17 in 17: Efficient 17-bit synthesis project

For discussion of specific patterns or specific families of patterns, both newly-discovered and well-known.
User avatar
Freywa
Posts: 589
Joined: June 23rd, 2011, 3:20 am
Location: Singapore
Contact:

Re: 17 in 17: Efficient 17-bit synthesis project

Post by Freywa » April 26th, 2019, 9:41 pm

AforAmpere wrote:xs17_mkie0dbz1 in 14:
Easily reducible to 13:

Code: Select all

x = 202, y = 44, rule = B3/S23
145bo$145bobo$145b2o$133bo$131bobo$132b2o4$135bo$134bo$134b3o2$125bo$
126b2o$125b2o$49bo$50bo131bo$48b3o129bobo$121b2o58b2o11bobo$121b2o61bo
10b2o$53bo129bobo9bo$48b2o3bobo127bobo13b2o$47bobo3b2o129bo8b2o3bo2bo$
49bo74bo68b2o4b2o$124bo61b2ob2o$124bo62bobo$2bo183bo2bo$obo3bo179b3o$b
2o2bo44b2o70b2o$5b3o42bobo69bobo60bob2o$51bo71bo61b2obo$55b2o70b2o$7b
2o46b2o70b2o$6bobo$8bo6$94b2o$93bobo$95bo!
Princess of Science, Parcly Taxel

BobShemyakin
Posts: 214
Joined: June 15th, 2014, 6:24 am

Re: 17 in 17: Efficient 17-bit synthesis project

Post by BobShemyakin » April 27th, 2019, 5:41 am

My base of syntheses of 6939 syntheses meets conditions 17 in 17:
17in17.rar
syntheses 17 in 17
(722.54 KiB) Downloaded 169 times
138 meet conditions 17 = 17:
17to17.rar
syntheses 17=17
(29.08 KiB) Downloaded 165 times
and 696 still lifes should be synthesized:
17.89 17.154 17.157 17.158 17.173 17.174 17.175 17.177 17.181 17.183
17.184 17.185 17.187 17.188 17.189 17.196 17.198 17.199 17.202 17.210
17.220 17.227 17.228 17.232 17.233 17.235 17.248 17.271 17.284 17.301
17.321 17.355 17.365 17.367 17.376 17.385 17.391 17.392 17.406 17.409
17.410 17.412 17.413 17.417 17.418 17.419 17.455 17.484 17.485 17.486
17.565 17.595 17.601 17.604 17.744 17.745 17.790 17.976 17.978 17.980
17.981 17.982 17.986 17.987 17.988 17.989 17.990 17.992 17.993 17.994
17.995 17.996 17.998 17.999 17.1002 17.1003 17.1004 17.1010 17.1015 17.1021
17.1026 17.1031 17.1037 17.1041 17.1042 17.1043 17.1044 17.1047 17.1049 17.1051
17.1053 17.1054 17.1055 17.1059 17.1067 17.1075 17.1076 17.1077 17.1081 17.1082
17.1084 17.1094 17.1095 17.1101 17.1103 17.1104 17.1105 17.1106 17.1107 17.1109
17.1110 17.1111 17.1115 17.1116 17.1119 17.1124 17.1125 17.1128 17.1129 17.1132
17.1133 17.1137 17.1138 17.1139 17.1140 17.1141 17.1142 17.1144 17.1152 17.1153
17.1158 17.1161 17.1162 17.1163 17.1164 17.1166 17.1171 17.1190 17.1191 17.1192
17.1193 17.1194 17.1199 17.1200 17.1201 17.1202 17.1205 17.1208 17.1224 17.1234
17.1239 17.1254 17.1265 17.1266 17.1279 17.1286 17.1287 17.1291 17.1295 17.1297
17.1306 17.1319 17.1324 17.1326 17.1327 17.1328 17.1329 17.1335 17.1336 17.1337
17.1339 17.1346 17.1347 17.1349 17.1400 17.1401 17.1402 17.1403 17.1426 17.1430
17.1441 17.1444 17.1447 17.1452 17.1458 17.1460 17.1462 17.1464 17.1465 17.1479
17.1481 17.1482 17.1484 17.1498 17.1507 17.1518 17.1531 17.1546 17.1549 17.1560
17.1561 17.1565 17.1566 17.1567 17.1584 17.1586 17.1587 17.1588 17.1589 17.1590
17.1591 17.1611 17.1615 17.1616 17.1617 17.1619 17.1620 17.1621 17.1632 17.1635
17.1636 17.1639 17.1641 17.1644 17.1646 17.1647 17.1675 17.1679 17.1681 17.1691
17.1694 17.1712 17.1717 17.1719 17.1722 17.1729 17.1730 17.1744 17.1748 17.1751
17.1752 17.1758 17.1763 17.1764 17.1775 17.1781 17.1785 17.1787 17.1824 17.1825
17.1850 17.1851 17.1859 17.1901 17.1902 17.1903 17.1904 17.1926 17.1927 17.1929
17.1930 17.1936 17.1944 17.1947 17.1957 17.1972 17.1973 17.1974 17.1976 17.2055
17.2056 17.2084 17.2086 17.2087 17.2088 17.2090 17.2091 17.2092 17.2095 17.2096
17.2097 17.2102 17.2103 17.2104 17.2115 17.2151 17.2153 17.2209 17.2225 17.2234
17.2253 17.2255 17.2262 17.2282 17.2288 17.2295 17.2297 17.2304 17.2305 17.2309
17.2310 17.2317 17.2337 17.2355 17.2368 17.2369 17.2398 17.2402 17.2413 17.2414
17.2447 17.2476 17.2477 17.2538 17.2540 17.2544 17.2548 17.2549 17.2578 17.2615
17.2632 17.2642 17.2643 17.2644 17.2761 17.2762 17.3166 17.3169 17.3170 17.3171
17.3173 17.3174 17.3176 17.3177 17.3180 17.3182 17.3183 17.3185 17.3187 17.3188
17.3189 17.3190 17.3191 17.3193 17.3194 17.3198 17.3199 17.3200 17.3201 17.3202
17.3203 17.3209 17.3214 17.3217 17.3220 17.3221 17.3224 17.3225 17.3226 17.3255
17.3256 17.3257 17.3262 17.3263 17.3265 17.3268 17.3270 17.3280 17.3281 17.3282
17.3283 17.3284 17.3287 17.3288 17.3301 17.3303 17.3306 17.3307 17.3308 17.3313
17.3317 17.3324 17.3325 17.3326 17.3335 17.3336 17.3338 17.3339 17.3347 17.3352
17.3354 17.3362 17.3365 17.3366 17.3369 17.3371 17.3387 17.3388 17.3392 17.3393
17.3394 17.3396 17.3411 17.3416 17.3419 17.3429 17.3430 17.3436 17.3437 17.3438
17.3439 17.3452 17.3457 17.3458 17.3479 17.3480 17.3481 17.3482 17.3483 17.3484
17.3487 17.3490 17.3494 17.3496 17.3501 17.3504 17.3505 17.3508 17.3509 17.3510
17.3511 17.3513 17.3517 17.3520 17.3523 17.3525 17.3544 17.3545 17.3549 17.3557
17.3565 17.3566 17.3567 17.3579 17.3591 17.3599 17.3601 17.3602 17.3603 17.3606
17.3609 17.3610 17.3611 17.3612 17.3623 17.3641 17.3643 17.3644 17.3648 17.3659
17.3661 17.3663 17.3666 17.3671 17.3672 17.3674 17.3696 17.3697 17.3703 17.3704
17.3710 17.3713 17.3714 17.3715 17.3732 17.3749 17.3750 17.3781 17.3783 17.3785
17.3795 17.3799 17.3800 17.3816 17.3818 17.3820 17.3821 17.3822 17.3830 17.3833
17.3835 17.3836 17.3837 17.3838 17.3841 17.3844 17.3848 17.3853 17.3854 17.3855
17.3856 17.3884 17.3888 17.3891 17.3896 17.3897 17.3906 17.3936 17.3937 17.3938
17.3939 17.3951 17.3952 17.3979 17.3981 17.3983 17.3984 17.3990 17.4000 17.4029
17.4050 17.4071 17.4084 17.4089 17.4096 17.4097 17.4098 17.4109 17.4120 17.4123
17.4130 17.4131 17.4133 17.4134 17.4135 17.4136 17.4146 17.4156 17.4165 17.4170
17.4172 17.4173 17.4188 17.4196 17.4214 17.4217 17.4219 17.4224 17.4228 17.4232
17.4233 17.4243 17.4247 17.4248 17.4249 17.4252 17.4253 17.4255 17.4257 17.4258
17.4259 17.4260 17.4263 17.4265 17.4266 17.4267 17.4270 17.4271 17.4272 17.4279
17.4287 17.4288 17.4294 17.4310 17.4323 17.4324 17.4325 17.4326 17.4372 17.4383
17.4385 17.4399 17.4405 17.4435 17.4444 17.4476 17.4490 17.4491 17.4496 17.4501
17.4503 17.4510 17.4512 17.4513 17.4518 17.4529 17.4531 17.4532 17.4534 17.4535
17.4537 17.4538 17.4543 17.4569 17.4570 17.4579 17.4595 17.4598 17.4604 17.4623
17.4632 17.4641 17.4655 17.4657 17.4683 17.4714 17.4720 17.4762 17.4765 17.4779
17.4844 17.4850 17.4899 17.4939 17.4992 17.4993 17.5000 17.5027 17.5031 17.5055
17.5059 17.5060 17.5064 17.5091 17.5112 17.5162 17.5216 17.5223 17.5229 17.5233
17.5236 17.5237 17.5261 17.5284 17.5360 17.5447 17.5560 17.5629 17.5634 17.5640
17.5729 17.5812 17.5851 17.5875 17.5946 17.5960 17.5962 17.5994 17.6049 17.6050
17.6053 17.6062 17.6063 17.6064 17.6068 17.6080 17.6113 17.6115 17.6116 17.6117
17.6163 17.6164 17.6169 17.6170 17.6172 17.6174 17.6175 17.6190 17.6215 17.6220
17.6221 17.6226 17.6247 17.6248 17.6257 17.6286 17.6313 17.6455 17.6482 17.6566
17.6612 17.6764 17.6780 17.6937 17.7110 17.7601


Bob Shemyakin

User avatar
calcyman
Posts: 2096
Joined: June 1st, 2009, 4:32 pm

Re: 17 in 17: Efficient 17-bit synthesis project

Post by calcyman » April 27th, 2019, 6:12 am

Very impressive! I'm running those 6939 syntheses through (a modification of) mniemiec.py to assimilate them into Shinjuku. This necessarily will result in an improvement, because Catagolue only knows how to synthesise 6923 of the 17-bit still-lifes in <= 16 gliders.

Thank you so much!
What do you do with ill crystallographers? Take them to the mono-clinic!

User avatar
Freywa
Posts: 589
Joined: June 23rd, 2011, 3:20 am
Location: Singapore
Contact:

Re: 17 in 17: Efficient 17-bit synthesis project

Post by Freywa » April 27th, 2019, 6:21 am

BobShemyakin wrote:My base of syntheses of 6939 syntheses meets conditions 17 in 17:
138 meet conditions 17 = 17:
I'll not bore over your files for a while, but for compatibility's sake, don't use RAR. Use 7-Zip (or .tar.gz, or just plain old zip) instead.

Golly can't open 17-490.rle – you are definitely using a non-standard program to write your RLEs. They look fragmented when I open them up in a text editor.

Lastly, regarding the manner in which you presented your syntheses, an RLE file by itself is quite hard to analyse for the steps. I created Shinjuku and its format to fill in that gap – and now it contains tens of thousands of directly machine-readable components. I would appreciate it if you just give component lines in the future.
Princess of Science, Parcly Taxel

User avatar
calcyman
Posts: 2096
Joined: June 1st, 2009, 4:32 pm

Re: 17 in 17: Efficient 17-bit synthesis project

Post by calcyman » April 27th, 2019, 7:59 am

Freywa wrote:(criticisms)
Bob: I found the files to be perfectly comprehensible, so please continue sharing your work in whichever format is most convenient for you. The automated conversion of archives of RLEs into Shinjuku lines is near-perfect (occasionally it has hiccups, but they're sufficiently rare that they can be done manually if necessary).

Interestingly, most of the syntheses agree with existing ones; out of 9981 synthesis components extracted from Bob's files, only 3265 are new.
What do you do with ill crystallographers? Take them to the mono-clinic!

User avatar
calcyman
Posts: 2096
Joined: June 1st, 2009, 4:32 pm

Re: 17 in 17: Efficient 17-bit synthesis project

Post by calcyman » April 27th, 2019, 9:33 am

EDIT: a more up-to-date version of the script is in the Shinjuku repository

I've hacked together a lifelib script called findpreds.py (based on Jeremy Tan's pullsoups.py) which should be useful for eyeballing predecessors of natural still-lifes. It does the following:
  • Downloads soups from Catagolue.
  • Determines the first generation in which an isolated copy of the still-life is present (provided it's after gen 50).
  • Rewinds that snapshot by 50 generations.
  • Systematically deletes as many kingwise-connected components as possible whilst still ensuring that the still-life is still produced.

Code: Select all

import sys, os, lifelib

lt = lifelib.load_rules('b3s23').lifetree(n_layers=1)

zero = lt.pattern()
sl = lt.pattern(sys.argv[1])

sym = 'C1'
if (len(sys.argv) >= 3):
    sym = sys.argv[2]

rew = 50
if (len(sys.argv) >= 4):
    rew = int(sys.argv[3])

def contains_sl(mess):

    without = mess.replace(sl, zero, halo='ooo$ooo$ooo!', orientations='rotate4reflect')
    return (mess != without)

mosaic = lt.pattern()
for (n, soup) in enumerate(sl.download_samples()[sym]):

    rew = 50

    s50 = soup[rew]

    lb = 0
    ub = 32768

    if contains_sl(s50[lb]):
        continue
    if not contains_sl(s50[ub]):
        continue

    # Binary search:
    while ((ub - lb) > 1):
        mb = (ub + lb) // 2
        if contains_sl(s50[mb]):
            ub = mb
        else:
            lb = mb

    print('#C soup %d gives birth at time %d' % (n, ub + rew))

    pred = soup[ub]
    comps = pred.components(halo='ooo$ooo$ooo!')

    for c in comps:
        if contains_sl((pred - c)[rew]):
            pred -= c

    mosaic += pred(n // 10 * 100, n % 10 * 100)

print(mosaic.rle_string())
For instance, if you run the following:

$ python findpreds.py xs17_0g8im853z343

then it yields this output:

Code: Select all

x = 80, y = 633, rule = B3/S23
39b2o$38bo3bo$38b2obobo3b3o$45b2o$46bo2bo$46b3o$40bo34b3o$40b2o31b3o$
39b2o7b3o17b3ob5o2bo$37b3o6b2o2bo4b3o9b6ob6o$29b3o9b3o3b3o16bob2o2bo$
29b3o3b2obo3b2o4bo17bo3b2o5b2o$28bo3bo2b2o3bob2o24bob3o5b2o$29b2o2bob
2obo26bo2b2o$29b4o7b2o20b2o3bo2bobo$31bo5b2o27bo3bobo$54bo6bo3bo5bo$
55bo5b3o6b2o$32b3o13bobo3bo11b4o$30b2ob2o16b2o11bo$30b2o16b2ob2o10bo$
31bo2bo13b2o14bob2o$32bobo30bo2b2o$33bo31bo4bo$67b4o69$57bo$38bo17bobo
$36b2o17b2ob2o$36b2o17bo2bob2o$33bo3bobo2bo12b2o2bo$32b3o2b5obo11bo2bo
4b3o$31b5o4b2obo10b2o3bo4b3o$30bo4b2o3bo2bo6bo2b3o2bo7b2o$29b2o5b2o3b
2o6bo5bo2b3o3bobo$28b2o5b2o12bobob3o2b8o$29bobo3b2o13bo3bo8bo$30b2o$
35bo5b2o4b3o$35bo4b3o4bobo$41bobo4b2o$25bo5bo10b2o$25bobo2bobo$25bo2bo
2bo3bo$27b2o5b2o$29b2ob2o$29bobobo$45b3o$30b3o74$12bo$12bo$12bo$12bo$
2bo10bo$b4o$2o2b4o6bo$bo3bo4b4o2bo$2bo6bo4b6o$3bo4b2o7bobob2o$23bo$17b
4obo$3b2o14b3o$3b2o5$18bo$17b3o$14b2o4bo$17bo2bo$14b3o2b2o$14bo$12bo$
12bo3$11b2o$11b2obo$12bobo$13bo61$28b3o8b3o$28bo2bo6b3o$21b2o5b2obo$
20bobo7bo4bo$22bo12bo$34bobo$35bo$35bo4$33bo$32b3o$31b2ob2o$30b2o$29b
3o$30b2o101$42b2o$33b3o6bobobo$33bo2b2o3b2obob2o$33b3obo4bobo$37b2o6b
2o$38b2o4b2o$34bobob2o$33b3ob2o$33bo3bo3b2o6b3o$34b4o2b5o2bo$39b2o3bo$
35bo3b2o3bob2o$36bo2bo2bob2o$36bo2bo5bo$37bob3o8b3o$36b2o2b2o8b3o$37bo
5b4o$45b2o197$37bo$37bo$37bo2$36bo$36bo$36bo3$28b2o$27bo2bo$28bobo$29b
o3$25b2o4b2o$25b2o2b4obo$29b3o2b3o$28bo5b2o$29b2o$29b3o!
Thence, it's easy to pick out a nice low-population predecessor that looks especially amenable to synthesis:

Code: Select all

x = 22, y = 17, rule = B3/S23
8b3o8b3o$8bo2bo6b3o$b2o5b2obo$obo7bo4bo$2bo12bo$14bobo$15bo$15bo4$13bo
$12b3o$11b2ob2o$10b2o$9b3o$10b2o!
Last edited by calcyman on June 8th, 2019, 9:04 am, edited 1 time in total.
What do you do with ill crystallographers? Take them to the mono-clinic!

AforAmpere
Posts: 1049
Joined: July 1st, 2016, 3:58 pm

Re: 17 in 17: Efficient 17-bit synthesis project

Post by AforAmpere » April 27th, 2019, 10:29 am

xs17_0c9jzca254c in 15:

Code: Select all

x = 199, y = 113, rule = B3/S23
6$192bo$191bo$191b3o25$98bo$99bo$97b3o40bo$139bo$139b3o3$152bo$152bobo
$152b2o5$86bo$87bo$85b3o35b2o$123b2o$97bobo$98b2o$98bo30bo$128bobo$
128bobo$109bo19bo$109bo13bo15b2o$109bo12bobo14b2o$121bo2bo5b3o$52bo69b
2o$51bo$51b3o50bo$103bobo22b2o$11bo34bo56bobo22bo3b2o$12bo34b2o55bo5b
2o9b2o2b2obo3b2o$10b3o33b2o62b2o5bo3bo2bo2bo$116bobo4b2o2bobo$116bobo
9b2o$117bo$37bobo$38b2o$38bo92b3o$48b2o3b3o$10b2o35b2o$9bobo37bo$11bo$
39b3o$41bo$40bo72b2o$112bobo$113bo13$154b3o$154bo$78b3o74bo$80bo$79bo!
I and wildmyron manage the 5S project, which collects all known spaceship speeds in Isotropic Non-totalistic rules.

Things to work on:
- Find a (7,1)c/8 ship in a Non-totalistic rule
- Finish a rule with ships with period >= f_e_0(n) (in progress)

BobShemyakin
Posts: 214
Joined: June 15th, 2014, 6:24 am

Re: 17 in 17: Efficient 17-bit synthesis project

Post by BobShemyakin » April 27th, 2019, 10:36 am

Freywa wrote:
BobShemyakin wrote:My base of syntheses of 6939 syntheses meets conditions 17 in 17:
138 meet conditions 17 = 17:
I'll not bore over your files for a while, but for compatibility's sake, don't use RAR. Use 7-Zip (or .tar.gz, or just plain old zip) instead.

Golly can't open 17-490.rle – you are definitely using a non-standard program to write your RLEs. They look fragmented when I open them up in a text editor.

Lastly, regarding the manner in which you presented your syntheses, an RLE file by itself is quite hard to analyse for the steps. I created Shinjuku and its format to fill in that gap – and now it contains tens of thousands of directly machine-readable components. I would appreciate it if you just give component lines in the future.
Sometimes it happens. New ptterns turn out by means of combination of the list of still lifes with the list of converters. Process can take a long time and an exit quite big. Sometimes basic data contain comments. To correct quite easily: by means of the text editor remove superfluous.
17.490 in 7G:

Code: Select all

x = 65, y = 96, rule = B3/S23
58b2o$58bo2b2obo$59bobob2o$o59bo$b2o4bo$2o3bobo$6b2o4$2b3o$4bo$3bo16$
11b2o$11bobo$11bo25$2bo$bobob2o$o2b2obo2b2o$2o7bobo$9bo$16bo$15bo$15b
3o$7b3o$9bo$8bo26$2bo$bobob2ob2o$o2b2obobo$2o6bo$8b2o!
Bob Shemyakin

User avatar
Freywa
Posts: 589
Joined: June 23rd, 2011, 3:20 am
Location: Singapore
Contact:

Re: 17 in 17: Efficient 17-bit synthesis project

Post by Freywa » April 27th, 2019, 1:32 pm

BobShemyakin wrote:Sometimes it happens. New ptterns turn out by means of combination of the list of still lifes with the list of converters. Process can take a long time and an exit quite big. Sometimes basic data contain comments. To correct quite easily: by means of the text editor remove superfluous.
Text editors only work for small numbers of files – and here's over 7000 of them.

I have a task for you. The attachment below contains 7773 RLEs, one for each 17-bit strict still life. Each RLE contains, as a comment in the first line, the cost in gliders of the still life within that RLE according to Shinjuku currently (unsynthesised still lifes have a cost marked as "infinity"). Your task is to process these files and reply back to me with only a smaller archive of RLEs, where each RLE there contains a strictly cheaper (better) synthesis of the still life within according to your archive. I will be expecting – I need to fill in gaps that were left by niemiec.py.
xs17.7z
17-bit still lifes, commented with their current Shinjuku cost
(114.75 KiB) Downloaded 167 times
Edit: And I've knocked out one already, xs17_0g84cj96z1226 in 13 (AforAmpere found the initial predecessor):

Code: Select all

x = 83, y = 33, rule = B3/S23
4bo$5bo$3b3o2$69bo$69bobo$69b2o2$66b3o$13bo18bobo$14b2o16b2o30bo5bo$
13b2o18bo30bo5bo$64bo5bo$81bo$66b3o11bo$80b3o$77b2o$69bo7b2o$69b3o$72b
o$9bo11bo9bo35bo3bo$7bobo12bo8bobo32bobobo$8b2o2b2o6b3o8b2o32bo2b2o$
11bo2bo50bobo$3o9b2o52b2o$2bo74b2o$bo74bo2bo$77b2o$80b3o$80bo$18b3o60b
o$20bo$19bo!
Princess of Science, Parcly Taxel

hkoenig
Posts: 98
Joined: June 20th, 2009, 11:40 am

Re: 17 in 17: Efficient 17-bit synthesis project

Post by hkoenig » April 27th, 2019, 2:02 pm

Bob-- Thanks for the archive.
freywa wrote: Your task is to process these files and reply back to me with only a smaller archive of RLEs, where each RLE there contains a strictly cheaper (better) synthesis of the still life within according to your archive. I will be expecting – I need to fill in gaps that were left by niemiec.py.
Perhaps you should welcome the help, no matter what the format. Writing conversions for formats isn't that hard, once you understand the data format. Yours is not some sort of special, canonical format that everyone is require to adhere to.

For example, it would have been nice for your list of 7773 files to also include the apgcode for the object, which would make object lookup a lot easier when loading into an SQLite database. On the other hand, a single file, with one record per object with something like this format would have sufficed--

Code: Select all

CREATE TABLE GliderCosts (
'apgcode' TEXT,
'gliderCost' TEXT
'rle' TEXT,
};

User avatar
Freywa
Posts: 589
Joined: June 23rd, 2011, 3:20 am
Location: Singapore
Contact:

Re: 17 in 17: Efficient 17-bit synthesis project

Post by Freywa » April 27th, 2019, 2:08 pm

As of this post, 86 still lifes remain unsynthed in Shinjuku:

Code: Select all

x = 207, y = 150, rule = B3/S23
b2ob2o20b2o12b2o18b2o3bo14b2obo19bo18bo2bo15bobobo16bo19bo21b2o$bob2o
18b2o2bo13bo18bo2b3o14bob4o15b3o2b2o14b4o14bob3obo14bobo17bobo16b2obob
o$6bo17bobo14bob2o16b2o3b2o18bo13bo5bo13b2o18bo5bo14bo2bo15bo2bo17bobo
$2b2ob2o15bo2bo16bobo18b2o2bo13b4obo14bo5bo13bob2obo14b3obo14b2o3bo14b
2o2b2o15bo2bo$2bo18bob2o20b2o16bob2o14bo2b2o16bo3b2o13bobob2o16bob2o
17b2o20bo13b2o3bo$obo19bo24bo55bobo16bo40bo20b2o16b3o$2o18bobo24b3o54b
2o55b3o20bo17bo$20b2o28bo109bo21bobo$49b2o109b2o20b2o12$b2o22b2o13b2o
4b2o12b2o3b2o15b2o18b2o16b2ob2o16b2o19b2o19bo16b2obo$bo2b2o20bo14bo4bo
bo11bo2bo2bo14bobo17bo2bob2o13bobo2bo13bob3o16bobo17b3o16bob2o$2bobo2b
o15b3o15bob2obobo12bob3o14bo19bob2o3bo12bo4b2o13bo4bo15bo2b2o14bo23b2o
$b2o3b2o15bo18bobo2bo14bo18b5o15bo2b3o14b2o18b2o2bo14b2o4bo13bob2o17b
3o2bo$3bo17b2o22b2o16bo21bo16bobo18bo19bob2o17b2o15bobo16bo4b2o$obo17b
o39b3o20b2o18bo16bobo20bo20bo19b2o14b2o$2o19b2o37bo22bo36b2o20b2o18bob
o20bo$22bo61bo77b2o21bobo$22bobo58b2o101b2o$23b2o11$3b2o21bo13b2ob2o
15b2o2b2o16b2o18b2o19bo17bo20b2o18bobo15b2o$2bobo19b3o14bobobo14bo2bo
2bo14bob3o15bo2bo15bobobo15bobo18bobo18b2obo14bo$bo21bo17bo4bo14bob3o
14bo5bo13bobo2bo14b2o2bo15bobo18bo4b2o12b2o3bo16bo$bob4o17bo17b4o16bo
18b5o15bob3o18b2o15b2ob2o14b2o5bo12bo2b2o16b2o$2bo3bo14b4o18bo19bo38bo
22bo17bobo16b2o2bo14b2o17bo$3bo16bo20bo18b3o20bo19bo16b2o2bo16b2o19bo
2bo16bo17b5o$3o18b2o18b2o17bo21bobo15b3o17bo2bo17bo21b2o15bo24bo$o21bo
60bo16bo20b2o20bo36b2o20bobo$21bo120b2o58b2o$21b2o11$4bo21bo13b2ob2o
15b2o21bo18b2o2b2o15b2o16bo21bob2o13b2o21bo$2b3o19b3o14bobobo14bobo18b
5o15bo2bo2bo12bo2bo16bobo20b2obo14bo19b3o$bo21bo17bo2bobo15bo17bo5bo
13bobobobo13b3o2bo14bobo18b2o17bo19bo$bob4o17bo17bobobo14b2ob2o15b2o2b
2o14bo2bob2o15b3o15b2ob2o14bo2b2obo13b2o19bo$2bo3bo14b4o16b2ob2o17bobo
17bo18b2o18bo20bobo15bobob2o15bo17b2o$3bo16bo40b2o17bobo40bo17b2o19bo
17b2o18bo4bo$3o18b2o38bo18b2o38b3o18bo38bo3b2o15bo2bobo$o21bo40bo56bo
21bo38bo2bo17bo2bo$20bo41b2o77b2o39bobo18b2o$20b2o161bo11$2b2o18b2o17b
2o2b2o13b2o21bo2bo16bobo15bo19b2ob2o17b2o16b2ob2o15b2o$3b3o15bo2bo15bo
2bo2bo13bobo18b6o14b4obo13bobo3b2o12bobobo18bobo15bob2obo13bo2bo$bo4bo
14b2o2bo15bob3o16bo17bo19bo5bo13bobo3bo13bo5bo14b2o2bo20bo13b2o2bo$b2o
bobo19bo15bo18b2ob2o15b2o2b2o14b2o3b2o13bob2obo14bo3b2o13bo4b2o15b4o
19bo$2bob2o15b5o17bo19bobo17bo2bo15bo20bobo16bo2bo15b2o17bo2bo16b2o2b
2o$o19bo2bo16b3o18b2o20b2o16bo19bo21bobo17bo16b2o18bo2bo$2o18b2o18bo
20bo39b2o18b2o21bo15bobo38bobo$62bo97b2o40bo$61b2o12$3b2o17b2o17bo18b
2ob2o18bo19bobo15bo19b2o21bo16b2o18b2o$2bo2bo15bobo16bobo2b2o13bob2obo
15b3o17b4obo13bobo3b2o12bobo2b2o16bobo15b2o3b2o14bo2b2o$bo2bobo14bo2b
2o15bobo2bo18bo14bo3b2o14bo5bo13bobo3bo13bo4bo15b3o2bo19bo14bo4bo$b2ob
obo15bobo18bobo15b4o16b2o2bo15b2o3b2o13bob2obo14bo5bo12bo3b3o14b4obo
13bo4bo$2bob2o15b2obo16bobob2o15bo20b2o17bo20bobo16bo3b2o13b3o16bo2bob
o14b2o2bo$o22bo16bobo17bo22bo16bo22bo19bobo18bo15b2o21bo$2o18bobo18bo
18b2o19bobo16b2o20b2o20b2o17b2o35b3o$20b2o59b2o117bo13$3bo18b2o16b2o
18b2o21b2o18b2o16bobo17b2o2b2o14b2ob2o18bo$2bobo18bo16bobo17bobob2o15b
3obo17bobo14bob2o16bobo2bo16bobo18bobo$bo2bo17bo19bo2b2o15bobo15bo5bo
13b2obo2bo13bo3b2o14bo5bo13bo4bo16bobobo$b2obob2o13bob3o16b2o2bo14b2o
2bo15bo5bo13bobob2o14b2o2bo15bo3b2o13b4obo16bobobo$2bob2obo13bo3bo18b
2o16bob2o16bo3b2o13bo2bo18b2o17bo2bo18bo15b2o3bo$o19b2o20b2o18bo20bobo
16bobo18bo19bobo16bo17bo$2o20b2o17bobo17b2o21b2o17bo20bo19bo17b2o17b3o
$22bobo17bo80b2o58bo$23bo12$6b2o13bo18bo19b2o21bo18b2obo15bo20b2obo15b
2o21bo$3b2o2bo13b3o16b3o17bo2b2o16b3o18bob2o14bobo3b2o12bo2b2obo15bo
20bobo$4bobo17bo18bo17b2o2bo14bo19b2o18bo2bo2bo13b2o4bo13bo21bobo$2bo
2bo15b2obo17bo2bo17b3o15bo19bob2obo14b2obobo14bob3o14b2o20bobobo$bob2o
16bo2b2o16b4o15b2o19bobo16bobob2o16bobo15bobo18bo17b2o3b2o$obo20bo16b
2o18bo2bo19b4o15bo20bo18bo17b2o18bo$obo18bobo16bo2bo17b2o24bo34b2o36bo
3b2o15b3o$bo18bobo19b2o41bobo73bo2bo18bo$21bo63b2o75bobo$163bo!
hkoenig wrote:Perhaps you should welcome the help, no matter what the format. Writing conversions for formats isn't that hard, once you understand the data format. Yours is not some sort of special, canonical format that everyone is require to adhere to.
The problem is that Shemyakin's RLE files are riddled with errors, so there is no inherent format in them. Many of the files start with something from Niemiec's database, and two lines are added in the middle that make the RLE invalid; the given rules for many files are also incorrect (though the cells are correct). Shinjuku was designed from the start to have as few errors in its files as possible, and to be easily machine-readable/writeable – open knowledge in its purest form.

I don't understand much of SQL. Perhaps you could tell me what files you need to easily load the synths into an SQLite database?
Princess of Science, Parcly Taxel

AforAmpere
Posts: 1049
Joined: July 1st, 2016, 3:58 pm

Re: 17 in 17: Efficient 17-bit synthesis project

Post by AforAmpere » April 27th, 2019, 5:19 pm

xs17_69ar2pm in 14:

Code: Select all

x = 144, y = 78, rule = B3/S23
$80bo$81bo$79b3o13$91bo$92bo14bo$90b3o14bobo$107b2o$100bo11bo$101bo9bo
$99b3o9b3o7$90bo$88bobo$89b2o3b3o$96bo$95bo2$3bo$4bo$2b3o2$91b3o$93bo
7b2o$92bo7bo2bo$101bobo$3bo98bo$bobo9b2o$2b2o9bobo$13bo90bo$103bobo$
103bo2bo$104b2o3$12b3o$12bo$13bo7$64b3o$66bo$65bo11$142bo$141b2o$141bo
bo!
EDIT, xs17_c9b8jdzw32 in 14:

Code: Select all

x = 144, y = 82, rule = B3/S23
7$67bo$68b2o$67b2o11$100bobo$100b2o$101bo$47bo$45bobo$46b2o$52bo55bo$
52bobo52bo$52b2o53b3o$94b2o5b3o$94b2o5bo$102bo2$104b3o$21bo28b2o9bo32b
2o8bo$22bo22bo4b2o8bo28bo4b2o9bo$20b3o21bobo13b3o25bobo$12b2o31b2o42b
2o$11bobo$13bo5b2o78b2o$18b2o79b2o$20bo36b3o$57bo$58bo8$132bo$131b2o$
131bobo18$130b3o$130bo$131bo!
I and wildmyron manage the 5S project, which collects all known spaceship speeds in Isotropic Non-totalistic rules.

Things to work on:
- Find a (7,1)c/8 ship in a Non-totalistic rule
- Finish a rule with ships with period >= f_e_0(n) (in progress)

User avatar
Billabob
Posts: 144
Joined: April 2nd, 2015, 5:28 pm

Re: 17 in 17: Efficient 17-bit synthesis project

Post by Billabob » April 27th, 2019, 7:35 pm

xs17_0br8jdz32 in 18 -> 16:

Code: Select all

x = 126, y = 58, rule = Life
73bo$74bo49bo$72b3o48bo$123b3o12$18bo41bo$17bo43bo$17b3o39b3o$11bobo
80bo$12b2o37bo7bo33bobo4b2o$12bo37bobo5b2o33bobo4b2o$50bobo5bobo33bo$
51bo$10b3o85b2o$obo7bo44b2o32b2o6bo2bo$b2o3bo4bo34b2o6bo2bo30bo2bo6b2o
$bo3bo39bo2bo6b2o32b2o$5b3o38b2o46bo$51bo41bobo4b2o$50bobo40bobo3bo2bo
$50bobo5b2o34bo5b2o$7b3o41bo6bobo$7bo50bo$8bo46b2o$54bobo$56bo39b2o$
91b2o3bobo$90bobo3bo$92bo2$94b2o$95b2o$94bo13$69b3o$71bo$70bo!
I'm amazed by the 3G synth finder script, if you told me 5 years ago something like this would exist I would call you a madman. But I was wondering if there's a way to filter results even further? It took me 20 minutes to find syntheses for each of those 2-beehives-at-specific-offsets.
▄▀
▀▀▀

fluffykitty
Posts: 638
Joined: June 14th, 2014, 5:03 pm

Re: 17 in 17: Efficient 17-bit synthesis project

Post by fluffykitty » April 27th, 2019, 8:33 pm

Synthesis of xs17_358m552z311 in 25G:

Code: Select all

x = 134, y = 134, rule = B3/S23
132bo$131bo$131b3o3$8bo$9bo$7b3o12$2bo$obo$b2o27$33bo33bo$34bo30bobo
24bo$32b3o4bo26b2o22b2o$40bo50b2o$38b3o4$46bo26bo$47b2o24bobo$46b2o25b
2o8$55bo$56bo$54b3o10$62b2o$61bobo$63bo$85b2o$85bobo$85bo2$102bo$101b
2o$101bobo$57b3o7b2o$59bo8b2o$58bo8bo2$35b2o$34bobo$36bo3$81bo$80b2o$
80bobo$46b2o$47b2o$46bo6$86b3o$86bo$35b2o50bo$34bobo$36bo5$126b3o$126b
o$127bo6$123b3o$123bo$124bo4$9b3o$11bo$10bo!
I like making rules

User avatar
dvgrn
Moderator
Posts: 5889
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: 17 in 17: Efficient 17-bit synthesis project

Post by dvgrn » April 27th, 2019, 9:09 pm

Billabob wrote:I'm amazed by the 3G synth finder script, if you told me 5 years ago something like this would exist I would call you a madman. But I was wondering if there's a way to filter results even further? It took me 20 minutes to find syntheses for each of those 2-beehives-at-specific-offsets.
You're using synthesise-constellation-ee9.py, right, not the synthesise-patt version? So the problem is not finding two beehives at specific offsets, it's finding one out of hundreds or thousands of those patterns where the reaction envelope doesn't get in the way of some pre-existing pattern.

That can definitely be done, but we might have to write some new functions in Python, or maybe switch over to lifelib or something. The three-glider collisions don't all produce the output constellation in the same orientation, so we'd have to run each one to stability, figure out which orientation it is, add whatever nearby obstacles we want to add to the initial state, and re-run the collision to see if everything still survives. It definitely seems worth having this as an option.

Also it seems like the current script is extremely fast, and so maybe it's time to make it work harder. We could process a sizable subset of four-glider syntheses and add all the interesting stuff to the search database -- all active reactions that drop to a bottleneck below N bits (for synthesise-patt.py) or that stabilize into three- or maybe four-object constellations (for synthesise-constellation.py), along with anything else we can think of to look for while we're doing the survey.

AforAmpere
Posts: 1049
Joined: July 1st, 2016, 3:58 pm

Re: 17 in 17: Efficient 17-bit synthesis project

Post by AforAmpere » April 27th, 2019, 9:23 pm

Dropping the sl that fluffykitty synthesized up above to 22 using a similar tactic:

Code: Select all

x = 448, y = 178, rule = B3/S23
22$323bo$322bo$322b3o7$202bo$203bo$201b3o13$199bo$200bo$198b3o34$97bo
59bo109bo$35bo60bobo57bobo107bobo$36bo59bobo57bobo78bo28bobo129b2o2b2o
$34b3o60bo59bo80bo28bo130bo2bo2bo$236b3o160bob3o$92b2o7b2o49b2o7b2o70b
o28b2o7b2o127bo$34b2o55bo2bo5bo2bo47bo2bo5bo2bo70bo26bo2bo5bo2bo127bo$
35b2o55b2o7b2o49b2o7b2o69b3o27b2o7b2o125b3o$34bo363bo$97bo59bo109bo$
96bobo57bobo107bobo$96bobo57bobo3bo85bo17bobo$92b2o3bo54b2o3bo4bo83bob
o4b3o6b2o3bo3b3o$91bo2bo56bo2bo7bo84b2o4bo7bo2bo$92b2o58b2o100bo7b2o$
249bo$249b2o$133bobo112bobo$108b2o24b2o$108bobo23bo254b2o14b2o$108bo
279bo2bo12bo2bo$98b3o288b2o14b2o$100bo46bo109bo124b2o$89bo9bo46bobo
107bobo122bo2bo$90b2o54bobo101b3o3bobo123b2o$89b2o56bo109bo$17b2o112b
2o102b3o$16bobo113b2o103bo$18bo70b3o39bo9b2o93bo14b2o$91bo49b2o108b2o$
90bo292b3o2$421b3o$421bo$406b2o14bo$405b2o$407bo11$63b2o$64b2o$63bo18$
402b3o$402bo$403bo!
Hopefully there is some way to reduce this.
I and wildmyron manage the 5S project, which collects all known spaceship speeds in Isotropic Non-totalistic rules.

Things to work on:
- Find a (7,1)c/8 ship in a Non-totalistic rule
- Finish a rule with ships with period >= f_e_0(n) (in progress)

User avatar
Billabob
Posts: 144
Joined: April 2nd, 2015, 5:28 pm

Re: 17 in 17: Efficient 17-bit synthesis project

Post by Billabob » April 27th, 2019, 9:54 pm

dvgrn wrote: You're using synthesise-constellation-ee9.py, right, not the synthesise-patt version? So the problem is not finding two beehives at specific offsets, it's finding one out of hundreds or thousands of those patterns where the reaction envelope doesn't get in the way of some pre-existing pattern.

That can definitely be done, but we might have to write some new functions in Python, or maybe switch over to lifelib or something. The three-glider collisions don't all produce the output constellation in the same orientation, so we'd have to run each one to stability, figure out which orientation it is, add whatever nearby obstacles we want to add to the initial state, and re-run the collision to see if everything still survives. It definitely seems worth having this as an option.

Also it seems like the current script is extremely fast, and so maybe it's time to make it work harder. We could process a sizable subset of four-glider syntheses and add all the interesting stuff to the search database -- all active reactions that drop to a bottleneck below N bits (for synthesise-patt.py) or that stabilize into three- or maybe four-object constellations (for synthesise-constellation.py), along with anything else we can think of to look for while we're doing the survey.
You're a life saver, thank you so much. I was indeed using synthesise-patt and scrolling through thousands of traffic lights to find the beehive constellations I needed. Quite embarrassing that I missed this script, although it was hard to find synthesise-patt in the first place.

That's a fantastic idea that would definitely come in handy. You could use LifeHistory like this, but I'm not sure how much slower it would be:

Code: Select all

x = 7, y = 13, rule = LifeHistory
5.D$4.D2B$3.B3D$2.4B$.4B$2.2B$2.2B$3.2AB$2.A2BA$.BA2BA$2DB2A$B2D$DB!
▄▀
▀▀▀

User avatar
Freywa
Posts: 589
Joined: June 23rd, 2011, 3:20 am
Location: Singapore
Contact:

Re: 17 in 17: Efficient 17-bit synthesis project

Post by Freywa » April 27th, 2019, 10:04 pm

As a first application of findpreds.py, xs17_4ai3gzx11d96 in 9:

Code: Select all

x = 129, y = 57, rule = B3/S23
6bo$4bobo$5b2o22$20bo$19bobo93bo$19bobo92bobo$20bo93bo2bo8bo$113b2o3bo
5bobo$117b2o6b2o$40bobo75bob2o6bo$40b2o76bo2bo6bo$41bo77b2o7bo$59bo$
40b3o14b2o$40bo17b2o$41bo13$8b3o$10bo$9bo3$b2o$obo$2bo!
xs17_39qb96z32 in 10:

Code: Select all

x = 124, y = 22, rule = B3/S23
2bo$bobo$o2bo50bo$b2o50bobo$52bo2bo66bo$53b2o66bo$121b3o$59bo9bo$58b2o
4bobo2bobo$58bobo3b2o3b2o47b2o$54b3o8bo43b2ob2o3bo2bo$56bo51bob2obo4bo
bo$55bo10b2o40bo10bo$66bobo40b4o$66bo44bo$113bo$10bo101b2o$10bobo$10b
2o$13b2o46b2o$13bobo45b2o$13bo!
xs17_3lkaacz1221 in 9:

Code: Select all

x = 200, y = 26, rule = B3/S23
2o123bo$b2o121bo$o4b2o117b3o$5bobo59bo130b2o$5bo60bobo6b3o45bo8bo62b2o
2bo$66bobo53bobo7bo61bo2b2o$67bo54bobo7bo61b3o$123bo73b2o$6bo189bo2bo$
5b2o185bo4b2o$5bobo184bo$192bo3$69bo62b3o58b3o$70bo5bobo109b3o$68b3o5b
2o46bo65bo$73bo3bo45bobo63bo$73b2o48bobo$72bobo49bo2$191b2o$191b2o$
187b3o$189bo$188bo!
xs17_6s1ra96zw1 in 11:

Code: Select all

x = 113, y = 42, rule = B3/S23
23bo$24b2o$23b2o3$35bo$34bo$20bo13b3o69bo$21b2o84bo$20b2o83b3o2$110bo$
109bobo$109b2o2$41bo$40bobo$41bo65bob2o$106bob2o2bo$106bo4b2o$33bo73b
3obo$34bo74bobo$32b3o75bo3$33bo$32b2o3b2o$32bobo2bobo$37bo11$2o$b2o$o!
xs17_178r9a4z032 in 14:

Code: Select all

x = 264, y = 30, rule = B3/S23
237bo$238bo$236b3o$240b2o$152bo87b2o$152bobo$152b2o$150bo$148bobo$149b
2o$156bo$157bo$155b3o$159bobo100bo$159b2o81b2ob2o14bo$160bo80bobobo15b
3o$240bo4bo12b2o$12bobo55bo162b2o6b4o12bo2bo$12b2o55bobo103bo57b2o8bo
14b2o$13bo55bobo102bobo52b3o13bo$obo56bo6bo3bo103bobo54bo12b2o$2o58bo
5bo104bo3bo54bo$bo3b2o51b3o5bo98bo5bo$4b2o158bobo4bo$6bo48b3o106bobo$
57bo107bo81b2o$56bo190b2o$125b2o123b3o$126b2o122bo$125bo125bo!
xs17_0cp3qicz32 in 7:

Code: Select all

x = 66, y = 16, rule = B3/S23
2o44b2o$obo43bobo14bo$bo45bo5bo8bobo$16bo36b2o7bo2bo$14bobo35bobo8b2o$
15b2o39b3o$56bo$19bo37bo$17b2o$18b2o2$55b2o$55b2o$18b2o$17b2o$19bo!
xs17_1no3tic in 10:

Code: Select all

x = 183, y = 21, rule = B3/S23
167bo$168bo$166b3o$181bo$170b3o7bo$3bo132bobo41b3o$4bo51bo58bo21b2o$2b
3o50bobo2b2o52bobo2b2o16bo40b2o$55bobo2b2o52bobo2b2o57b2o$5b2o49bo58bo
19b2o$2o4b2o126b2o$b2o2bo115bo14bo$o119bobo41bo$63b2o55bobo41b3obo$63b
obo55bo45b2o$63bo100b2o$164bob3o$60b2o61b2o40bo2bo$59bobo3bo57b2o41b2o
$61bo2b2o$64bobo!
xs17_39u0mqz023 in 10:

Code: Select all

x = 170, y = 22, rule = B3/S23
118bo$119b2o$118b2o3$120bo3b2o$53bo64bobo2b2o34b2o$53bobo63b2o4bo28bo
2bo2bo$53b2o99b6o$51bo$49bobo52b2o50bob2o$bo48b2o51bo2bo49b2obo$2bo
101b2o$3o57bo52bo$59bobo50bobo50b2o$3b3o53bobo50bobo49bo2bo$5bo54bo52b
o51b2o$4bo$168b2o$14bo54bo52bo39b3o3b2o$13bobo52bobo50bobo40bo$14b2o
53b2o51b2o39bo!
From Goldtiger997, xs17_32qkgoz3146 in 9:

Code: Select all

x = 79, y = 38, rule = B3/S23
69bo$70bo3b2o$16bo51b3o2bobo$16bobo55bo$16b2o$obo$b2o$bo$24bo$22b2o$
23b2o3$72b2o3b2o$73bo3bo$13bo57bo2b2obo$12b2o57b2obobo$12bobo59bo$19b
2o53b2o$18b2o$20bo8$3bo7b3o$3b2o6bo$2bobo7bo5$14b3o$14bo$15bo!
Last edited by Freywa on April 28th, 2019, 1:01 am, edited 1 time in total.
Princess of Science, Parcly Taxel

Hunting
Posts: 1117
Joined: September 11th, 2017, 2:54 am
Location: Ponyville, Equestria

Re: 17 in 17: Efficient 17-bit synthesis project

Post by Hunting » April 28th, 2019, 12:58 am

Freywa wrote:As a first application of findpreds.py, xs17_4ai3gzx11d96 in 9:

Code: Select all

x = 129, y = 57, rule = B3/S23
6bo$4bobo$5b2o22$20bo$19bobo93bo$19bobo92bobo$20bo93bo2bo8bo$113b2o3bo
5bobo$117b2o6b2o$40bobo75bob2o6bo$40b2o76bo2bo6bo$41bo77b2o7bo$59bo$
40b3o14b2o$40bo17b2o$41bo13$8b3o$10bo$9bo3$b2o$obo$2bo!
Sorry for another off-topic posting, but that synthesis includes a catalysis beehive.
This post was brought to you by the Element of Magic.

Plz correct my grammar mistakes. I'm still studying English.

Working on:

Nothing.

Favorite gun ever:

Code: Select all

#C Favorite Gun. Found by me.
x = 4, y = 6, rule = B2e3i4at/S1c23cijn4a
o2bo$4o3$4o$o2bo!

User avatar
Freywa
Posts: 589
Joined: June 23rd, 2011, 3:20 am
Location: Singapore
Contact:

Re: 17 in 17: Efficient 17-bit synthesis project

Post by Freywa » April 28th, 2019, 1:21 am

It's been raining syntheses left and right with the new script. xs17_4a9baa4zw252 in 9:

Code: Select all

x = 102, y = 18, rule = B3/S23
50bo$51bo$49b3o31bo15bo$84bo13bobo$52bo29b3o10bo2bobo$51bo34bo7bobobob
2o$51b3o31bobo7bo2bo2bo$5b2o79b2o10bobo$2bo2bobo38bo52bo$obo2bo39bobo$
b2o42bobo$46bo2$4b3o36bo41bo$4bo38bo5b2o35bo$5bo37bo4bobo2b3o28b3o$50b
o2bo34b2o$54bo33b2o!
Now all xs17s named by Flammenkamp have a synthesis.
Princess of Science, Parcly Taxel

User avatar
Freywa
Posts: 589
Joined: June 23rd, 2011, 3:20 am
Location: Singapore
Contact:

Re: 17 in 17: Efficient 17-bit synthesis project

Post by Freywa » April 28th, 2019, 6:09 am

I am staring at what could be the most insane developments in glider synthesis with the advent of the findpreds.py script. I ploughed through a dozen of the still lifes in a far shorter time than normal, and now only 64 xs17s remain unsynthed:

Code: Select all

x = 147, y = 150, rule = B3/S23
b2ob2o20bo13b2o3b2o16bo16b2ob2o16bo21bob2o14b2ob2o$bob2o19b3o13bo2bo2b
o14b3o2b2o13bobo2bo13bobo20b2obo14bob2obo$6bo16bo17bob3o14bo5bo13bo4b
2o13bobo18b2o23bo$2b2ob2o17bo17bo18bo5bo13b2o18b2ob2o14bo2b2obo15b4o$
2bo18b4o18bo18bo3b2o15bo19bobo15bobob2o13bo2bo$obo17bo19b3o20bobo14bob
o18b2o19bo17b2o$2o19b2o17bo23b2o14b2o19bo$22bo80bo$20bo81b2o$20b2o11$b
2o19b2o16b2o20b2o19bo17bo21b2o19bo$bo2b2o15bo2bo15bobo18bo2bob2o12bobo
bo15bobo20bobo17bobo$2bobo2bo13b2o2bo16bo17bob2o3bo12b2o2bo15bobo18b2o
2bo16bobobo$b2o3b2o18bo14b2ob2o15bo2b3o17b2o15b2ob2o14bo4b2o15bobobo$
3bo17b5o17bobo16bobo20bo17bobo15b2o17b2o3bo$obo17bo2bo17b2o20bo16b2o2b
o16b2o20bo16bo$2o18b2o19bo38bo2bo17bo18bobo18b3o$43bo37b2o19bo17b2o21b
o$42b2o57b2o12$3b2o17b2o16b2o20b2o19b2o16b2ob2o18bo19bo$2bobo16bobo16b
obo18bo2bo15bo2bo16bobobo18bobo17bobo$bo19bo2b2o16bo17bobo2bo14b3o2bo
14bo5bo14b3o2bo15bobo$bob4o15bobo16b2ob2o15bob3o17b3o15bo3b2o13bo3b3o
15bobobo$2bo3bo14b2obo18bobo16bo19bo19bo2bo15b3o16b2o3b2o$3bo19bo17b2o
20bo19bo19bobo18bo15bo$3o17bobo18bo18b3o17b3o21bo18b2o16b3o$o19b2o20bo
17bo19bo62bo$41b2o12$3bo18b2o16b2obo18b2o2b2o13bo19b2o18b2o21b2o$2bobo
18bo16bob4o15bo2bo2bo12bobo3b2o12bobo2b2o15bo17b2obobo$bo2bo17bo23bo
13bobobobo13bobo3bo13bo4bo14bo20bobo$b2obob2o13bob3o15b4obo14bo2bob2o
13bob2obo14bo5bo12b2o19bo2bo$2bob2obo13bo3bo15bo2b2o16b2o19bobo16bo3b
2o14bo17b2o3bo$o19b2o59bo21bobo14b2o20b3o$2o20b2o57b2o21b2o14bo3b2o16b
o$22bobo96bo2bo$23bo98bobo$123bo11$6b2o12b2o20b2o19bobo15bo19b2o2b2o
15bo17b2obo$3b2o2bo13bo19bobo17b4obo13bobo3b2o12bobo2bo15bobo16bob2o$
4bobo14bob2o15bo19bo5bo13bobo3bo13bo5bo13bo2bo20b2o$2bo2bo16bobo16b5o
15b2o3b2o13bob2obo14bo3b2o13b2o2b2o15b3o2bo$bob2o20b2o18bo16bo20bobo
16bo2bo20bo13bo4b2o$obo24bo15b2o16bo21bo19bobo18b2o14b2o$obo24b3o13bo
17b2o19b2o20bo19bo$bo28bo13bo77bobo$29b2o12b2o77b2o12$6b2o12b2o4b2o15b
o2bo16bobo15bobo18bo20bo16b2o$3b2o2bo13bo4bobo12b6o14b4obo13bob2o17bob
o17b3o16bo$4bobo14bob2obobo11bo19bo5bo13bo3b2o15bo2bo15bo21bo$2bo2bo
16bobo2bo13b2o2b2o14b2o3b2o13b2o2bo14b2o3bo14bob2o17b2o$bob2o20b2o16bo
2bo15bo20b2o19b2o15bobo16bo$2bo40b2o15bo22bo19bo20b2o14b5o$obo57b2o22b
o16b3o21bo19bo$2o81b2o15bo24bobo14bobo$100b2o24b2o14b2o12$5b2o14b2o2b
2o16bo19b2o16bo20b2o18bobo18bo$6bo13bo2bo2bo14b3o19bobo14bobo3b2o13bob
o18b2obo15b3o$3b3o15bob3o14bo3b2o14b2obo2bo13bo2bo2bo14bo2b2o14b2o3bo
14bo$3bo18bo18b2o2bo15bobob2o14b2obobo13b2o4bo13bo2b2o16bo$b2o20bo19b
2o16bo2bo18bobo18b2o15b2o17b2o$o19b3o20bo18bobo18bo20bo17bo17bo4bo$b2o
17bo20bobo19bo18b2o18bobo15bo20bo2bobo$2bo38b2o59b2o16b2o20bo2bo$2bobo
138b2o$3b2o11$6bo14bo21b2o17b2obo15b2o19b2o16b2o19b2o$4b3o13bobo2b2o
14b3obo16bob2o14bob3o16bobo17bo18bo2bo$3bo17bobo2bo13bo5bo13b2o18bo4bo
15bo4b2o12bo19b2o2bo$4bo18bobo15bo5bo13bob2obo14b2o2bo14b2o5bo12b2o23b
o$b4o16bobob2o15bo3b2o13bobob2o16bob2o15b2o2bo15bo17b2o2b2o$o19bobo20b
obo16bo20bo18bo2bo14b2o18bo2bo$b2o18bo22b2o36b2o19b2o15bo3b2o15bobo$2b
o118bo2bo17bo$bo120bobo$b2o120bo!
In fact, I should sit back for a while and receive syntheses from others as they come, and spend time on other things.
Princess of Science, Parcly Taxel

AforAmpere
Posts: 1049
Joined: July 1st, 2016, 3:58 pm

Re: 17 in 17: Efficient 17-bit synthesis project

Post by AforAmpere » April 28th, 2019, 11:38 am

xs17_69jwrdzx121 in 14:

Code: Select all

x = 191, y = 60, rule = B3/S23
7$12bo$11bo$11b3o3$54bo33bobo8bo$53bobo33b2o7bobo$53b2o34bo8b2o$41bo$
40bo$40b3o38b2o$4b2o13bo61bobo62b2o$5b2o11bo15b2o46bobo60bo2bo$4bo13b
3o14b2o46bo60bo2b2o$34bo4b3o101bo$39bo102bo$40bo20bo44bo36b3ob2o15bo$
60bobo29b3o10bobo37b2obo3b2o10bo$60b2o32bo10b2o44bobo10bo$93bo58bo3$
11b2o80b2o$12b2o80b2o$11bo81bo9bo$102b2o$102bobo2$174b3o$174bo$159b3o
13bo$159bo$160bo7$114b2o$114bobo$114bo!
I and wildmyron manage the 5S project, which collects all known spaceship speeds in Isotropic Non-totalistic rules.

Things to work on:
- Find a (7,1)c/8 ship in a Non-totalistic rule
- Finish a rule with ships with period >= f_e_0(n) (in progress)

User avatar
Freywa
Posts: 589
Joined: June 23rd, 2011, 3:20 am
Location: Singapore
Contact:

Re: 17 in 17: Efficient 17-bit synthesis project

Post by Freywa » April 28th, 2019, 11:58 am

We're now down to 21:

Code: Select all

x = 87, y = 90, rule = B3/S23
3b2o15b2o21b2o16bo21bo$2bobo15bobo17bo2bo16bobo18b3o$bo20bo17b3o2bo14b
obo17bo$bob4o14b2ob2o17b3o15b2ob2o15bo$2bo3bo16bobo16bo20bobo14b2o$3bo
17b2o20bo17b2o17bo4bo$3o18bo18b3o18bo19bo2bobo$o21bo17bo21bo19bo2bo$
21b2o38b2o20b2o12$b2o2b2o15b2o17bo19b2ob2o$o2bo2bo14bo2bob2o12bobo3b2o
12bobobo$bob3o14bob2o3bo12bobo3bo13bo5bo$2bo18bo2b3o14bob2obo14bo3b2o$
3bo18bobo18bobo16bo2bo$3o20bo17bo21bobo$o40b2o21bo14$bo21bobo15bo20bo$
obo2b2o14b4obo13bobo3b2o13bobo$bobo2bo13bo5bo13bobo3bo14bo2bo$3bobo15b
2o3b2o13bob2obo13b2o3bo$bobob2o15bo20bobo18b2o$obo18bo21bo19bo$bo19b2o
19b2o17b3o$60bo$60b2o12$2o3b2o16bobo15bo19b2o$o2bo2bo14b4obo13bobo3b2o
14bo$bob3o14bo5bo13bo2bo2bo13bo$2bo18b2o3b2o13b2obobo13b2o$3bo18bo20bo
bo16bo$3o17bo22bo16b2o$o19b2o20b2o16bo3b2o$61bo2bo$62bobo$63bo11$2o21b
o17bo18b2o$obo17bobobo15bobo18bo$2bo17b2o2bo15bobo17bo$b2ob2o18b2o15b
2ob2o14b2o$3bobo19bo17bobo16bo$b2o17b2o2bo16b2o17b2o$bo18bo2bo17bo18bo
3b2o$3bo17b2o20bo17bo2bo$2b2o38b2o18bobo$63bo!
Last edited by Freywa on April 30th, 2019, 10:12 am, edited 5 times in total.
Princess of Science, Parcly Taxel

AforAmpere
Posts: 1049
Joined: July 1st, 2016, 3:58 pm

Re: 17 in 17: Efficient 17-bit synthesis project

Post by AforAmpere » April 28th, 2019, 12:12 pm

xs17_0mlhikz3201 in 9G:

Code: Select all

x = 31, y = 32, rule = B3/S23
3$6bo$7b2o14bo$6b2o14bo$22b3o4$15bo$16b2o$15b2o2$26bo$25bo$25b3o$17bo$
15b2o$4b3o9b2o$6bo$5bo5$3b2o22b3o$4b2o21bo$3bo10b3o11bo$14bo$15bo!
I and wildmyron manage the 5S project, which collects all known spaceship speeds in Isotropic Non-totalistic rules.

Things to work on:
- Find a (7,1)c/8 ship in a Non-totalistic rule
- Finish a rule with ships with period >= f_e_0(n) (in progress)

User avatar
BlinkerSpawn
Posts: 1907
Joined: November 8th, 2014, 8:48 pm
Location: Getting a snacker from R-Bee's

Re: 17 in 17: Efficient 17-bit synthesis project

Post by BlinkerSpawn » April 28th, 2019, 6:25 pm

Freywa wrote:As a first application of findpreds.py...
A few reductions:

Code: Select all

x = 151, y = 53, rule = B3/S23
90bo$88b2o$89b2o10$145bo$122bo23bo$19bo100bobo21b3o$18bo102b2o$18b3o
60bobo63bo$60bo21b2o62bo$17bo8bo32bobo2b2o16bo63b3o$16bobo7bo32bobo2b
2o$16bobo7bo33bo19b2o59bo$17bo61b2o59bobo$66bo14bo58bobo$65bobo73bo$
65bobo$66bo71bo$138bo5b2o$138bo4bobo2b3o$26b3o39b2o75bo2bo$68b2o79bo$
18bo$17bobo$17bobo$18bo17$bo$b2o$obo!
LifeWiki: Like Wikipedia but with more spaceships. [citation needed]

Image

Post Reply