148$1912b3o$1912bo$1913bo148$
2054b3o$2054bo$2055bo141$2220b3o$2220bo$2221bo!
Everything works out quite nicely in this recipe, except that the trigger glider has to be on the wrong color...!

Probably something similar could be dug up with a good trigger glider. I have a couple of dozen possible recipes that I'd have to sort through. Most likely I won't experiment with them any more, unless it seems like it might be useful.

User avatar
codeholic
Moderator
Posts: 1147
Joined: September 13th, 2011, 8:23 am
Location: Hamburg, Germany

Re: Half-bakery reaction with glider

Post by codeholic » July 7th, 2014, 6:54 pm

A clean seed with 6 (?) scans and 17 gliders from a traffic light, that is perhaps better than starting with a loaf.

Code: Select all

x = 412, y = 424, rule = B3/S23
407bo$407bo$407bo2$403b3o3b3o2$407bo$407bo$407bo17$377b3o$379bo$378bo
23$356b3o$358bo$357bo23$335b3o$337bo$336bo23$318b3o$320bo$319bo23$297b
3o$299bo$298bo22$243bo$243b2o$242bobo24$225b3o$227bo$226bo22$223bo$
223b2o$222bobo24$183b3o$185bo$184bo22$143bo$143b2o$142bobo23$126bo$
126b2o$125bobo24$106b3o$108bo$107bo22$72bo$72b2o$71bobo24$46b3o$48bo$
47bo22$34bo$34b2o$33bobo24$16b3o$18bo$17bo19$bo$b2o$obo!
Ivan Fomichev

User avatar
calcyman
Moderator
Posts: 2932
Joined: June 1st, 2009, 4:32 pm

Re: Half-bakery reaction with glider

Post by calcyman » July 7th, 2014, 7:13 pm

Everything works out quite nicely in this recipe, except that the trigger glider has to be on the wrong color...!
No, the trigger glider is the only thing on the correct colour; the other fifteen gliders are wrong. The following diagram shows that the parities work out correctly for regenerating the SW-directed `A1' glider in the correct location:

Code: Select all

x = 339, y = 460, rule = LifeHistory
286.2B$285.4B$284.5B.B$284.9B$284.10B$283.11B$282.13B2.5B$281.24B$
280.24B2A$279.25B2A$278.27B$277.4B2.19B$276.4B2.16B2.2B$275.4B4.15B$
274.4B4.6B4.6B$273.4B6.6B4.6B$272.4B6.6B8.4B$271.4B8.6B8.4B$270.4B8.
6B10.4B$269.4B10.6B10.4B$268.4B10.6B12.4B$267.4B12.6B12.4B$266.4B12.
6B14.4B$265.4B14.6B14.4B$264.4B14.6B16.4B$263.4B16.6B16.4B$262.4B16.
6B18.4B$261.4B18.6B18.4B$260.4B18.6B20.4B$259.4B20.6B20.4B$258.4B20.
6B22.4B$257.4B22.6B22.4B$256.4B22.6B24.4B$255.4B24.6B24.4B$254.4B24.
6B26.4B$253.4B26.6B26.4B$252.4B26.6B28.4B$251.4B28.6B28.4B$250.4B28.
6B30.4B$249.4B30.6B30.4B$248.4B5.4B21.6B32.4B$247.4B5.4B23.6B32.4B$
246.4B5.4B23.6B34.4B$245.4B5.4B25.6B34.4B$244.4B5.4B25.6B36.4B$243.4B
5.4B27.6B36.4B$242.4B5.4B27.6B38.4B$241.4B5.4B29.6B38.4B$240.4B5.4B
29.6B40.4B$239.4B5.4B31.6B40.4B$238.4B5.4B31.6B42.4B$237.4B5.4B33.6B
42.4B$236.4B5.4B33.6B44.4B$235.4B5.4B35.6B44.4B$234.4B5.4B35.6B46.4B$
233.4B5.4B37.6B46.4B$232.4B5.4B37.6B48.3B$231.4B5.4B39.6B48.B$230.4B
5.4B39.6B$229.4B5.4B41.6B$228.4B5.4B41.6B$227.4B5.4B43.6B$226.4B5.4B
43.6B$225.4B5.4B45.6B$224.4B5.4B45.6B$223.4B5.4B47.6B$222.4B5.4B47.6B
$221.4B5.4B49.6B$220.4B5.4B49.6B$219.4B5.4B51.6B$218.4B5.4B51.6B$217.
4B5.4B53.6B$216.4B5.4B53.6B$215.4B5.4B55.6B$214.4B5.4B55.6B$213.4B5.
4B57.6B$212.4B5.4B57.6B$211.4B5.4B59.6B$210.4B5.4B59.6B$209.4B5.4B61.
6B$208.4B5.4B61.6B$207.4B5.4B63.6B$206.4B5.4B63.6B$205.4B5.4B65.6B$
204.4B5.4B65.6B$203.4B5.4B67.6B$202.4B5.4B67.6B$201.4B5.4B69.6B$200.
4B5.4B69.6B$199.4B5.4B71.6B$198.4B5.4B71.6B$197.4B5.4B73.6B$196.4B5.
4B73.6B$195.4B5.4B75.6B$194.4B5.4B75.6B$193.4B5.4B77.6B$192.4B5.4B77.
6B$191.4B5.4B79.6B$190.4B5.4B79.6B$189.4B5.4B81.6B$188.4B5.4B81.6B$
187.4B5.4B83.6B$186.4B5.4B83.6B$185.4B5.4B85.6B$184.4B5.4B85.6B$183.
4B5.4B87.6B$182.4B5.4B87.6B$181.4B5.4B89.6B$180.4B5.4B89.6B$179.4B5.
4B91.6B$178.4B5.4B91.6B$177.4B5.4B93.6B$176.4B5.4B93.6B$175.4B5.4B95.
6B$174.4B5.4B95.6B$173.4B5.4B97.6B$172.4B5.4B97.6B$171.4B5.4B99.6B$
170.4B5.4B99.6B$169.4B5.4B101.6B$168.4B5.4B101.6B$167.4B5.4B103.6B$
166.4B5.4B103.6B$165.4B5.4B105.6B$164.4B5.4B105.6B$163.4B5.4B107.6B$
162.4B5.4B107.6B$161.4B5.4B109.6B$160.4B5.4B109.6B$159.4B5.4B111.6B$
158.4B5.4B111.6B$157.4B5.4B113.6B$156.4B5.4B113.6B$155.4B5.4B115.6B$
154.4B5.4B115.6B$153.4B5.4B117.6B$152.4B5.4B117.6B$151.4B5.4B119.6B$
150.4B5.4B119.6B$149.4B5.4B121.6B$148.4B5.4B121.6B$147.4B5.4B123.6B$
146.4B5.4B123.6B$145.4B5.4B125.6B$144.4B5.4B125.6B$143.4B5.4B127.6B$
142.4B5.4B127.6B$141.4B5.4B129.6B$140.4B5.4B129.6B$139.4B5.4B131.6B$
138.4B5.4B131.6B$137.4B5.4B133.6B$136.4B5.4B133.6B$135.4B5.4B135.6B$
134.4B5.4B135.6B$133.4B5.4B137.6B$132.4B5.4B137.6B$131.4B5.4B139.6B$
130.4B5.4B139.6B$129.4B5.4B141.6B$128.4B5.4B141.6B$127.4B5.4B143.6B$
126.4B5.4B143.6B$125.4B5.4B145.6B$124.4B5.4B145.6B$123.4B5.4B147.6B$
122.4B5.4B147.6B$121.4B5.4B149.6B$120.4B5.4B149.6B$119.4B5.4B151.6B$
118.4B5.4B151.6B$117.4B5.4B153.6B$116.4B5.4B153.6B$115.4B5.4B155.6B$
114.4B5.4B155.6B$113.4B5.4B157.6B$112.4B5.4B157.6B$111.4B5.4B159.6B$
110.4B5.4B159.6B$109.4B5.4B161.6B$108.4B5.4B161.6B$107.4B5.4B163.6B$
106.4B5.4B163.6B$105.4B5.4B165.6B$104.4B5.4B165.6B$103.4B5.4B167.6B$
102.4B5.4B167.6B$101.4B5.4B169.6B$100.4B5.4B169.6B$99.4B5.4B171.6B$
98.4B5.4B171.6B$97.4B5.4B173.6B$96.4B5.4B173.6B$95.4B5.4B175.6B$94.4B
5.4B175.6B$93.4B5.4B177.6B$92.4B5.4B177.6B$91.4B5.4B179.6B$90.4B5.4B
179.6B$89.4B5.4B181.6B$88.4B5.4B181.6B$87.4B5.4B183.6B$86.4B5.4B183.
6B$85.4B5.4B185.6B$84.4B5.4B185.6B$83.4B5.4B187.6B$82.4B5.4B187.6B$
81.4B5.4B189.6B$80.4B5.4B189.6B$79.4B5.4B191.6B$78.4B5.4B191.6B$77.4B
5.4B193.6B$76.4B5.4B193.6B$75.4B5.4B195.6B$74.4B5.4B195.6B$73.4B5.4B
197.6B$72.4B5.4B197.6B$71.4B5.4B199.6B$70.4B5.4B199.6B$69.4B5.4B201.
6B$68.4B5.4B201.6B$67.4B5.4B203.6B$66.4B5.4B203.6B$65.4B5.4B205.6B$
64.4B5.4B205.6B$63.4B5.4B207.6B$62.4B5.4B207.6B$61.4B5.4B209.6B$60.4B
5.4B209.6B$59.4B5.4B211.6B$58.4B5.4B211.6B$57.4B5.4B213.6B$56.4B5.4B
213.6B$55.4B5.4B215.6B$54.4B5.4B215.6B$53.4B5.4B217.6B$52.4B5.4B217.
6B$51.4B5.4B219.6B$50.4B5.4B219.6B$49.4B5.4B221.6B$48.4B5.4B221.6B$
47.4B5.4B223.6B$46.4B5.4B223.6B$45.4B5.4B225.6B$44.4B5.4B225.6B$43.4B
5.4B227.6B$42.4B5.4B227.6B$41.4B5.4B229.6B$40.4B5.4B229.6B$39.4B5.4B
231.6B$38.4B5.4B231.6B$37.4B5.4B233.6B$36.4B5.4B233.6B$35.4B5.4B235.
6B$34.4B5.4B235.6B$33.4B5.4B237.6B$32.4B5.4B237.6B$31.4B5.4B239.6B$
30.4B5.4B239.6B$29.4B5.4B241.6B$28.4B5.4B241.6B$27.4B5.4B243.6B$26.4B
5.4B243.6B$25.4B5.4B245.6B$24.4B5.4B245.6B$23.4B5.4B247.6B$22.4B5.4B
247.6B$21.4B5.4B249.6B$20.4B5.4B249.6B$20.3B5.4B251.6B$20.2B5.4B251.
6B$26.4B253.6B$25.4B253.6B$24.4B255.6B$23.4B255.6B$22.4B257.6B$21.4B
257.6B$20.4B259.6B$17.6B259.6B$10.2A5.5B261.6B$9.A2BA3.8B258.6B$9.ABA
B2.10B258.6B$7.2ABA3B.10B257.6B$6.A2BA4B.11B257.6B$6.ABA18B255.6B$7.A
5B2D13B255.6B$9.3BD2BD13B253.6B$9.3BDBD7B.7B253.6B$8.2B2DBD8B2.7B251.
6B$7.2BD2BD9B3.7B251.6B$8.BDBD10B4.7B249.6B$8.2BD11B5.7B249.6B$7.14B
7.7B247.6B$6.13B10.7B247.6B$5.11B14.7B245.6B$4.11B16.7B245.6B$3.4B.5B
19.7B243.6B$2.4B3.3B21.7B243.6B$.4B29.7B241.6B$4B31.7B241.6B$3B33.7B
239.6B$2B35.7B239.6B$B37.7B237.6B$39.7B237.6B$2B38.7B235.6B$3B38.7B
235.6B$3B39.7B233.6B$4B39.7B233.6B$5B39.7B231.6B$6B39.7B231.6B$7B39.
7B229.6B$.7B39.7B229.6B$2.7B39.7B227.6B$3.7B39.7B227.6B$4.7B39.7B225.
6B$5.7B39.7B225.6B$6.7B39.7B223.6B$7.7B39.7B223.6B$8.7B39.7B221.6B$9.
7B39.7B221.6B$10.7B39.7B219.6B$11.7B39.7B219.6B$12.7B39.7B217.6B$13.
7B39.7B217.6B$14.7B39.7B215.6B$15.7B39.7B215.6B$16.7B39.7B213.6B$17.
7B39.7B213.6B$18.7B39.7B211.6B$19.7B39.7B211.6B$20.7B39.7B209.6B$21.
7B39.7B209.6B$22.7B39.7B207.6B$23.7B39.7B207.6B$24.7B39.7B205.6B$25.
7B39.7B205.6B$26.7B39.7B203.6B$27.7B39.7B203.6B$28.7B39.7B201.6B$29.
7B39.7B201.6B$30.7B39.7B199.6B$31.7B39.7B199.6B$32.7B39.7B197.6B$33.
7B39.7B197.6B$34.7B39.7B195.6B.2B$35.7B39.7B195.8B$36.7B39.7B193.9B$
37.7B39.7B193.8B$38.7B39.7B192.8B$39.7B39.7B193.6B$40.7B39.7B191.8B$
41.7B39.7B189.9B$42.7B39.7B187.4B2.2B$43.7B39.7B185.4B$44.7B39.7B183.
4B$45.7B39.7B181.4B$46.7B39.7B179.4B$47.7B39.7B177.4B$48.7B39.7B175.
4B$49.7B39.7B173.4B$50.7B39.7B171.4B$51.7B39.7B169.4B$52.7B39.7B167.
4B$53.7B39.7B165.4B$54.7B39.7B163.4B$55.7B39.7B161.4B$56.7B39.7B159.
4B$57.7B39.7B157.4B$58.7B39.7B155.4B$59.7B39.7B153.4B$60.7B39.7B151.
4B$61.7B39.7B149.4B$62.7B39.7B147.4B$63.7B39.7B145.4B$64.7B39.7B143.
4B$65.7B39.7B141.4B$66.7B39.7B139.4B$67.7B39.7B137.4B$68.7B39.7B135.
4B$69.7B39.7B133.4B$70.7B39.7B131.4B$71.7B39.7B129.4B$72.7B39.7B127.
4B$73.7B39.7B125.4B$74.7B39.7B123.4B$75.7B39.7B121.4B$76.7B39.7B119.
4B$77.7B39.7B117.4B$78.7B39.7B115.4B$79.7B39.7B113.4B$80.7B39.7B111.
4B$81.7B39.7B109.4B$82.7B39.7B107.4B$83.7B39.7B105.4B$84.7B39.7B103.
4B$85.7B39.7B101.4B$86.7B39.7B99.4B$87.7B39.7B97.4B$88.7B39.7B95.4B$
89.7B39.7B93.4B$90.7B39.7B91.4B$91.7B39.7B89.4B$92.7B39.7B87.4B$93.7B
39.7B85.4B$94.7B39.7B83.4B$95.7B39.7B81.4B$96.7B39.7B79.4B$97.7B39.7B
77.4B$98.7B39.7B75.4B$99.7B39.7B73.4B$100.7B39.7B71.4B$101.7B39.7B69.
4B$102.7B39.7B67.4B$103.7B39.7B65.4B$104.7B39.7B63.4B$105.7B39.7B61.
4B$106.7B39.7B59.4B$107.7B39.7B57.4B$108.7B39.7B55.4B$109.7B39.7B53.
4B$110.7B39.7B51.4B$111.7B39.7B49.4B$112.7B39.7B47.4B$113.7B39.7B45.
4B$114.7B39.7B43.4B$115.7B39.7B41.4B$116.7B39.7B39.4B$117.7B39.7B37.
4B$118.7B39.7B35.4B$119.7B39.7B33.4B$120.7B39.7B31.4B$121.7B39.7B29.
4B$122.7B39.7B27.4B$123.7B39.7B25.4B$124.7B39.7B23.4B$125.7B39.7B21.
4B$126.7B39.7B19.4B$127.7B39.7B17.4B$128.7B39.7B15.4B$129.7B39.7B13.
4B$130.7B39.7B11.4B$131.7B39.7B9.4B$132.7B39.7B7.4B$133.7B39.9B3.4B$
134.7B39.14B$135.7B39.12B$136.7B39.10B$137.7B39.10B$138.7B39.10B$139.
7B39.10B$140.7B37.12B$141.7B37.12B$142.7B37.12B$143.7B38.11B$144.7B
37.3B2.7B$145.7B42.7B$146.7B42.7B$147.7B42.7B$148.7B42.7B$149.7B42.7B
!
Also, the NW-directed glider that collides with the N-directed MWSS cannot be generated by a boat reflector, since it is on the wrong parity.

Finally, we needn't worry about the leftover block, since it can be trivially shot down by a NW-directed glider from the B, A2 or C trails, provided we do the edge-shooting 180° reflector malarkey for the B/A2/C side before producing the MWSS+G collision for the A1 side.

And to re-iterate, the gliders feeding the B/A2/C side should be done from the outside inwards, i.e. C first. This circumvents any awkward crossing problems.

So we just need some pre-processing bolted onto the beginning of knightlife's 17-glider MWSS recipe to generate the initial traffic light whilst cleanly displacing a block by (-3, -6).
What do you do with ill crystallographers? Take them to the mono-clinic!

chris_c
Posts: 966
Joined: June 28th, 2014, 7:15 am

Re: Half-bakery reaction with glider

Post by chris_c » July 8th, 2014, 12:01 am

Here is an MWSS in six scans with reconstruction included:

Code: Select all

x = 592, y = 615, rule = B3/S23
568b2o$568b2o3$588b2o$588b2o15$570b2o$570b2o3$548b3o39b2o$550bo39b2o$
549bo12$536b3o$538bo$537bo18$538b3o$540bo$539bo12$526b3o$528bo$527bo
18$526b3o$528bo$527bo18$514b3o$516bo$515bo17$497bo$497b2o$496bobo34$
451b3o$453bo$452bo18$433b3o$435bo$434bo18$417b3o$419bo$418bo18$409b3o$
411bo$410bo19$366b3o$368bo$367bo32$312b3o$314bo$313bo17$295bo$295b2o$
294bobo19$276b3o$278bo$277bo17$259bo$259b2o$258bobo19$240b3o$242bo$
241bo78$144b3o$146bo$145bo10$134b3o$136bo$135bo10$124b3o$126bo$125bo
10$118b3o$120bo$119bo12$118b3o$120bo$119bo6$68b3o$70bo$69bo6$118b3o$
120bo$119bo19$47b3o$49bo$48bo27$30b3o$32bo$31bo23$17b3o$19bo$18bo27$3o
$2bo$bo!
It's annoying that one entire scan is spent to delete a single block but I couldn't find any way that is better. The block to two block transformation on the right and the rather large block move on the left were both found courtesy of codeholic's search program (which I find pretty much invaluable, thanks!)

User avatar
dvgrn
Moderator
Posts: 10610
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Half-bakery reaction with glider

Post by dvgrn » July 8th, 2014, 11:44 am

chris_c wrote:Here is an MWSS in six scans with reconstruction included...
Looks good. So now we just need to build this in parallel with a six-scans-or-less color-preserving 90-turner, adjusted to put the glider in the right place relative to the MWSS. I think there are enough degrees of freedom for the placement of the target block that the timing won't be an issue:

Code: Select all

x = 167, y = 191, rule = LifeHistory
162.2D$162.2D5$165.2C$165.2C7$161.3C$163.C$162.C11$150.C$150.2C$149.C
.C3$139.C$139.2C$138.C.C8$137.C$137.2C$136.C.C21$126.C$126.2C$125.C.C
13$99.3C$101.C$100.C20$81.3C$83.C$82.C17$60.3C$62.C$61.C19$53.C$53.2C
$52.C.C4$21.C$21.2C$20.C.C38$.C$.2C$C.C!
I've done a quick check on glider parities and colors, but could have gotten something wrong -- an independent review would be appreciated as usual.

It might be about time to write a compiler script that can merge a set of parallel recipes, and produce an actual HBK body pattern that shoots those recipes. Anyone care to volunteer? I haven't been able to decide on a good input format for the script -- can't even agree with myself, let alone anyone else...!

EDIT: Tried my hand at combining the two recipes --

Code: Select all

x = 1082, y = 1225, rule = LifeHistory
$929.D.D$929.2D$930.D55$1077.2D$1077.2D5$1080.2C$1080.2C153$882.D3$
885.D3$888.D3$891.D3$894.D2$946.2C$897.D48.2C3$900.D65.2C$966.2C2$
903.D3$906.D3$909.D3$912.D3$915.D$948.2C$948.2C$918.D2$968.2C$921.D
46.2C$890.3C$892.C$891.C32.D3$927.D3$930.D3$933.D2$879.2C$880.2C54.D$
879.C2$939.D3$942.D3$945.D3$948.D3$951.D3$954.D3$957.D3$960.D3$963.D
3$966.D3$969.D3$972.D6$868.2A$869.2A$868.A12$856.3A$858.A$857.A18$
858.2A$859.2A$858.A12$846.3A$848.A$847.A$732.D3$735.D3$738.D3$741.D3$
744.D3$747.D2$846.3A$750.D97.A$847.A2$753.D3$756.D3$759.D3$762.D3$
765.D3$768.D$834.3A$836.A$771.D63.A2$738.2C$739.2C33.D$738.C2$777.D3$
780.D3$736.2C45.D$737.2C$736.C$786.D3$789.D26.2A$817.2A$816.A$792.D3$
795.D3$798.D3$801.D3$804.D2$725.2C$726.2C79.D$725.C2$810.D3$813.D3$
816.D3$819.D3$822.D62$711.2A$712.2A$711.A9$561.D3$564.D3$567.D3$570.D
122.3A$695.A$694.A$573.D3$576.D3$579.D3$582.D3$585.D3$588.D2$677.3A$
591.D87.A$678.A2$594.D3$597.D3$600.D3$603.D3$606.D3$609.D$669.3A$671.
A$612.D57.A3$615.D3$618.D3$621.D3$624.D3$558.3C66.D$560.C$559.C$630.D
3$633.D3$636.D3$639.D3$642.D3$645.D3$648.D$540.3C$542.C$541.C109.D39$
426.D3$429.D3$432.D3$435.D3$438.D3$441.D3$444.D3$447.D3$450.D3$453.D
3$456.D3$459.D3$462.D3$465.D3$468.D3$471.D3$474.D3$477.D38.3A$518.A$
517.A$480.D3$483.D$439.3C$441.C$440.C45.D3$489.D3$492.D3$495.D3$498.D
3$501.D3$504.D$432.2C$433.2C$432.C74.D3$510.D3$513.D3$516.D45$412.2A$
413.2A$412.A18$394.3A$396.A$395.A16$241.D2$376.2A$244.D132.2A$376.A2$
247.D3$250.D3$253.D3$256.D3$259.D3$262.D$358.3A$360.A$265.D93.A3$268.
D3$271.D3$274.D3$277.D3$280.D3$283.D56.2A$341.2A$340.A$286.D3$239.D3.
D2.2D2.D3.D.4D29.D$239.2D.2D.D2.D.D3.D.D$239.D.D.D.D2.D.D3.D.D$239.D
3.D.D2.D2.D.D2.3D33.D$239.D3.D.D2.D2.D.D2.D$239.D3.D.D2.D3.D3.D$239.D
3.D2.2D4.D3.4D35.D3$298.D2$249.D.D$247.D5.D47.D2$246.D3.2C2.D$249.C.C
52.D$246.D4.C2.D2$247.D5.D53.D$249.D.D2$310.D3$313.D3$316.D3$319.D3$
322.D3$325.D3$328.D3$331.D12$230.2C$231.2C$230.C68$194.3A$196.A$195.A
10$184.2A$185.2A$184.A8$51.D2$174.3A$54.D121.A$175.A2$57.D3$60.D3$63.
D2$168.2A$66.D102.2A$168.A2$69.D3$72.D3$75.D3$78.D$168.3A$170.A$81.D
87.A3$84.D3$87.D3$90.D2$168.D.D$93.D72.D5.D2$165.D2.3A2.D$96.D73.A$
165.D3.A3.D2$99.D66.D5.D$168.D.D2$102.D3$105.D3$108.D3$111.D3$114.D3$
117.D3$120.D3$123.D3$126.D3$129.D3$132.D3$135.D2$68.2A$69.2A67.D$68.A
2$141.D25$47.2A$48.2A$47.A27$30.2A$31.2A$30.A23$17.2A$18.2A$17.A27$2A
$.2A$A!
Dividing lines have been added between the six scans, and the two key gliders are circled. The glider that triggers the 90-turner is labeled "MOVE" -- it's in the wrong phase, and therefore needs to be sent during a different scan. Not sure yet whether it will be the one before or after the current scan; it will be easier to sort out the math when we have an actual HBK body producing all of these gliders. Even without the final adjustments to get the timing right, the current MWSS seed and 90-turner together take up most of a minimum 39-lane shooting range.

It still seems more efficient to do all the constructions in parallel, even if we have to do a double or triple or quadruple reset (for shooting range width 157, 235, or 313hd). But just to keep the design simpler, it may make sense to do these six resets first (i.e., singleton resets, producing a 79hd / 39fd shooting range wherever we turn out to need it) and then shift the shooting range over to deal with the six 180-turners for B/A2/C. (?)

User avatar
calcyman
Moderator
Posts: 2932
Joined: June 1st, 2009, 4:32 pm

Re: Half-bakery reaction with glider

Post by calcyman » July 8th, 2014, 7:53 pm

The glider that triggers the 90-turner is labeled "MOVE" -- it's in the wrong phase, and therefore needs to be sent during a different scan.
Can't we simply change the phase of the MOVE glider by one tick (mod 2), change the phase of all of the green gliders by one tick (mod 2), and translate all of the green gliders by 1fd south-east?
What do you do with ill crystallographers? Take them to the mono-clinic!

User avatar
dvgrn
Moderator
Posts: 10610
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Half-bakery reaction with glider

Post by dvgrn » July 8th, 2014, 9:03 pm

calcyman wrote:
The glider that triggers the 90-turner is labeled "MOVE" -- it's in the wrong phase, and therefore needs to be sent during a different scan.
Can't we simply change the phase of the MOVE glider by one tick (mod 2), change the phase of all of the green gliders by one tick (mod 2), and translate all of the green gliders by 1fd south-east?
Hmm. Yes. Almost. You also have to move the MOVE glider 1fd northeast, because otherwise it's too slow to catch the LWSS (because the LWSS travels twice as fast as the gliders):

Code: Select all

#C revised combined recipe pattern
x = 1081, y = 1224, rule = LifeHistory
929.D$929.D.D$929.2D54$1076.2D$1076.2D5$1079.2C$1079.2C153$881.D3$
884.D3$887.D3$890.D3$893.D3$896.D49.2C$946.2C2$899.D$966.2C$966.2C$
902.D3$905.D3$908.D3$911.D3$914.D2$948.2C$917.D30.2C3$920.D47.2C$889.
3C76.2C$891.C$890.C32.D3$926.D3$929.D3$932.D2$878.2C$879.2C54.D$878.C
2$938.D3$941.D3$944.D3$947.D3$950.D3$953.D3$956.D3$959.D3$962.D3$965.
D3$968.D3$971.D7$868.3A$870.A$869.A11$857.A$857.2A$856.A.A19$858.3A$
860.A$859.A11$847.A$847.2A$846.A.A$731.D3$734.D3$737.D3$740.D3$743.D
3$746.D2$847.A$749.D97.2A$846.A.A2$752.D3$755.D3$758.D3$761.D3$764.D
3$767.D$835.A$835.2A$770.D63.A.A2$737.2C$738.2C33.D$737.C2$776.D3$
779.D3$735.2C45.D$736.2C$735.C$785.D3$788.D$816.3A$818.A$791.D25.A3$
794.D3$797.D3$800.D3$803.D2$724.2C$725.2C79.D$724.C2$809.D3$812.D3$
815.D3$818.D3$821.D63$711.3A$713.A$712.A8$560.D3$563.D3$566.D3$569.D
124.A$694.2A$693.A.A$572.D3$575.D3$578.D3$581.D3$584.D3$587.D2$678.A$
590.D87.2A$677.A.A2$593.D3$596.D3$599.D3$602.D3$605.D3$608.D$670.A$
670.2A$611.D57.A.A3$614.D3$617.D3$620.D3$623.D3$557.3C66.D$559.C$558.
C$629.D3$632.D3$635.D3$638.D3$641.D3$644.D3$647.D$539.3C$541.C$540.C
109.D39$425.D3$428.D3$431.D3$434.D3$437.D3$440.D3$443.D3$446.D3$449.D
3$452.D3$455.D3$458.D3$461.D3$464.D3$467.D3$470.D3$473.D3$476.D40.A$
517.2A$516.A.A$479.D3$482.D$438.3C$440.C$439.C45.D3$488.D3$491.D3$
494.D3$497.D3$500.D3$503.D$431.2C$432.2C$431.C74.D3$509.D3$512.D3$
515.D46$412.3A$414.A$413.A17$395.A$395.2A$394.A.A16$240.D3$243.D132.
3A$378.A$377.A$246.D3$249.D3$252.D3$255.D3$258.D3$261.D$359.A$359.2A$
264.D93.A.A3$267.D3$270.D3$273.D3$276.D3$279.D3$282.D$340.3A$342.A$
285.D55.A3$288.D3$291.D3$294.D3$297.D3$248.D.D49.D$246.D5.D2$245.D2.
2C3.D49.D$249.2C$245.D2.C4.D$306.D$246.D5.D$248.D.D$309.D3$312.D3$
315.D3$318.D3$321.D3$324.D3$327.D3$330.D12$229.2C$230.2C$229.C68$195.
A$195.2A$194.A.A11$184.3A$186.A$185.A7$50.D2$175.A$53.D121.2A$174.A.A
2$56.D3$59.D3$62.D3$65.D102.3A$170.A$169.A$68.D3$71.D3$74.D3$77.D$
169.A$169.2A$80.D87.A.A3$83.D3$86.D3$89.D2$168.D.D$92.D73.D5.D2$165.D
3.A3.D$95.D73.2A$165.D2.A.A2.D2$98.D67.D5.D$168.D.D2$101.D3$104.D3$
107.D3$110.D3$113.D3$116.D3$119.D3$122.D3$125.D3$128.D3$131.D3$134.D
3$68.3A66.D$70.A$69.A$140.D26$47.3A$49.A$48.A27$30.3A$32.A$31.A23$17.
3A$19.A$18.A27$3A$2.A$.A!
But since these aren't real slow recipes produced by an HBK spaceship body, some retiming will have to be done anyway when we get to that point.

EDIT: So here's one minimum step for adjusting the spacing between the two recipes, without changing the timing of the output glider: move the 90-turner recipe (white gliders) 4 cells to the left, and move the white block northwest, 2 cells diagonally.

Moving the output lane 1fd closer to the recipe gliders just requires moving the whole 90-turner recipe (including the white block target) southwest by 1fd. Conversely, of course, moving it all northeast 1fd pushes the output glider lane away 1fd NW. Shouldn't be too hard to pick a convenient offset that allows both circled gliders to be generated by HBK rakes.

chris_c
Posts: 966
Joined: June 28th, 2014, 7:15 am

Re: Half-bakery reaction with glider

Post by chris_c » July 9th, 2014, 9:03 pm

I got the MWSS recipe down to 5 scans and 25 gliders:

Code: Select all

x = 736, y = 787, rule = B3/S23
710b2o$710b2o19$694b3o15b2o$696bo15b2o$695bo10$732b2o$732b2o$682b3o$
684bo$683bo16$734b2o$734b2o$682b3o$684bo$683bo17$665bo$665b2o$664bobo
19$646b3o$648bo$647bo17$629bo$629b2o$628bobo19$610b3o$612bo$611bo53$
571b3o$573bo$572bo93$426b3o$428bo$427bo15$417b3o$419bo$418bo16$411b3o$
413bo$412bo37$390bo$390b2o$389bobo18$376bo$376b2o$375bobo76$334b3o$
336bo$335bo37$305bo$305b2o$304bobo35$208bo$208b2o$207bobo14$198bo$198b
2o$197bobo26$200b3o$202bo$201bo17$195bo$195b2o$194bobo44$125b3o$127bo$
126bo17$108b3o$110bo$109bo12$96b3o$98bo$97bo73$15b3o$17bo$16bo15$6b3o$
8bo$7bo16$3o$2bo$bo!
Also I found a nice NE->NW turner that uses just 2 scans:

Code: Select all

x = 277, y = 320, rule = B3/S23
273b2o$273b2o19$275b2o$275b2o3$253b2o$252bobo$254bo10$243b2o$244b2o$
243bo21$240b2o$239bobo$241bo18$222b2o$223b2o$222bo18$204b2o$203bobo$
205bo18$194b2o$195b2o$194bo18$176b2o$175bobo$177bo43$89b2o$90b2o$89bo
16$79b2o$80b2o$79bo14$75b2o$76b2o$75bo19$62b2o$61bobo$63bo24$46b2o$47b
2o$46bo28$20b2o$21b2o$20bo20$2o$b2o$o!
I hope the color is convenient for the MWSS tricks. I haven't checked.

chris_c
Posts: 966
Joined: June 28th, 2014, 7:15 am

Re: Half-bakery reaction with glider

Post by chris_c » July 9th, 2014, 10:30 pm

Westbound LWSS from a hive using one reset. It looks good but is it useful?

Code: Select all

x = 182, y = 207, rule = B3/S23
179b2o$178bo2bo$179b2o17$161bo$161b2o$160bobo22$141bo$141b2o$140bobo
22$125bo$125b2o$124bobo23$102b3o$104bo$103bo21$81bo$81b2o$80bobo22$61b
o$61b2o$60bobo39$12b3o$14bo$13bo22$3o$2bo$bo!

User avatar
codeholic
Moderator
Posts: 1147
Joined: September 13th, 2011, 8:23 am
Location: Hamburg, Germany

Re: Half-bakery reaction with glider

Post by codeholic » July 10th, 2014, 2:56 am

chris_c wrote:Westbound LWSS from a hive using one reset. It looks good but is it useful?
If you also find a cheap recipe for a northbound LWSS.

Code: Select all

x = 22, y = 16, rule = B3/S23
18b2o$17b2ob2o$18b4o$19b2o8$b3o$o2bo$3bo$3bo$obo!
Ivan Fomichev

User avatar
dvgrn
Moderator
Posts: 10610
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Half-bakery reaction with glider

Post by dvgrn » July 10th, 2014, 7:53 am

codeholic wrote:
chris_c wrote:Westbound LWSS from a hive using one reset. It looks good but is it useful?
If you also find a cheap recipe for a northbound LWSS...
Or if a (-3,-6) beehive can be cheaply regenerated somehow, and if the width of the recipe doesn't matter as much as the number of scans. Could probably line up a row of 90-turners to build something like this --

Code: Select all

x = 62, y = 57, rule = B3/S23
14b2o$13b2ob2o$14b4o$15b2o8$bo$2o$obo13$26b3o$26bo$27bo21$38b3o$38bo$
39bo3$59b3o$59bo$60bo!
-- but probably with just two or three slow gliders, not four. Of course a similar trick might be made to work with the seven-glider slow salvo, lining up a row of 90-turners to be triggered one after another in a single scan -- but with more than a few gliders it seems likely that early targets will get in the way of later gliders.

There's always the possibility that a lucky LWSS+G collision near an ash object will regenerate the same ash object at (-3,-6), plus a glider, but that's probably too unlikely to be worth looking for.

Seems as if we might as well settle for the 5-scan MWSS for now, unless a cheaper or narrower "natural" seed shows up.

chris_c
Posts: 966
Joined: June 28th, 2014, 7:15 am

Re: Half-bakery reaction with glider

Post by chris_c » July 10th, 2014, 11:03 pm

A three scan westward LWSS based on the previously mentioned one:

Code: Select all

x = 316, y = 354, rule = B3/S23
300b2o$300b2o19$284bo17b2o$284b2o16b2o$283bobo12$272bo$272b2o$271bobo
7$313b2o$312bo2bo$313b2o15$266bo$266b2o$265bobo18$248b2o$247bobo$249bo
18$230bo$230b2o$229bobo18$220b2o$219bobo$221bo14$164bo$164b2o$163bobo
2$202bo$202b2o$201bobo12$154bo$154b2o$153bobo4$192bo$192b2o$191bobo14$
144bo$144b2o$143bobo15$135bo$135b2o$134bobo18$119bo$119b2o$118bobo18$
107bo$107b2o$106bobo16$91b2o$90bobo$92bo18$73b2o$74b2o$73bo18$57b2o$
58b2o$57bo38$9b2o$8bobo$10bo18$b2o$obo$2bo!
I wonder if something like this could work? If the three SE rakes can be teased into this configuration at some point then a well timed LWSS can be turned into a boat to be used at leisure in the next iteration. There is no freedom to change the color of the boat with this exact salvo but hopefully opposite color reflectors can be produced some other way if necessary.

Code: Select all

x = 155, y = 151, rule = B3/S23
43b2o11b2o$43b2o10bo2bo$56bobo$57bo50$3o$2bo$bo90bobo$92b2o$93bo11$71b
o$70bo$70b3o6$72b2o$71bo2bo$71bobo$69b2obo$68bo2bo$68bobo$69bo2$56b2o$
55bo2bo63bobo$55bobo64b2o$53b2obo66bo$52bo2bo$52bobo$53bo8$101bo$100bo
$100b3o6$102b2o$101bo2bo$101bobo$99b2obo$98bo2bo$98bobo$99bo2$86b2o65b
o$85bo2bo63bo$85bobo64b3o$83b2obo$82bo2bo$82bobo$83bo8$132bo$130b2o$
131b2o6$132b2o$131bo2bo$131bobo$129b2obo$128bo2bo$128bobo$129bo2$116b
2o$115bo2bo$115bobo$113b2obo$112bo2bo$112bobo$113bo!

chris_c
Posts: 966
Joined: June 28th, 2014, 7:15 am

Re: Half-bakery reaction with glider

Post by chris_c » July 11th, 2014, 6:05 am

Hmmm... well the idea does seem at least vaguely sound. Here is a more complete example pattern. The boat reappears at (-3,-6) shift and a second glider from the SE is used to delete the blinker. The input NE glider is of the opposite color to the output NE glider as would be the case in the final spaceship.

Code: Select all

x = 657, y = 449, rule = B3/S23
529bo$528bo$528b3o12$508bo$506b2o$507b2o44$652b4o$652bo3bo$136bo515bo$
135bobo515bo2bo$135b2o39$104b2o$103bo2bo$103bobo$101b2obo$100bo2bo$
100bobo$101bo6$94b2o$93bo2bo$93bobo$91b2obo$90bo2bo$90bobo$91bo16$212b
2o$211bobo$212bo4$545bobo$545b2o$546bo11$524bo$523bo$523b3o45$395bo$
394bo$394b3o2$4b2o$3bo2bo$3bobo$b2obo$o2bo$obo$bo4$374bo$372b2o$373b2o
2$309b2o$308bo2bo$308bobo$306b2obo$305bo2bo$305bobo$306bo2$293b2o$292b
o2bo$292bobo$290b2obo$289bo2bo$289bobo$290bo22$87b3o$89bo251b2o$88bo
251bo2bo$340bobo$338b2obo$337bo2bo$337bobo$338bo2$325b2o$324bo2bo$324b
obo$322b2obo$321bo2bo$321bobo$322bo82$178b2o$177bo2bo$177bobo$175b2obo
$174bo2bo$174bobo$175bo2$162b2o$161bo2bo$161bobo$159b2obo$158bo2bo$
158bobo$159bo22$211b2o$210bo2bo$210bobo$208b2obo$207bo2bo$207bobo$208b
o2$195b2o$194bo2bo$194bobo$192b2obo$191bo2bo$191bobo$192bo30$255b2o$
254bo2bo$254bobo$252b2obo$251bo2bo$251bobo$252bo2$239b2o$238bo2bo$238b
obo$236b2obo$235bo2bo$235bobo$236bo!

User avatar
dvgrn
Moderator
Posts: 10610
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Half-bakery reaction with glider

Post by dvgrn » July 11th, 2014, 3:49 pm

chris_c wrote:Hmmm... well the idea does seem at least vaguely sound. Here is a more complete example pattern. The boat reappears at (-3,-6) shift and a second glider from the SE is used to delete the blinker. The input NE glider is of the opposite color to the output NE glider as would be the case in the final spaceship.
I just tried dropping the three-scan LWSS recipe into this sample pattern. It seems to need gliders of the wrong color -- i.e., not the color of the gliders produced by the rakes:

Code: Select all

x = 512, y = 439, rule = LifeHistory
411.2C$411.2C19$413.2C$413.2C6$492.C.C$492.2C$493.C11$471.C$470.C$
470.3C7$413.2C11.2C$413.2C10.C2.C$426.C.C$136.C290.C$135.C.C$135.2C
39$104.2C$103.C2.C$103.C.C$101.2C.C$100.C2.C$100.C.C$101.C6$94.2C$93.
C2.C$93.C.C131.D.D132.D.D$91.2C.C130.D5.D128.D5.D$90.C2.C$90.C.C131.D
2.3D2.D126.D2.3C2.D$91.C137.D134.C$224.D3.D3.D126.D3.C3.D2$225.D5.D
128.D5.D$227.D.D132.D.D12$212.2C$211.C.C$212.C40$509.C$509.C.C$509.2C
12$487.C.C$487.2C$488.C12$4.2C$3.C2.C$3.C.C$.2C.C$C2.C$C.C$.C7$124.C$
124.2C183.2C$123.C.C182.C2.C$308.C.C$306.2C.C$305.C2.C$305.C.C$306.C
2$293.2C$292.C2.C$292.C.C$290.2C.C$289.C2.C$289.C.C$290.C5$358.C.C$
358.2C$359.C11$337.C$336.C$336.3C3$341.2C$340.C2.C$340.C.C$338.2C.C$
337.C2.C$337.C.C$338.C2$325.2C$324.C2.C$324.C.C$322.2C.C$321.C2.C$
321.C.C$322.C82$178.2C$177.C2.C$177.C.C$175.2C.C$174.C2.C$174.C.C$
175.C2$162.2C$161.C2.C$161.C.C$159.2C.C$158.C2.C$158.C.C$159.C22$211.
2C$210.C2.C$210.C.C$208.2C.C$207.C2.C$207.C.C$208.C2$195.2C$194.C2.C$
194.C.C$192.2C.C$191.C2.C$191.C.C$192.C30$255.2C$254.C2.C$254.C.C$
252.2C.C$251.C2.C$251.C.C$252.C2$239.2C$238.C2.C$238.C.C$236.2C.C$
235.C2.C$235.C.C$236.C!
Can't shift the recipe over by just 1hd, because the LWSS only returns to the same phase every 2hd. Am I missing a trick here?

A possible nice thing about these reactions based on *WSSes and lateral gliders is that they can easily reach through half-bakery trails from 180-turners. If the timing works out badly for producing a glider for one of the trails, working from outside to inside, then an *WSS+glider trick could work equally well in the southeast.

I'm working on a script to build a complete HBK body, with a config file to allow the choice to send/not send a glider at each point of each scan. For starters I'm going to try getting a 24-scan HBK connected to a working suppression unit. Might 24 scans actually be enough to complete a knightship, with massively parallel construction?

I'll try to build in the option to do double/triple/quadruple-width scans, with multiple eight-half-bakery resets per scan... but changing the scan width also changes the required initial width between the A and B trails, which changes the way the suppression has to be done -- it gets a little bit complicated.

EDIT: The A1:BA2C design is even more annoying than I thought for long HBKs with many resets. I guessed, somewhat at random, that the distance from A1 to B is going to have to be something like 1000fd, given the distance that the whole structure is going to drift along its length. The shooting range, however wide it is, will be very close to B in the northeast, and very close to A1 in the southwest (because the shooting range is really an exact diagonal, whereas the half-bakery trails are only almost-diagonal).

At 1000fd total A1-to-B width, a knightship could support only about 1000 / (39+8) = ~21 single resets before the available shooting range narrows down to nothing. For a double-width shooting range it would be 1000 / (39+2*8) = ~18, and so on. So we might need an even wider A1-to-B distance.

But if A1-to-B is wider, we have to make B-to-C a little wider as well, to let A2 drift between them. A1 and A2 have to move in step with each other, so that the half-bakery kickback always works. B and C stay in step with each other also, but at a slightly different angle -- they have the 39N half-bakeries for each scan, but they skip the 8N half-bakeries for each reset. So B and C drift slightly with respect to A1 and A2; you just need 8N extra lanes per reset, or about 170 free lanes between B and C for A2 to drift in, for a 1000fd A1-to-B design:

Code: Select all

x = 3538, y = 2723, rule = B3/S23
1464bo$1464bobo$1464b2o12$3531bo$3530bo$3530b3o12$3510bo$3508b2o$3509b
2o388$3536bo$3535bo$3535b3o12$3515bo$3513b2o$3514b2o830$201b2o$200bo2b
o$200bobo$198b2obo$197bo2bo$197bobo$198bo18$179b2o$178bo2bo$178bobo$
176b2obo$175bo2bo$175bobo$176bo32$143b2o$142bo2bo$142bobo$140b2obo$
139bo2bo$139bobo$140bo6$133b2o$132bo2bo$132bobo$130b2obo$129bo2bo$129b
obo$130bo7$122b2o$121bo2bo$121bobo$119b2obo$118bo2bo$118bobo$119bo6$
112b2o$111bo2bo$111bobo$109b2obo$108bo2bo$108bobo$109bo6$102b2o$101bo
2bo$101bobo$99b2obo$98bo2bo$98bobo$99bo6$92b2o$91bo2bo$91bobo$89b2obo$
88bo2bo$88bobo$89bo6$82b2o$81bo2bo$81bobo$79b2obo$78bo2bo$78bobo$79bo
6$72b2o$71bo2bo$71bobo$69b2obo$68bo2bo$68bobo$69bo42$26b2o$25bo2bo$25b
obo$23b2obo$22bo2bo$22bobo$23bo18$4b2o$3bo2bo$3bobo$b2obo$o2bo$obo$bo
606$1483bo$1482bo$1482b3o12$1462bo$1460b2o$1461b2o180$1255b2o$1254bo2b
o$1254bobo$1252b2obo$1251bo2bo$1251bobo$1252bo2$1239b2o$1238bo2bo$
1238bobo$1236b2obo$1235bo2bo$1235bobo$1236bo4$1270b2o$1269bo2bo$1269bo
bo$1267b2obo$1266bo2bo$1266bobo$1233b2o32bo$1232bo2bo$1232bobo19b2o$
1230b2obo19bo2bo$1229bo2bo20bobo$1229bobo19b2obo$1230bo19bo2bo$1250bob
o$1217b2o32bo$1216bo2bo$1216bobo$1214b2obo$1213bo2bo$1213bobo$1214bo4$
1248b2o$1247bo2bo$1247bobo$1245b2obo$1244bo2bo$1244bobo$1245bo2$1232b
2o$1231bo2bo$1231bobo$1229b2obo$1228bo2bo$1228bobo$1229bo9$1227b2o$
1226bo2bo$1226bobo$1224b2obo$1223bo2bo$1223bobo$1224bo6$1217b2o$1216bo
2bo$1216bobo$1214b2obo$1213bo2bo$1213bobo$1214bo7$1206b2o$1205bo2bo$
1205bobo$1203b2obo$1202bo2bo$1202bobo$1203bo6$1196b2o$1195bo2bo$1195bo
bo$1193b2obo$1192bo2bo$1192bobo$1193bo5$1171b2o$1170bo2bo12b2o$1170bob
o12bo2bo$1168b2obo13bobo$1167bo2bo12b2obo$1167bobo12bo2bo$1168bo13bobo
$1183bo5$1161b2o$1160bo2bo12b2o$1160bobo12bo2bo$1158b2obo13bobo$1157bo
2bo12b2obo$1157bobo12bo2bo$1158bo13bobo$1173bo6$1150b2o14b2o$1149bo2bo
12bo2bo$1149bobo13bobo$1147b2obo12b2obo$1146bo2bo12bo2bo$1146bobo13bob
o$1147bo15bo6$1140b2o14b2o$1139bo2bo12bo2bo$1139bobo13bobo$1137b2obo
12b2obo$1136bo2bo12bo2bo$1136bobo13bobo$1137bo15bo6$1130b2o$1129bo2bo$
1129bobo$1127b2obo$1126bo2bo$1126bobo$1127bo6$1120b2o$1119bo2bo$1119bo
bo$1117b2obo$1116bo2bo$1116bobo$1117bo6$1110b2o$1072b2o35bo2bo$1071bo
2bo34bobo$1071bobo33b2obo$1069b2obo33bo2bo$1068bo2bo34bobo$1068bobo36b
o$1069bo2$1056b2o$1055bo2bo$1055bobo$1053b2obo43b2o$1052bo2bo43bo2bo$
1052bobo44bobo$1053bo43b2obo$1096bo2bo$1096bobo$1097bo2$1458b2o$1457bo
2bo$1457bobo$1455b2obo$1454bo2bo$1050b2o402bobo$1049bo2bo402bo$1049bob
o43b2o$1047b2obo43bo2bo344b2o$1046bo2bo44bobo344bo2bo$1046bobo43b2obo
345bobo$1047bo43bo2bo344b2obo$1091bobo344bo2bo$1034b2o56bo345bobo$
1033bo2bo402bo$1033bobo43b2o$1031b2obo43bo2bo$1030bo2bo44bobo$1030bobo
43b2obo$1031bo43bo2bo$1075bobo$1076bo3$1436b2o$1435bo2bo$1435bobo$
1433b2obo$1432bo2bo$1432bobo$1433bo$1073b2o$1072bo2bo344b2o$1072bobo
344bo2bo$1070b2obo345bobo$1069bo2bo344b2obo$1069bobo344bo2bo$1070bo
345bobo$1417bo$1057b2o$1056bo2bo$1056bobo$1054b2obo$1053bo2bo$1053bobo
$1054bo142$1275b2o$1274bo2bo$1274bobo$1272b2obo$1271bo2bo$1271bobo$
1272bo2$1259b2o$1258bo2bo$1258bobo$1256b2obo$1255bo2bo$1255bobo$1256bo
10$1253b2o$1252bo2bo$1252bobo$1250b2obo$1249bo2bo$1249bobo$1250bo2$
1237b2o$1236bo2bo$1236bobo$1234b2obo$1233bo2bo$1233bobo$1234bo!
Most of these adjustments break our current suppression mechanisms, or at least require a lot of custom fixes, so I probably won't worry about suppression until we're sure we have the body design right.

I'm tempted to go back and look at the nice simple seven-glider A:BC setup again, with none of these B/A2/C complications as the A-to-B distance gets wider. Couldn't we build all three A gliders on the NW side, using MWSS+G or LWSS+G tricks, with just half a dozen resets or so?

chris_c
Posts: 966
Joined: June 28th, 2014, 7:15 am

Re: Half-bakery reaction with glider

Post by chris_c » July 12th, 2014, 2:29 am

dvgrn wrote: Can't shift the recipe over by just 1hd, because the LWSS only returns to the same phase every 2hd. Am I missing a trick here?
Well isn't the circled red glider on the left out of position by 3 squares? The two NE gliders in white look like they are of the same color to me (in the technical sense as well as in the visual sense :)). Having said that, this aspect of spaceship timing is something that I overlooked... unlike gliders, two spaceships on the same line are not necessarily translations of each other in time.

It really seems that my luck is in today:

1. Boat produced by the LWSS + G collision in the NW is of the correct color.
2. LWSS produced by the monochromatic salvo is of the correct color, a phenomenon that I did not even consider.
3. The required 90 degree turner NE->NW matches the color of the two scan design that I posted recently.
dvgrn wrote: I'm working on a script to build a complete HBK body, with a config file to allow the choice to send/not send a glider at each point of each scan. For starters I'm going to try getting a 24-scan HBK connected to a working suppression unit. Might 24 scans actually be enough to complete a knightship, with massively parallel construction?
I think it's very difficult to say anything about the back end of the ship until we have a concrete sequence of lane numbers that we want to fire gliders on. If I were to make a HBK compiler (and I am tempted) then the width of the range is not something I would have the user worry about. As input I would just have the sequence of lane numbers and it would be up to the compiler to figure out how to space everything most tightly. I agree that making a suppression mechanism when you are unsure about how the spacing will turn out is quite daunting.

Also if the input was "0 2 6 12 20 30 42" I would expect to get back chains that looked like this:

Code: Select all

x = 493, y = 496, rule = B3/S23
442b2o$441bobo46bobo$442bo47b2o$491bo11$469bo$468bo$468b3o3$473b2o$
472bo2bo$472bobo$420b2o48b2obo$419bobo47bo2bo$420bo48bobo$470bo2$457b
2o$456bo2bo$456bobo$454b2obo$453bo2bo$453bobo$454bo9$452b2o$451bo2bo$
451bobo$449b2obo$448bo2bo$448bobo$449bo2$436b2o$435bo2bo$435bobo$433b
2obo$432bo2bo$432bobo$433bo9$431b2o$430bo2bo$375b2o53bobo$374bobo51b2o
bo$375bo51bo2bo$427bobo$428bo3$414b2o$413bo2bo$413bobo$411b2obo$410bo
2bo$410bobo$411bo9$409b2o$408bo2bo$408bobo$406b2obo$405bo2bo$405bobo$
406bo2$393b2o$392bo2bo$392bobo$390b2obo$389bo2bo$389bobo$390bo9$388b2o
$387bo2bo$387bobo$385b2obo$384bo2bo$384bobo$385bo6$378b2o$377bo2bo$
377bobo$375b2obo$374bo2bo$374bobo$362b2o11bo$361bo2bo$310b2o49bobo$
309bobo47b2obo$310bo47bo2bo$358bobo$359bo6$352b2o$351bo2bo$351bobo$
349b2obo$348bo2bo$348bobo$349bo9$347b2o$346bo2bo$346bobo$344b2obo$343b
o2bo$343bobo$344bo2$331b2o$330bo2bo$330bobo$328b2obo$327bo2bo$327bobo$
328bo9$326b2o$325bo2bo$325bobo$323b2obo$322bo2bo$322bobo$323bo6$316b2o
$315bo2bo$315bobo$313b2obo$312bo2bo$312bobo$313bo6$306b2o$305bo2bo$
305bobo$303b2obo$290b2o10bo2bo$289bo2bo9bobo$289bobo11bo$287b2obo$286b
o2bo$286bobo$287bo4$224b2o$223bobo$224bo55b2o$279bo2bo$279bobo$277b2ob
o$276bo2bo$276bobo$277bo6$270b2o$269bo2bo$269bobo$267b2obo$266bo2bo$
266bobo$267bo9$265b2o$264bo2bo$264bobo$262b2obo$261bo2bo$261bobo$262bo
2$249b2o$248bo2bo$248bobo$246b2obo$245bo2bo$245bobo$246bo9$244b2o$243b
o2bo$243bobo$241b2obo$240bo2bo$240bobo$241bo6$234b2o$233bo2bo$233bobo$
231b2obo$230bo2bo$230bobo$231bo6$224b2o$223bo2bo$223bobo$221b2obo$220b
o2bo$220bobo$221bo6$214b2o$199b2o12bo2bo$198bo2bo11bobo$198bobo10b2obo
$196b2obo10bo2bo$195bo2bo11bobo$195bobo13bo$196bo6$189b2o$188bo2bo$
188bobo$186b2obo$185bo2bo$118b2o65bobo$117bobo66bo$118bo5$179b2o$178bo
2bo$178bobo$176b2obo$175bo2bo$175bobo$176bo6$169b2o$168bo2bo$168bobo$
166b2obo$165bo2bo$165bobo$166bo9$164b2o$163bo2bo$163bobo$161b2obo$160b
o2bo$160bobo$161bo2$148b2o$147bo2bo$147bobo$145b2obo$144bo2bo$144bobo$
145bo9$143b2o$142bo2bo$142bobo$140b2obo$139bo2bo$139bobo$140bo6$133b2o
$132bo2bo$132bobo$130b2obo$129bo2bo$129bobo$130bo6$123b2o$122bo2bo$
122bobo$120b2obo$119bo2bo$119bobo$120bo6$113b2o$98b2o12bo2bo$97bo2bo
11bobo$97bobo10b2obo$95b2obo10bo2bo$94bo2bo11bobo$94bobo13bo$95bo5$
103b2o$88b2o12bo2bo$87bo2bo11bobo$87bobo10b2obo$85b2obo10bo2bo$84bo2bo
11bobo$84bobo13bo$85bo5$b2o$obo75b2o$bo75bo2bo$77bobo$75b2obo$74bo2bo$
74bobo$75bo6$68b2o$67bo2bo$67bobo$65b2obo$64bo2bo$64bobo$65bo6$58b2o$
57bo2bo$57bobo$55b2obo$54bo2bo$54bobo$55bo9$53b2o$52bo2bo$52bobo$50b2o
bo$49bo2bo$49bobo$50bo2$37b2o$36bo2bo$36bobo$34b2obo$33bo2bo$33bobo$
34bo!
From the above pattern I believe that the minimum distance required to insert 0, 1, 2, 3, 4 and 5 pairs of "blank" HB's is 22, 45, 65, 86, 106 and 117fd respectively. From 4 onwards the distance always grows by 11fd per HB pair.

If the gap between the produced gliders ends up being too small it might be good to allow control of that by sending an extra token in the input. Input of "0 2 6 d100 12" might indicate that we want gliders on lanes 0, 2, 6 and 12 but we want a gap of at least 100fd between the HB's responsible for producing glider 6 and those producing glider 12. The minimum achievable gap would be by adding two blank HB's which requires 65fd according to my reckoning.

I would just keep the input format as simple as possible at this stage. That would make life much easier if later on someone were to write some higher level code that could shift and interleave different scan patterns etc.
dvgrn wrote: I'm tempted to go back and look at the nice simple seven-glider A:BC setup again, with none of these B/A2/C complications as the A-to-B distance gets wider. Couldn't we build all three A gliders on the NW side, using MWSS+G or LWSS+G tricks, with just half a dozen resets or so?
The westward LWSS trick that I gave should be able to work with all of the "A" gliders in a plain ABC construction:

Code: Select all

x = 673, y = 365, rule = B3/S23
o$b2o$2o72$588bobo$588b2o$589bo11$567bo$566bo$566b3o18$670b2o$669b4o$
668b2ob2o$669b2o47$546bobo$546b2o$547bo9$495b2o$494bo2bo$494bobo28bo$
492b2obo28bo$491bo2bo29b3o$491bobo$492bo2$479b2o$478bo2bo$478bobo$476b
2obo$475bo2bo$475bobo$476bo9$513b2o$512bo2bo$512bobo$510b2obo$509bo2bo
$509bobo$510bo2$497b2o$496bo2bo$496bobo$494b2obo$493bo2bo$493bobo$494b
o16$485b2o$484bo2bo$484bobo$482b2obo$481bo2bo$481bobo$482bo8$473b2o$
472bo2bo$472bobo$470b2obo$469bo2bo$469bobo$470bo8$461b2o$460bo2bo$460b
obo$458b2obo$457bo2bo$457bobo$458bo8$449b2o$448bo2bo$448bobo$446b2obo$
445bo2bo$445bobo$429b2o15bo$428bo2bo$428bobo$426b2obo$425bo2bo$425bobo
$426bo8$417b2o$416bo2bo$416bobo$414b2obo$413bo2bo$413bobo$414bo8$405b
2o$404bo2bo$404bobo$402b2obo$401bo2bo$401bobo$402bo8$393b2o$392bo2bo$
392bobo$390b2obo$389bo2bo$389bobo$390bo7$341b2o$340bo2bo$340bobo$338b
2obo$337bo2bo$337bobo$338bo2$325b2o$324bo2bo$324bobo$322b2obo$321bo2bo
$321bobo$322bo12$364b2o$363bo2bo$363bobo$361b2obo$360bo2bo$360bobo$
361bo2$348b2o$347bo2bo$347bobo$345b2obo$344bo2bo$344bobo$345bo!
To delete the blinker it is necessary that the second glider is on a higher lane than the first. Therefore it is necessary that the first glider comes from the rear rake (C). Therefore it is lucky that C has the same color as A2 otherwise the color of the boat would be changed. Furthermore, it is lucky that both halves of A2 have the same color as A1 since we know that the color of boat that worked for A1 will also work for A2 when that is moved back up to the NW.

chris_c
Posts: 966
Joined: June 28th, 2014, 7:15 am

Re: Half-bakery reaction with glider

Post by chris_c » July 12th, 2014, 7:19 am

chris_c wrote: Well isn't the circled red glider on the left out of position by 3 squares?
Oh... I now see that the left hand red glider indicates the position of the output glider of the next iteration (ie. after 2,000 generations or so). But that should be of the opposite color to the input glider to the LWSS seed, because that LWSS was created on the previous iteration, before the entire pattern was shifted by (-3,-6).

I changed the timing of the B and C salvos to show that the output glider is of the correct color to begin the reconstruction of the LWSS seed:

Code: Select all

x = 621, y = 523, rule = B3/S23
601bobo$601b2o$602bo11$580bo$579bo$579b3o36$442b2o$442b2o19$444b2o$
444b2o28$444b2o11b2o$444b2o10bo2bo$457bobo$167bo290bo$166bobo$166b2o
37$618bo$618bobo$135b2o481b2o$134bo2bo$134bobo$132b2obo$131bo2bo$131bo
bo$132bo6$125b2o469bobo$124bo2bo468b2o$124bobo470bo$122b2obo$121bo2bo$
121bobo269b3o$122bo272bo$394bo15$243b2o$242bobo$243bo68$35b2o$34bo2bo$
34bobo$32b2obo$31bo2bo$31bobo$32bo7$155bo$155b2o183b2o$154bobo182bo2bo
$339bobo$337b2obo$336bo2bo$336bobo$337bo2$324b2o$323bo2bo$323bobo$321b
2obo$320bo2bo$320bobo$321bo5$4b2o383bobo$3bo2bo382b2o$3bobo384bo$b2obo
$o2bo$obo$bo7$368bo$367bo$367b3o3$372b2o$371bo2bo$371bobo$369b2obo$
368bo2bo$368bobo$369bo2$356b2o$355bo2bo$355bobo$353b2obo$352bo2bo$352b
obo$353bo82$209b2o$208bo2bo$208bobo$206b2obo$205bo2bo$205bobo$206bo2$
193b2o$192bo2bo$192bobo$190b2obo$189bo2bo$189bobo$190bo19$178b2o$177bo
2bo$177bobo$175b2obo63b2o$174bo2bo63bo2bo$174bobo64bobo$175bo63b2obo$
238bo2bo$162b2o74bobo$161bo2bo74bo$161bobo$159b2obo63b2o$158bo2bo63bo
2bo$158bobo64bobo$159bo63b2obo$222bo2bo$222bobo$223bo19$211b2o$210bo2b
o$210bobo$208b2obo$207bo2bo$207bobo$208bo2$195b2o$194bo2bo$194bobo$
192b2obo90b2o$191bo2bo90bo2bo$191bobo91bobo$192bo90b2obo$282bo2bo$282b
obo$283bo2$270b2o$269bo2bo$269bobo$267b2obo$266bo2bo$266bobo$267bo19$
255b2o$254bo2bo$254bobo$252b2obo$251bo2bo$251bobo$252bo2$239b2o$238bo
2bo$238bobo$236b2obo$235bo2bo$235bobo$236bo!

User avatar
dvgrn
Moderator
Posts: 10610
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Half-bakery reaction with glider

Post by dvgrn » July 12th, 2014, 8:51 am

chris_c wrote:Well isn't the circled red glider on the left out of position by 3 squares?
...
Oh... I now see that the left hand red glider indicates the position of the output glider of the next iteration (ie. after 2,000 generations or so). But that should be of the opposite color to the input glider to the LWSS seed, because that LWSS was created on the previous iteration, before the entire pattern was shifted by (-3,-6).
Very nice! I withdraw my confused objection.
chris_c wrote:It really seems that my luck is in today:

1. Boat produced by the LWSS + G collision in the NW is of the correct color.
2. LWSS produced by the monochromatic salvo is of the correct color, a phenomenon that I did not even consider.
3. The required 90 degree turner NE->NW matches the color of the two scan design that I posted recently.
4. All three A gliders are the same color. That's a big piece of luck, too, which really pushes me to consider switching back to the original A:BC design, just because it's so much simpler to work with. Four nested 180-turners instead of six will be easier to deal with, I think, especially since the A1:BA2C design had some gliders so far advanced or delayed with respect to others.

I've started coding a quick HBK builder, based now on the A:BC design (can always change it later, but it will be good to see what a multi-reset system actually looks like). It may take a few more days to finish since I'm traveling today.

Hadn't thought about the adjustable delay between successive rake gliders at all -- you get +8N delay just by moving a rake unit N full diagonals southwest, which is pretty much equivalent to moving the target by N fd northeast, but sometimes one will be more convenient than the other. Guess I'll have to adjust my input format to allow for that. More eventually --

User avatar
calcyman
Moderator
Posts: 2932
Joined: June 1st, 2009, 4:32 pm

Re: Half-bakery reaction with glider

Post by calcyman » July 12th, 2014, 7:45 pm

Attached is a script to build a (-3, -6)c/P knightship for any sufficiently large P. xxx
Attachments
semibake.zip
HBK assembler
(2.82 KiB) Downloaded 647 times
What do you do with ill crystallographers? Take them to the mono-clinic!

Post Reply