(Engineered) diehards

[77551enpassant]

The amount of digits in the lifespan of it is too long to fit in a forum post.

[bibunsekibun]

Then, what is the number of digits of the number of digits of the number of digits?

[EvinZL]

repeat that 13 times, that's roughly the scale

[bibunsekibun]

You can't talk about it unless you have digits of×13? The latest technology is amazing!

[Pavgran]

I'm not believing in the viability of my plan so much any more. Seems like one-time turners are inevitably going to take up much more room than is going to be available in that little rectangle, even if it's been made a little bigger.

Nevertheless, the plan is quite valid.
Edit: except the block is in wrong place... That mirror symmetry played a joke on me. The diehard is not working in this state

Code: Select all

x = 111, y = 90, rule = LifeSuper
21.6Q23.8Q.Q$19.10Q21.2Q.Q.Q4.3Q24.2M15.Q$21.Q4.Q26.Q3.2Q28.2M8.2Q5.Q
$15.2pA26.2M13.2Q2.Q33.Q2.Q4.Q$16.pA27.M5.Q6.Q39.2Q8.2Q$16.pA.pA25.M.
M6.2M3.Q.Q.Q21.2M21.Q.Q$17.2pA26.2M6.M.M28.2M22.Q$4.pA.pA5.pA43.M7.2M
$4.2pA5.pA.pA25.3O11.M2.M7.2M21.2M8.M$5.pA5.pA2.pA15.2Q6.3O15.M30.2M
6.3M4.2M$12.2pA16.2Q13.pA7.M.M38.M7.2M$7.pA36.pA.pA6.2M39.2M$7.3pA16.
pA18.pA$10.pA14.pA.pA14.pA5.pA50.2M7.3M$9.2pA15.pA4.O9.pA.pA3.pA.pA
18.Q30.2M6.M.M$31.2O7.pA.pA5.pA3.2pA13.Q.Q37.M$13.2pA16.2O6.pA.pA3.pA
6.2pA14.2Q32.M2.M.M.2M$13.2pA17.O5.pA.pA3.pA.pA9.pA5.2Q22.M.M13.2M3.
2M.M$19.pA17.pA.pA5.pA9.pA.pA3.Q2.Q23.M17.M.M.M$18.pA.pA4.pA10.pA.pA
17.pA.pA3.2Q20.M.M21.2M$2pA6.2pA9.pA4.pA.pA8.pA.pA19.pA.pA24.M23.M$pA
.pA5.2pA6.pA7.pA2.pA6.pA.pA21.pA24.M.M$pA14.pA.pA7.2pA6.pA.pA6.2O5.2O
4.pA27.2M$16.pA.pA13.pA.pA7.2O5.O4.pA.pA3.2pA20.M$17.pA.pA13.pA13.O.O
5.pA4.pA.pA20.2M$18.pA.pA26.2O12.2pA13.2M6.M$19.pA.pA15.2O37.2M$20.pA
.pA7.2pA5.2O$21.pA.pA6.2pA23.2M22.2M$16.2pA4.pA.pA30.M23.2M$.3pA13.pA
5.pA.pA16.M10.M.M13.2pA$.pA2.pA6.2pA4.pA.pA4.pA.pA14.M.M9.2M14.pA.pA
13.2S$.pA.2pA4.pA2.pA5.2pA5.pA.pA6.2M4.M.2M26.2pA4.2M6.S.S4.3M$9.2pA
15.pA.pA5.2M3.2M.2M22.pA9.2M6.2S5.3M$15.pA11.pA.pA10.M.2M21.pA.pA13.S
8.M3.M$6.pA8.3pA10.pA12.M.M22.pA6.S6.S.S8.5M10.Q.Q$5.2pA.pA9.pA12.pA
10.M29.S.S4.S.S10.4M11.4Q$17.2pA11.pA.pA39.S.S3.S.S24.Q.2Q$25.2O4.pA.
pA39.S5.S6.M18.Q4.Q$27.O4.pA.pA25.2S8.S5.S7.3M18.3Q.2Q$24.O8.pA25.S.S
7.S.S3.S.S5.M7.2M13.3Q.Q$25.2O31.S2.S.S6.S5.S6.2M6.2M12.Q3.2Q$58.2S2.
2S3.S5.S33.Q.Q$16.W3.O16.2pA16.O10.S.S3.S.S12.Q17.Q2.Q.Q$15.W4.3O8.W
5.2pA15.O.O10.S5.S7.M5.Q18.Q.Q$15.3W5.O7.2W22.O8.S5.S9.3M4.Q$22.2O7.
2W11.2U17.S.S3.S.S7.M3.M8.Q$32.W10.U2.U17.S5.S7.M.3M.M7.Q$15.W.W.2W6.
W16.U.2U13.S5.S11.5M8.Q$18.W.W6.2W16.U.U12.S.S3.S.S$15.2W.W8.2W17.2U
13.S5.S$22.W5.W35.S$15.W.W3.W.W39.S.S$15.2W4.W2.W39.S$15.W6.W.W47.S
25.2Q$23.W48.3S22.Q2.Q$11.W33.O29.S22.Q.Q$5.W3.2W8.2W23.O.O5.S21.2S
15.2M6.Q$4.W5.2W5.W.2W23.2O5.S.S29.2M6.2M$4.3W12.2W3.M5.M21.S3.2S25.M
$15.W3.W3.M7.M23.S.S26.3M$15.2W6.M.4M26.2S29.M7.2M$15.2W7.2M3.M2.M8.O
9.S42.2M9.Q2.Q.Q$16.W7.3M2.M2.M8.2O7.S.S53.Q.Q.Q$11.W15.M4.M8.2O8.2S
38.2M14.Q$11.2W14.3M.M10.O48.2M17.Q$11.2W14.2M.2M76.Q$W11.W14.2M4.3M
70.2Q2.Q$W.W4.W25.3M10.2pA57.3Q.2Q$2W5.2W27.2M8.2pA4.pA7.2S5.Q37.2Q3.
Q$7.2W27.2M13.pA.pA5.S.S4.Q.Q36.Q.Q.2Q$8.W27.2M14.pA6.S6.2Q38.Q$W39.
3M6.pA8.2S45.3Q2.Q$.3W.W32.2M4.M3.pA.pA54.2Q.Q.Q$.W2.2W32.M6.M3.pA$4.
W34.M.2M11.2U.2U14.M$W.2W.W31.3M3.M10.U.2U2.U2.2M3.M4.2M$2W35.4M2.M
15.2U2.2M2.2M5.2M$W40.M.M24.2M4.M26.3M4.M$.W.W37.3M6.2U.2U14.M3.M26.M
2.M4.2M$41.2M2.M4.U.2U2.U43.M2.M2.M.M.M$43.2M10.2U50.2M.M$105.M.M.2M$
9.4W52.M2.2M23.M13.M$9.W3.W30.W3.W14.3M2.2M2.M16.M.5M3.2M5.2Q.Q$9.W
29.W4.W3.2W16.2M2.3M18.2M.2M4.M4.Q.2Q.Q$10.W2.W26.W.W3.2W.3W14.2M2.3M
19.2M2.2M11.Q$39.W10.2W14.2M6.M2.M9.2M2.2M.M14.Q$39.W4.2W2.3W16.M6.3M
10.M3.2M.6M2.2M$41.8W39.M2.M3.5Q2.2M!

[ihatecorderships]

How do we tell if a diehard works or not when we can't run it for a respectable portion of its lifespan in a respectable amount of time?

[dvgrn]

How do we tell if a diehard works or not when we can't run it for a respectable portion of its lifespan in a respectable amount of time?

The same way we check whether the 0E0P metacell works. Conway's Life is much more predictable than Real Life, so we can carefully run all of the key reactions and see if they work. If all the pieces function as designed, and we check the math to make sure that things will always bounce back from the c/5 spaceship in the correct phase to interact with the oscillators at "home base", then we can complete a believable proof of diehardiness.

If you want, you can even write a script to take out the c/5 spaceship and replace it with permanent glider eaters a safe distance away -- then carefully paste in a temporary c/5 spaceship every now and then and re-remove it once it has caught a key glider. That might allow the diehard to be run to completion in a "respectable amount of time".

[Pavgran]

How do we tell if a diehard works or not when we can't run it for a respectable portion of its lifespan in a respectable amount of time?

I have assembled a demo pattern which is the same diehard except some things are missing and the ship is in different place.
You can run it in reasonable amount of time (it takes my Golly about 5 minutes to run it with base step 64 and hyperspeed turned on) and see that it reaches 0 population. Then you can confirm that changing the position and/or phase of c/5 ship does nothing but changing initial block positions. Then you can confirm that the original pattern starts off well and first sawtooth manages to produce a block. You can then confirm that broken snakes successfully eat the block during block-moving reaction. All of that requires no pattern manipulation.
For the second sawtooth, you can manipulate block position and delete blocking glider from the first sawtooth to release the glider pair in second sawtooth. You can confirm that toads and beehives also eat the block cleanly, and that initial block (it's not present in the most recent version of the diehard) successfully manages to eat some gliders from the second sawtooth. Hopefully that should give you enough insight of how the pattern works to be make you convinced that it is indeed a diehard.

[Pavgran]

On the idea of ignoring the bounding box constraint and just trying to make reasonably small and very long-lasting diehard.

I've assembled a pattern that is repeatable with vertical displacement of 83 and time displacement of 111+256*n, which implements stacked recursive filters.

Code: Select all

x = 230, y = 503, rule = LifeSuper
198.2C$197.B2C2B3.2C$198.4B2.B2CB$194.B.6B3.2B$183.2O7.10B2.2B$174.2C
8.O7.11B2C2B$173.B2CB7.O.O4.12B2C3B.B$174.2B9.2O4.18B2C$169.B3.2B17.
15B.B2C$168.2CB.4B11.4H15B3.B$168.2C8B10.BA3BA12B$169.B.B2C6B2.2B2.BA
2BAB3A10B$172.2C15BDABABA10B$172.16BADBD4B.7B$168.19B2A3D4B2.6B$168.
20BABAD2BA2B.7B$167.2C19BABA7B.6B6.S$167.2C14B.4B7.9B5.3S$168.B.11B2.
4B9.8B4.S$170.10B2.4B11.8B3.2S$171.14B13.7B.H$170.14B15.7BH$171.13B
15.2B3D2BH$171.13B15.2BD4BH$173.2B.4B3DB15.B3D4B.2B$176.H4BD2B15.13B$
176.H2B3D2B15.13B$176.H7B15.14B$176.H.7B13.14B$173.2O3.8B11.4B2.10B$
174.O4.8B9.4B2.11B.B$171.3O5.9B7.4B.14B2C$171.O6.6B.29B2C$178.7B.5BD
23B$179.6B2.4B3D21B$179.7B.4BDBD17B$178.15BD15B2C$178.19B2.2B2.6B2CB.
B$177.18B10.8B2C$173.B3.15B4H11.4B.B2C$172.2CB.15B17.2B3.B$172.2C18B
15.2B$173.B.3B2C12B14.B2CB$176.2B2C11B16.2C$177.2B2.10B$176.2B3.6B.B$
175.B2CB2.4B$176.2C3.2B2CB$183.2C8$212.H4.2A5.2A$212.HD.2B2A5.A$212.H
D4B3.BA.A$212.HD6B.B2A$213.D4B2A2B$213.5BABAB$174.2A36.H6BAB.2A$173.A
.A36.H6B3.3B$167.2A4.A38.HB3.B.BA3BA$165.A2.A2.2A.4A34.H7.A4BA$165.2A
.A.A.A.A2.A43.BABA.A$168.A.ABABAB47.A.ABAB$168.A.AB2AB49.A4BA$169.AB.
2B51.A3BA$172.3B50.3B$172.4B6.2A42.2A$170.3B2AB6.A$170.D2B2AB3.BA.A$
168.HBD7B.B2A$168.H11B$168.H3B3D5B$168.H3BD7B$170.3BD4B$172.6B$170.9B
$170.2A4.4D$171.A5.4H$168.3A27.2C$168.A28.B2C2B3.2C$198.4B2.B2CB$194.
B.6B3.2B$192.10B2.2B$174.2C16.11B2C2B$173.B2CB14.12B2C3B.B$174.2B15.
18B2C$169.B3.2B17.15B.B2C$168.2CB.4B11.4H15B3.B$168.2C8B10.18B$169.B.
B2C6B2.2B2.19B$172.2C15BD15B$172.17BDBD4B.7B$168.21B3D4B2.6B$168.23BD
5B.7B$167.2C29B.6B6.S$167.2C14B.4B7.9B5.3S$168.B.11B2.4B9.8B4.S$170.
10B2.4B11.8B3.2S$171.14B13.7B.H$170.14B15.7BH$171.13B15.2B3D2BH$171.
13B15.2BD4BH$173.2B.4B3DB15.B3D4B.2B$176.H4BD2B15.13B$176.H2B3D2B15.
13B$176.H7B15.14B$176.H.7B13.14B$178.8B11.4B2.10B$179.8B9.4B2.11B.B$
179.9B7.4B.14B2C$178.6B.29B2C$178.7B.5BD23B$179.6B2.4B3D21B$179.7B.4B
DBD17B$178.5B3A7BD15B2C$178.4BABABABA8B2.2B2.6B2CB.B$177.5BA3BAB2A5B
10.8B2C$173.B3.5B2A2BABA3B4H11.4B.B2C$172.2CB.8BA5BA17.2B3.B$172.2C
10BA2BAB2AB15.2B$173.B.3B2C5BAB2A3B14.B2CB$176.2B2C11B16.2C$177.2B2.
10B$176.2B3.6B.B$175.B2CB2.4B$176.2C3.2B2CB$183.2C7$161.2M$161.M.M48.
H4.2A5.2A$163.M48.HD.2B2A5.A$163.2M47.HD4B3.BA.A$212.HD6B.B2A$213.D4B
2A2B$213.5BABAB$174.2A36.H6BAB.2B$173.A.A36.H6B3.BAB$167.2A4.A38.HB3.
B.6B$165.A2.A2.2A.4A34.H7.BAB3A$165.2A.A.A.A.A2.A43.2B2A.A$168.A.ABAB
AB47.A.2A2B$76.2A90.A.AB2AB49.3ABAB$77.A91.AB.2B51.5B$77.A.AB2.B5.H
82.3B50.BAB$78.2AB.3B3.DH82.4B6.2A42.2B$80.6B.BDH80.3B2AB6.A$80.8BDH
16.2A62.D2B2AB3.BA.A$80.2B2A4BD18.A9.H50.HBD7B.B2A$81.A2BA3B19.A.AB5.
DH50.H11B$81.BABA3BH19.2AB.4BDH50.H3B3D5B$82.BA4BH21.6BDH50.H3BD7B$
82.3B2.BH20.7BD53.3BD4B$82.3B3.H19.8B56.6B$80.3B15.A8.10B53.9B$75.B2.
2A3B15.3A7.9B53.2A4.4D$74.2ABA2BAB2A17.A5.10B54.A5.4H$74.A4BA2B2A16.
2A4.11B.H49.3A27.2C$75.3BAB2.B17.5B.11BDH49.A28.B2C2B3.2C$75.ABA16.4H
4.3B.11BDH79.4B2.B2CB$95.4D2.16BDH75.B.6B3.2B$94.23BD74.10B2.2B$94.
23B57.2C16.10BACD2B$94.22B57.B2CB14.2A3B3A3BACD3B.B$94.21B59.2B15.2A
3B2AB3A2BA4B2C$95.19B55.B3.2B17.5BAB5A3B.B2C$96.17B55.2CB.4B11.4H8BAB
3A2B3.B$96.16B56.2C8B10.18B$96.15B58.B.B2C6B2.2B2.19B$95.15B62.2C15BD
15B$94.15B63.17BDBD4B.3BA3B$93.16BH58.21B3D4B2.B4AB$92.17BH58.23BD5B.
BA3BAB$92.H14B.BH57.2C29B.BA2BAB6.S$92.H13B3.H57.2C14B.4B7.5B3AB5.3S$
92.H12B63.B.11B2.4B9.8B4.S$92.H11B66.10B2.4B11.8B3.2S$94.9B68.14B13.
7B.H$94.2B.4HB68.14B15.7BH$171.13B15.2B3D2BH$171.13B15.2BD4BH$173.2B.
4B3DB15.B3D4B.2B$176.H4BD2B15.13B$176.H2B3D2B15.13B$176.H7B15.14B$
176.H.7B13.14B$178.8B11.4B2.10B$179.8B9.4B2.11B.B$179.9B7.4B.14B2C$
178.6B.29B2C$178.7B.5BD23B$179.6B2.4B3D21B$179.7B.4BDBD17B$178.15BD
15B2C$178.19B2.2B2.6B2CB.B$177.17B11.8B2C$173.B3.15B4H11.4B.B2C$172.
2CB.15B17.2B3.B$172.2C18B15.2B$173.B.3B2C12B14.B2CB$176.2B2C11B16.2C$
177.2B2.10B$176.2B3.6B.B$175.B2CB2.4B$176.2C3.2B2CB$183.2C7$161.2M$
161.M.M48.H4.2A5.2A$163.M48.HD.2B2A5.A$163.2M47.HD4B3.BA.A$212.HD6B.B
2A$213.D4B2A2B$213.5BABAB$174.2A36.H6BAB.2B$173.A.A36.H6B3.3B$167.2A
4.A38.HB3.B.3B2AB$165.A2.A2.2A.4A34.H7.2B2ABA$165.2A.A.A.A.A2.A43.4B.
A$168.A.ABABAB47.A.4B$76.2A90.A.AB2AB49.AB2A2B$77.A91.AB.2B51.B2A2B$
14.2M61.A.AB2.B5.H82.3B50.3B$14.2M62.2AB.3B3.DH82.4B6.2A42.2B$80.6B.B
DH80.3B2AB6.A$80.8BDH16.2A62.D2B2AB3.BA.A$80.2B2A4BD18.A9.H50.HBD7B.B
2A$81.A2BA3B19.A.AB5.DH50.H11B$81.BABA3BH19.2AB.4BDH50.H3B3D5B$82.BA
4BH21.6BDH50.H3BD7B$6.2M74.3B2.BH20.7BD53.3BD4B$6.2M74.3B3.H19.8B56.
6B$18.M61.3B15.A8.10B53.9B$17.3M55.B2.2A3B15.3A7.9B53.2A4.4D$17.3M54.
2A4B4A17.A5.10B54.A5.4H$74.2A2B2AB3A16.2A4.11B.H49.3A27.2C$16.4M55.3B
AB2.B17.5B.11BDH49.A28.B2C2B3.2C$16.5M54.3B16.4H4.3B.11BDH79.4B2.B2CB
$17.M.2M74.4D2.16BDH75.B.6B3.2B$94.23BD74.10B2.2B$2.M43.2M46.23B57.2C
16.11B2C2B$.M3.M36.3M.2M46.22B57.B2CB14.12B2C3B.B$M5.M35.M3.2M46.21B
59.2B15.18B2C$.2M.2M11.M.M23.2M.M48.19B55.B3.2B17.15B.B2C$16.M2.M25.M
50.17B55.2CB.4B11.4H15B3.B$5.M10.M3.M75.16B56.2C8B10.18B$2.4M9.2M2.3M
74.15B58.B.B2C6B2.2B2.19B$2.M.M11.M3.M74.15B62.2C15BD15B$2.M14.3M74.
15B63.17BDBD4B.7B$.3M14.M74.16BH58.21B3D4B2.6B$2M.2M87.17BH58.23BD5B.
7B$.3M88.H14B.BH57.2C29B.6B6.S$2.M89.H13B3.H57.2C14B.4B7.9B5.3S$92.H
12B63.B.11B2.4B9.8B4.S$92.H11B66.10B2.4B11.8B3.2S$39.2M53.9B68.14B13.
7B.H$30.2M7.2M53.2B.4HB68.14B15.7BH$16.3M10.M.M139.13B15.2B3D2BH$32.M
138.13B15.2BD4BH$14.M3.M5.M3.M2.M141.2B.4B3CB15.B3D4B.2B$14.4M5.M.M2.
M147.H4BC2B15.13B$14.M9.M3.3M145.H2B3C2B15.13B$28.2M146.H7B15.14B$
176.H.7B13.14B$178.8B11.4B2.10B$179.8B9.4B2.11B.B$179.9B7.4B.14B2C$
178.6B.29B2C$178.7B.5BD23B$179.6B2.4B3D21B$179.7B.4BDBD17B$178.15BD
15B2C$178.19B2.2B2.6B2CB.B$177.17B11.8B2C$173.B3.15B4H11.4B.B2C$172.
2CB.15B17.2B3.B$172.2C18B15.2B$173.B.3B2C12B14.B2CB$176.2B2C11B16.2C$
177.2B2.10B$176.2B3.6B.B$175.B2CB2.4B$176.2C3.2B2CB$183.2C7$161.2M$
161.M.M48.H4.2A5.2A$163.M48.HD.2B2A5.A$163.2M47.HD4B3.BA.A$212.HD6B.B
2A$213.D4B2A2B$213.5BABAB$22.2M150.2A36.H6BAB.2B$22.2M149.A.A36.H6B3.
3B$167.2A4.A38.HB3.B.3B3A$165.A2.A2.2A.4A34.H7.2B3AB$165.2A.A.A.A.A2.
A43.B3A.B$168.A.ABABAB47.B.3AB$76.2A90.A.AB2AB49.B3A2B$77.A91.AB.2B
51.3A2B$14.2M61.A.AB2.B5.H82.3B50.3B$14.2M62.2AB.3B3.DH82.4B6.2A42.2B
$80.6B.BDH80.3B2AB6.A$80.8BDH16.2A62.D2B2AB3.BA.A$80.2B2A4BD18.A9.H
50.HBD7B.B2A$81.A2BA3B19.A.AB5.DH50.H11B$47.2M32.BABA3BH19.2AB.4BDH
50.H3B3D5B$47.3M32.BA4BH21.6BDH50.H3BD7B$48.2M32.3B2.BH20.7BD53.3BD4B
$26.M22.M3.M28.3B3.H19.8B56.6B$25.3M25.2M25.3B15.A8.10B53.9B$24.2M2.M
17.2M4.M2.2M18.B2.B4A15.3A7.9B53.2A4.4D$13.M12.3M18.3M3.M.2M17.2A7BA
17.A5.10B54.A5.4H$10.2M.M33.4M2.3M18.2A2B2A3BA16.2A4.11B.H49.3A27.2C$
8.2M38.3M24.3B2A2.A17.5B.11BDH49.A28.B2C2B3.2C$6.M2.M2.2M5.M55.3B16.
4H4.3B.11BDH79.4B2.B2CB$5.M.M.2M2.2M3.M.M74.4D2.16BDH75.B.6B3.2B$4.3M
.2M8.M.M5.M28.2M37.23BD74.10B2.2B$5.2M5.M13.M28.2M37.23B57.2C16.11B2C
2B$6.M19.M67.22B57.B2CB14.12B2C3B.B$7.5M82.21B59.2B15.18B2C$8.2M85.
19B55.B3.2B17.15B.B2C$9.M86.17B55.2CB.4B11.4H15B3.B$96.16B56.2C8B10.
18B$96.15B58.B.B2C6B2.2B2.19B$28.M18.2M46.15B62.2C15BD15B$27.M.M17.2M
45.15B63.17BDBD4B.7B$27.M2.M62.16BH58.21B3D4B2.6B$29.2M61.17BH58.23BD
5B.7B$25.M3.2M2.2M.2M54.H14B.BH57.2C29B.6B6.S$25.M3.2M4.M.2M53.H13B3.
H57.2C14B.4B7.9B5.3S$25.M3.2M2.2M.M55.H12B63.B.11B2.4B9.8B4.S$92.H11B
66.10B2.4B11.8B3.2S$94.9B68.14B13.7B.H$94.2B.4HB68.14B15.3BA3BH$171.
13B15.2B2DC2BH$171.13B15.2BDBABAH$173.2B.4B3DB15.B3DA3B.2B$176.H4BD2B
15.5BABA5B$176.H2B3D2B15.7BA5B$176.H7B15.5BABA6B$176.H.7B13.6B2A6B$
178.8B11.4B2.10B$179.8B9.4B2.11B.B$179.9B7.4B.14B2C$178.6B.29B2C$178.
7B.5BD23B$179.6B2.4B3D21B$179.7B.4BDBD17B$178.15BD15B2C$178.19B2.2B2.
6B2CB.B$177.17B11.8B2C$173.B3.15B4H11.4B.B2C$172.2CB.15B17.2B3.B$172.
2C18B15.2B$173.B.3B2C12B14.B2CB$176.2B2C11B16.2C$177.2B2.10B$176.2B3.
6B.B$175.B2CB2.4B$176.2C3.2B2CB$183.2C2$205.2O$205.O.O$207.O$207.2O$
30.2M$30.2M129.2M$161.M.M$44.M118.M$42.3M118.2M$44.M$35.M2.2M5.M$39.
4M2.M$22.2M11.3M2.2M3.M$22.2M6.2M10.2M$29.M11.2M$42.M4.2M3.3M$47.M3.
4M$46.M2.3M.M$52.M23.2A$47.M4.M24.A$50.2M5.M19.A.AB2.B5.H$56.3M19.2AB
.3B3.DH$56.M.2M20.6B.BDH$59.M20.8BDH16.2A$59.3M18.2B2A4BD18.A9.H$23.M
57.A2BA3B19.A.AB5.DH$32.3M46.BABA3BH19.2AB.4BDH$18.M.2M10.M.M47.BA4BH
21.6BDH$18.M.M6.M2.2M2.M47.3B2.BH20.7BD$17.M4.2M3.M2.3M30.2M17.3B3.H
19.8B$18.M8.M2.3M30.2M15.2AB15.A8.10B$18.M7.M48.B2.2B3A15.3A7.9B$74.
2A2BA2BABA17.A5.10B$22.M51.2A2B2A2B2A16.2A4.11B.H$75.3B2A2.B17.5B.11B
DH$75.3B16.4H4.3B.11BDH$95.4D2.16BDH$55.2M37.23BD$55.2M37.23B$41.M.M
50.22B$36.2M.3M.2M49.21B$36.4M4.M50.19B$36.2M5.3M50.17B$43.M52.16B$
34.3M59.15B$95.15B$94.15B$93.16BH$92.17BH$92.H14B.BH$92.H13B3.H$92.H
12B$92.H11B$94.9B$94.2B.4HB16$48.3M$48.3M$47.M3.M$48.2M.2M$49.M.2M$
48.2M.M2$48.3M$139.2O$38.2M99.O.O$38.2M101.O$141.2O2$56.M$35.2M13.M.
6M$34.3M13.M7.M$32.2M.M.M11.M7.M$32.3M.3M11.2M5.M$31.M2.M12.2M.M3.M.M
$47.M.M8.3M$47.M10.M2.M$57.M2.2M$57.M$58.3M$59.3M$61.2M$61.2M$62.3M$
63.2M.2M$36.2M29.M$38.M28.M$30.2M.M30.M.M$31.M6.M25.M.M$28.M2.M$28.M
2.M4.4M31.2M$28.M2.3M.M2.M.M30.2M$31.M2.M4.M$30.2M5$63.2M$49.M13.2M$
49.2M$50.M$47.M.M$46.M2.M$46.M.M$46.M.M$47.M!

It utilizes a simple 3fd block pull and a robust reaction with a 2-engine cordership.
It took several iterations to achieve correct timing with minimal amount of reflectors.

Now, there is an open field of ideas that could help in transforming this into actual diehard. How exactly would it be constructed? Single-channel technology? c/2 rakes? c/10 rakes? What would trigger a stop in filter construction? Would the construction be re-triggerable? How exactly would the pattern trigger self-destruction?

I think I would rather not tackle this project right now, as I have still a lot of work to do in HaskLife project. Would someone like to volunteer to complete this idea?

[Sarah]

Random thought: I was thinking about how firing a gun at some faraway object could produce a really long lived diehard, then thought that that object could be a spaceship, then thought to put that here. Then realized that was what the diehard did already lol(well, on two layers).

[AlbertArmStain]

This thread got me wondering, is it posable to make a diehard that dies if and only if a statement: A, is true.
If this could be posable, then a diehard with and unknown lifespan exists.

[77551enpassant]

It should be possible to create a Universal Turing Machine in CGOL that halts iff a theorem such as the Riemann hypothesis is true/false. Then it would be trivial to wire it to self-destruct circuitry that destroys the entire machine. Or you could extend the lifespan even further by wiring it to repeated copies of the recursive filter Pavgran mentioned. The size of the bounding box of this pattern would be much more than 10000, so at that point it would become a problem of optimizing a UTM to fit within a small bounding box.
Edit: or just a regular Turing machine specially optimized for the problem of computing the Riemann hypothesis. It would also need to utilize sliding-block memory that can store aribtrary values, I think.
Edit 2: I should also mention that extremely large numbers like Graham's number have the potential to be calculated using relatively simple algorithms, so a pattern that calculates that would have a lifespan much bigger than anything discussed here.

[dvgrn]

This thread got me wondering, is it posable to make a diehard that dies if and only if a statement: A, is true.
If this could be posable, then a diehard with and unknown lifespan exists.

This idea got covered at the beginning of April on Discord, I think, and also here in this thread.

[calcyman]

Edit 2: I should also mention that extremely large numbers like Graham's number have the potential to be calculated using relatively simple algorithms, so a pattern that calculates that would have a lifespan much bigger than anything discussed here.

lumi (on the Discord) is building a machine with a lifespan of f_ε0(n) in the fast-growing hierarchy, which is vastly greater than Graham's number.

[Pavgran]

Laver tables are a good candidate on the role of the computation with unknown termination.

[GUYTU6J]

lumi (on the Discord) is building a machine with a lifespan of f_ε0(n) in the fast-growing hierarchy, which is vastly greater than Graham's number.

How is the machine designed and how large is it?

[gameoflifemaniac]

lumi (on the Discord) is building a machine with a lifespan of f_ε0(n) in the fast-growing hierarchy, which is vastly greater than Graham's number.

How is the machine designed and how large is it?

Hi, I am lumi from Discord. The way it will work is it uses a GPC to run a googological notation called the primitive sequence system (for which I have already made APGsembly code), and once it terminates, it activates a self-destruction mechanism. If dvgrn finishes his new APGompiler, my prediction is that the resulting pattern will fit in a bounding box below 30k x 30k. Would probably fit into 15k by 15k if I ever learned how to build computers by hand

[AlbertArmStain]

lumi (on the Discord) is building a machine with a lifespan of f_ε0(n) in the fast-growing hierarchy, which is vastly greater than Graham's number.

How is the machine designed and how large is it?

Hi, I am lumi from Discord. The way it will work is it uses a GPC to run a googological notation called the primitive sequence system (for which I have already made APGsembly code), and once it terminates, it activates a self-destruction mechanism. If dvgrn finishes his new APGompiler, my prediction is that the resulting pattern will fit in a bounding box below 30k x 30k. Would probably fit into 15k by 15k if I ever learned how to build computers by hand

It has been reduced in the “Smaller Pi Calculator Challenge” thread. It would be nice if it was reduced even further though.

[dvgrn]

Would probably fit into 15k by 15k if I ever learned how to build computers by hand...

I definitely wouldn't want to discourage anyone to do some experimental construction of those finite-state Life machines manually. What calcyman said on Discord is all perfectly true -- it's tremendously inefficient to build things by hand, instead of just writing and using a script. But on the other hand, Nathaniel and I wouldn't have been able to successfully reverse-engineer the pi calculator and write a chapter about it, without a lot of experimenting with the patterns in Golly; it's a good learning experience to see what breaks the calculator, and what doesn't, and why.

Select and Shift+S in Golly is the main tool that's needed for tightening up these patterns by hand. If it looks like there's a diagonal gap between two components, either in the components stack or in the computer, then the pattern can probably be compressed perfectly well by moving everything diagonally.

In some cases, the new smaller pattern might not run quite as well in HashLife, and it might take more space when saved in .mc format, because the components won't be aligned on a power-of-two grid any more. But if bounding box is all that matters, then compression of these calculator patterns is surprisingly easy.

[gameoflifemaniac]

How is the machine designed and how large is it?

Hi, I am lumi from Discord. The way it will work is it uses a GPC to run a googological notation called the primitive sequence system (for which I have already made APGsembly code), and once it terminates, it activates a self-destruction mechanism. If dvgrn finishes his new APGompiler, my prediction is that the resulting pattern will fit in a bounding box below 30k x 30k. Would probably fit into 15k by 15k if I ever learned how to build computers by hand

It has been reduced in the “Smaller Pi Calculator Challenge” thread. It would be nice if it was reduced even further though.

The GPC is old, unsuited for the new version of APGsembly. There's no compiler for APGsembly 2.0 either. And the components (unary, binary, printer) themselves weren't actually updated in the GPC, because they were compactified only in 2020

[dvgrn]

The GPC is old, unsuited for the new version of APGsembly. There's no compiler for APGsembly 2.0 either. And the components (unary, binary, printer) themselves weren't actually updated in the GPC, because they were compactified only in 2020

Just to set expectations a little bit -- the current APGsembly compiler isn't terribly far away from working with APGsembly 2.0, and in fact it can be made to compile 2.0-ish things with a little bit of hand-editing. The lack of support for #COMPONENTS and #REGISTERS directives is annoying, but not an insurmountable problem.

A working compiler for all of the code snippets printed in the Life textbook -- and, presumably, whatever else anyone wants to throw at it -- is definitely back on my to-do list. It won't happen this weekend, though, because there's a new beta build of Golly coming on Monday, and a pile of pattern changes are due to get checked in.

It will probably take a few weeks at least to get the 2.0 APGompiler done, unless someone else wants to tackle the rewrite.

[gameoflifemaniac]

I made an upgrade to my primitive sequence system code:

Code: Select all

#REGISTERS {"B0": [10, "11000000000"], "U3": 5}
# State    Input    Next state    Actions
# ---------------------------------------
INITIAL; *; STATE_1_INITIAL; NOP
STATE_1_INITIAL; *; STATE_3; READ B0
STATE_3; Z; STATE_4_WHILE_BODY; NOP
STATE_3; NZ; STATE_2; NOP
STATE_4_WHILE_BODY; *; STATE_1_INITIAL; TDEC B0, INC U0
STATE_2; *; STATE_6; TDEC U0
STATE_6; Z; STATE_5; INC U3, NOP
STATE_6; NZ; STATE_7; INC U0, NOP
STATE_7; *; STATE_8; INC U0, NOP
STATE_8; *; STATE_11; READ B0
STATE_11; Z; STATE_12_WHILE_BODY; NOP
STATE_11; NZ; STATE_10; NOP
STATE_12_WHILE_BODY; *; STATE_13; TDEC B0, INC B1
STATE_13; *; STATE_14; TDEC U0
STATE_14; Z; STATE_8; NOP
STATE_14; NZ; STATE_8; INC U1, NOP
STATE_10; *; STATE_15; SET B1, NOP
STATE_15; *; STATE_17; TDEC U0
STATE_17; Z; STATE_16; NOP
STATE_17; NZ; STATE_9; NOP
STATE_16; *; STATE_18; TDEC U1
STATE_18; Z; STATE_8; NOP
STATE_18; NZ; STATE_19_WHILE_BODY; NOP
STATE_19_WHILE_BODY; *; STATE_16; INC U0, NOP
STATE_9; *; STATE_20; INC B1, NOP
STATE_20; *; STATE_22; TDEC U3
STATE_22; Z; STATE_21; NOP
STATE_22; NZ; STATE_23_WHILE_BODY; NOP
STATE_23_WHILE_BODY; *; STATE_24; INC U4, NOP
STATE_24; *; STATE_26; TDEC B1
STATE_26; Z; STATE_25; NOP
STATE_26; NZ; STATE_27_WHILE_BODY; NOP
STATE_27_WHILE_BODY; *; STATE_29; READ B1
STATE_29; Z; STATE_28; NOP
STATE_29; NZ; STATE_30; SET B0, SET B1, NOP
STATE_30; *; STATE_32; TDEC U6
STATE_32; Z; STATE_31; NOP
STATE_32; NZ; STATE_33_WHILE_BODY; NOP
STATE_33_WHILE_BODY; *; STATE_30; INC B0, INC U5, NOP
STATE_31; *; STATE_34; TDEC U5
STATE_34; Z; STATE_28; NOP
STATE_34; NZ; STATE_35_WHILE_BODY; NOP
STATE_35_WHILE_BODY; *; STATE_31; INC U6, NOP
STATE_28; *; STATE_24; INC B0, INC U2, NOP
STATE_25; *; STATE_36; TDEC B0
STATE_36; *; STATE_38; TDEC U2
STATE_38; Z; STATE_37; NOP
STATE_38; NZ; STATE_39_WHILE_BODY; NOP
STATE_39_WHILE_BODY; *; STATE_36; INC B1, NOP
STATE_37; *; STATE_41; TDEC U0
STATE_41; Z; STATE_40; NOP
STATE_41; NZ; STATE_42_WHILE_BODY; NOP
STATE_42_WHILE_BODY; *; STATE_37; INC U5, INC U6, NOP
STATE_40; *; STATE_43; TDEC U5
STATE_43; Z; STATE_20; NOP
STATE_43; NZ; STATE_44_WHILE_BODY; NOP
STATE_44_WHILE_BODY; *; STATE_40; INC U0, NOP
STATE_21; *; STATE_45; INC U4, NOP
STATE_45; *; STATE_47; TDEC U4
STATE_47; Z; STATE_46; NOP
STATE_47; NZ; STATE_48_WHILE_BODY; NOP
STATE_48_WHILE_BODY; *; STATE_45; INC U3, NOP
STATE_46; *; STATE_50; TDEC B1
STATE_50; Z; STATE_49; NOP
STATE_50; NZ; STATE_51_WHILE_BODY; NOP
STATE_51_WHILE_BODY; *; STATE_46; READ B1
STATE_49; *; STATE_53; TDEC U1
STATE_53; Z; STATE_52; NOP
STATE_53; NZ; STATE_49; NOP
STATE_52; *; STATE_55; TDEC U0
STATE_55; Z; STATE_54; NOP
STATE_55; NZ; STATE_52; NOP
STATE_54; *; STATE_57; TDEC U5
STATE_57; Z; STATE_56; NOP
STATE_57; NZ; STATE_54; NOP
STATE_56; *; STATE_58; TDEC U6
STATE_58; Z; STATE_5; NOP
STATE_58; NZ; STATE_56; NOP
STATE_5; *; STATE_59; TDEC B0
STATE_59; Z; STATE_END; NOP
STATE_59; NZ; STATE_1_INITIAL; NOP
STATE_END; *; STATE_END; HALT_OUT

The code now reaches a level of φ(ω,0) in the fast-growing hierarchy. This is a major efficiency improvement for the expected bounding box.

[Sarah]

Demo for the delay I made to make it easier to modify that.

Code: Select all

x = 58, y = 60, rule = B3/S23Super
33.2M$33.2M3$30.2M$30.2M2$33.2M8.M$34.M6.3M4.2M$33.M6.M7.2M$40.2M$35.
M$32.M12.2M$32.M2.M9.2M$31.3M$31.3M14.2M$31.M2.M13.2M7.M$32.M.M20.3M$
32.3M19.M$54.2M$16.M$16.M.M$16.2M$47.M$12.2M8.2M22.M$11.M.M8.2M21.3M.
M$11.M33.2M2.2M$10.2M13.2M18.2M2.2M$25.2M$15.2pA28.2M$15.pA.pA27.M4.M
$16.2pA4.2M6.2S13.M4.M$12.pA9.2M6.2S5.2M7.M.M$11.pA.pA13.S9.2M$12.pA6.
S6.S.S$18.S.S4.S.S12.2M$18.S.S3.S.S13.2M$19.S5.S6.M$6.2S8.S5.S7.3M$5.
S.S7.S.S3.S.S5.M7.2M$4.S2.S.S6.S5.S6.2M6.2M$4.2S2.2S3.S5.S$.O10.S.S3.
S.S$O.O10.S5.S6.2M$.O8.S5.S$9.S.S3.S.S$10.S5.S$7.S5.S10.2M3.2M$6.S.S3.
S.S9.2M3.2M$7.S5.S$10.S15.3M$9.S.S14.3M$10.S16.M6$27.2M$27.2M!

[stheia]

Wow. I never knew that engineered diehards could live that long.

[…]

Code: Select all

Diehard

[…]

Just as an alert for anyone who is watching this, the spaceship crashes into a wall before the would-be diehard would…well, die. It should work better in Golly.

Lurking.
The diehard still dies in Lifeviewer by "all spaceships crash into wall".
I don't know how, but when each spaceship crashes into the LifeViewer wall, it leaves behind no debris. Can someone enlighten me?

[confocaloid]

Lurking.
The diehard still dies in Lifeviewer by "all spaceships crash into wall".
I don't know how, but when each spaceship crashes into the LifeViewer wall, it leaves behind no debris. Can someone enlighten me?

I think the spaceship disappears completely at once at the moment it touches the wall. This seems to agree with the page lazyslug.com/lifeview/plugin/version.txt saying

Build 177
- patterns of any size can now be deleted in one go when they hit the boundary