Page 1 of 1

True knight trail

Posted: December 11th, 2010, 9:42 pm
by ynotds
Leaving blocks every (4, 2) cells period 10, using the same 22 state rule provided in the "intermediate neighbourhood" thread:

Code: Select all

x = 32, y = 13, rule = WMPVN-45678bc_459a_22
2A20.SR$2A15.2T2.TUQNM$4.2A11.U.2O.UOPLIH$4.2A13.P.2JSPJKGDC$8.2A9.QR
KR2ENKEFBA$8.2A7.RQURLMFM2AIF3A$11.R2A2.RSTMLPMGHAH2AD2A$11.S2A2.QSMN
OHGKHBCAC3A$12.TU.H2A2NHIJCBFC2A$15.I2ALMGICDE4A$16.JK2.2A2D2A.A$20.
2ABC2A$21.A!
This actually works for all numbers of states of this rule > 10, but there does not appear to be any simple way to turn it into a knight ship.

Re: True knight trail

Posted: December 12th, 2010, 1:47 am
by Axaj
Well, I've successfully linked them together, eliminating debris inbetween, but there is still some at the end.

Code: Select all

x = 43, y = 31, rule = WMPVN-45678bc_459a_22
32.URQ$28.2S2.STPML$28.T.2N.TNOKHG$28.U.O.2IROIJFCB$18.POP6.U2.PQJQ2D
MJDEA$18.Q.Q7.QPTQKLEL2AHE2A$19.R.PFEF2.QRSLKOLFGAG2AC2A$18.TSU.GRGU.
PRLMNGFJGABAB2A$18.U2.PQHTF3A2MGHIBAEB2A$20.URJIK.AHAKLFHBCD4A$16.2S
2.STKMLFGAJ2.2A2C2A$16.T.2N.TLNKH3A2.3AB2A$16.U.O.2IROIJACB$6.POP6.U
2.PQJQ2DMJBDA$6.Q.Q7.QPTQKLEL2AHE2A$7.R.PFEF2.QRSLKOLFGAG2AC2A$6.TSU.
GRGU.PRLMNGFJGABAB2A$6.U2.PQHTF3A2MGHIBAEB2A$8.URJIK.AHAKLFHBCD4A$4.
2S2.STKMLFGAJ2.2A2C2A$4.T.2N.TLNKH3A2.3AB2A$4.U.O.2IROIJACB$3.U2.PQJQ
2DMJBDA$4.QPTQKLEL2AHE2A$.U.QRSLKOLFGAG2AC2A$L2A2RLMNGFJGABAB2A$M2APQ
KMGHIBAEB2A$.NO.B2A2HBCD4A$4.C2AFGAC2A$5.DE2.4A$10.A!
EDIT: Found a different way.

Code: Select all

x = 22, y = 34, rule = WMPVN-45678bc_459a_22
11.UT$12.SPO$8.2Q2.QRNKJ$8.R.2LURLMIFE$8.STMT2GPMGHD2A$6.TS.TNOHO2BKH
BCA$5.SU.ONROIJCJ2AFC2A$5.RQNPQJIMJDEAE4A$7.MLIKLEDHE2A.2A$9.HGDFG2AC
2A$10.UCB2AB2A$10.TQP3A$6.2R2.RSOLK$6.S.2M.SMNJGF$6.TUNU2HQNHIEBA$4.U
T.UOPIP2CLICDA$3.T2.POSPJKDK2AGD2A$3.SROQRKJNKEFAF2ABA$5.NMJLMFEIF6A$
7.IHEGH2AD3A$9.DCABC3A$8.URQ4A$4.2S2.STPML$4.T.2N.TNOKHG$4.U.O.2IROIJ
FCB$3.U2.PQJQ2DMJDEA$4.QPTQKLEL2AHE2A$.U.QRSLKOLFGAG2AC2A$L2A2RLMNGFJ
GABAB2A$M2APQKMGHIBAEB2A$.NO.B2A2HBCD4A$4.C2AFGAC2A$5.DE2.4A$10.A!

Knight meets his dragon

Posted: December 13th, 2010, 9:35 am
by ynotds
dragon.zip
Open in Golly 2.2
(65.31 KiB) Downloaded 464 times
I'd expected something like this, but nothing so spectacular. Found in the wild from the first seed I ran on for WMPVN-45678bc_459a_23 and cleaned up, hence the zip with the rule tree for _23. The same initial mechanism in _22 produces a comparably elaborate but otherwise very different looking dragon. I don't consider it urgent to test in other rules, though as mentioned previously the knight track layer exists in every rule from _11 up.

Re: True knight trail

Posted: December 13th, 2010, 10:50 am
by ebcube
A way cleaner dragon on C22 (dragonrake?):

Code: Select all

x = 104, y = 57, rule = WMPVN-45678bc_459a_22
$98.K$97.LJL$97.MKM$97.NONPS$95.LM2ORQR$94.KJKONUQ$94.HIOMPM$91.FGHGJ
IQ2L$90.AB2HGJQOURS$90.ADC2JIRP2UT$87.3ACBJINKLTU.U$87.2A2CBKI2NMTST$
87.ACBGDEMNONOR$86.2ADB2GFMLMPQ$86.3AFGHGHK$88.AFEFIJ$84.2A.3AD$84.2A
2.BC$80.2A$80.2A$76.2A$76.2A$72.2A$72.2A$68.2A$68.2A$64.2A$64.2A$60.
2A$60.2A$56.2A$56.2A$52.2A$52.2A$48.2A$48.2A$44.2A$44.2A$40.2A$40.2A$
36.2A$36.2A$10.RS20.2A$8.MNQUT2.2T15.2A$6.HILPOU.2O.U11.2A$4.CDGKJPS
2J.P13.2A$3.ABFEKN2ERKRQ9.2A$2.3AFI2AMFMLRUQR7.2A$2.2AD2AHAHGMPLMTSR
2.2AR$2.3ACACBHKGHONMSQ2.2AS$5.2ACFBCJIH2N2AH.UT$5.4AEDCIGML2AI$6.A.
2A2D2A2.KJ$8.2ACB2A$12.A!
You can toggle between dragons (again, on C22) with blocks:

Code: Select all

x = 101, y = 57, rule = WMPVN-45678bc_459a_22
$97.K$96.LJL$96.MKM$96.NONPS$94.LM2ORQR$93.KJKONUQ$93.HIOMPM$90.FGHGJ
IQ2L$89.AB2HGJQOURS$89.ADC2JIRP2UT$86.3ACBJINKLTU.U$86.2A2CBKI2NMTST$
86.ACBGDEMNONOR$85.2ADB2GFMLMPQ$85.3AFGHGHK$87.AFEFIJ$83.2A.3AD$83.2A
2.BC$79.2A$61.2A16.2A$61.2A12.2A$75.2A$71.2A$71.2A$67.2A$67.2A$63.2A$
63.2A$59.2A$59.2A$55.2A$55.2A$51.2A$51.2A$47.2A$47.2A$43.2A$43.2A$39.
2A$21.2A16.2A$21.2A12.2A$35.2A$9.RS20.2A$7.MNQUT2.2T15.2A$5.HILPOU.2O
.U11.2A$3.CDGKJPS2J.P13.2A$2.ABFEKN2ERKRQ9.2A$.3AFI2AMFMLRUQR7.2A$.2A
D2AHAHGMPLMTSR2.2AR$.3ACACBHKGHONMSQ2.2AS$4.2ACFBCJIH2N2AH.UT$4.4AEDC
IGML2AI$5.A.2A2D2A2.KJ$7.2ACB2A$11.A!

And now a true knight ship in WMPVN-45678_45xbx_7+

Posted: April 23rd, 2011, 9:15 am
by ynotds
knightship.zip
Open in Golly 2.2
(992 Bytes) Downloaded 715 times
The attached seed is a p2x1y5t ship in born 45b, 45bc, 45ab, 45abc for all state counts greater than 7. An analogous p4x2y10t form works for state count 7.

Re: True knight trail

Posted: April 23rd, 2011, 4:13 pm
by ebcube
Not a knightship, but a rather interesting ship found on 45678_45b_8:

Code: Select all

x = 16, y = 8, rule = WMPVN-45678_45b_8
6.B.A.B$3.AB2C3A2CBA$.3A3CA.A3C3A$.2AD2AB.A.B2AD2A$.3AF2.3A2.F3A$3.A
2FA.A.A2FA$3.5A.5A$5.A5.A!

Re: True knight trail

Posted: April 23rd, 2011, 4:52 pm
by ynotds
ebcube wrote:(A) rather interesting ship found on 45678_45b_8:
I'd only just found as analogous ship in _28 but a version with a persistent asymmetry in its trailing debris which I thought hinted at the existence of a symmetric version like you found. And I've since only done a minimal survey which shows that the basic engine does things differently in adjacent rules, e.g. in _15 it leaves either a bilaterally symmetric or a left-right alternating widish trail of blocks. For higher numbers it also seems largely immune to interactions with much more common trailing c/2 ships, although that too warrants more exhaustive testing.