ConwayLife.com

Hi All, On my first post to this thread I mentioned that I tried to start the diehard by sending a glider over the top and that I couldn't get it to work. I found a way to make that work: x = 112, y = 89, rule = B3/S23 2o19b8o20b12o4b5o2bo15b2o15bo$b2o19bo4bo22bo3b2o8b2obob4obo13b2o8b2o5b o$o43b2o4...

Compacted 112*89 diehard: x = 112, y = 89, rule = LifeSuper 21.8Q20.12Q4.5Q2.Q15.2M15.Q$22.Q4.Q22.Q3.2Q8.2Q.Q.4Q.Q13.2M8.2Q5.Q$ 44.2M4.Q4.2Q9.4Q27.Q2.Q4.Q$45.M4.2Q46.2Q8.2Q$45.M.M8.Q3.M24.2M21.Q.Q$ 46.2M3.Q.2Q.Q2.M25.2M22.Q$51.2Q6.3M3.2M$40.3O8.2Q12.2M21.2M8.M$31.2Q 6.3O46.2M6.3M4.2M$31.2Q13.pA9.3M...

EDIT: One more first sawtooth pull before the first fuse is lit: There was a single-cell bug in one of the blobs of space dust which caused the final shutdown to fail. Here's the corrected version, colourised: x = 112, y = 89, rule = LifeSuper 21.8Q20.12Q4.5Q2.Q15.2M15.Q$22.Q4.Q22.Q3.2Q8.2Q.Q.4Q.Q1...

EDIT: One more first sawtooth pull before the first fuse is lit: Very nice! I've been looking at this area here, and noticed that if an extra eater head is attached to one of the block destroyers, then it produces a mild spark to the west: x = 60, y = 53, rule = LifeHistory 45.2A$45.2A$15.2A.2A38.2...

No, EvinZL's does not work. Mine and b3s23love's do, though (I checked). Right, but they have the same height of the second tower, don't they? In terms of improving the height of the second tower, we can use a fuse to extract a one-time signal into the currently unused northwest corner, and then we...

however, this does I'm not entirely convinced -- if you compare your version (left) to Tim Coe's version (right), the crabstretcher and fuse both start out translated southwest by (2,2), which means that it takes 8 extra generations for a glider to ignite the fuse. But the crab requires the approac...

b3s23love: And I was incorrect about your contribution. The 507 is the number of loaves and you increased their number by 10, so you increased a 497 in the equation to 507. This also means that I was incorrect in my estimate of the second fuse loaves with my inclusion of 2 toads. It should be 121^(...

Hi All, I found a way to ignite a fuse with a pull. The configurations of still lives that do the lighting are actually fairly common. The difficult parts in getting it to work were working around the unmovable eater (required for C/5 spaceship destruction) and getting the phase right so the burn t...

i'm back and i have a signature I've given you a 24-hour temporary ban for causing considerable clutter through repeated violations of rules 3a and 4 of the forum rules: https://conwaylife.com/forums/viewtopic.php?f=2&t=2092 Please can you familiarise yourself with the forum rules, notability guide...

It seems to me that if an engineered tetrational diehard has lifespan t ticks, then the bounding box area of history/envelope has O( t ^2) cells since most of the time there is a spaceship sailing diagonally. Is a tetrational diehard design with envelope strictly below O( t ) cells possible? Yes, y...

So what are the remaining tasks for the optimalised RCT15? Destruction of the DBCA is something that hasn't been investigated much yet. Here's a method of cleanly deactivating the p256 guns by asynchronously deleting 5 blocks: x = 3002, y = 2579, rule = B3/S23 1924b2o$1924bo2bo$1910bo$1910b3o11bo2b...

Yes, it works: And now together ... This is awesome! For those who haven't downloaded and run the pattern in Golly yet, you totally should do: it's a complete construction (from the original arm) of the DBCA, including the meteor-shower activation and self-modification from a monochromatic arm into...

probably OK, just the collision timing with the block was wrong. ... maybe we have to shift the starting block position to achieve a cheaper recipe ... oh no that could solve phase off by 4, but not off by 7. No, moving the starting block doesn't solve any phase issues: it might change the phases o...

Hippo.69 wrote: ↑

August 6th, 2023, 3:59 am

The checkeboard parity ... you mean I built the DBCA in "the other color"? So all lines should be offset to the other parity

Yes, exactly: the glider lanes need to be shifted by an odd number, and the glider phases (mod also need to be shifted.

Here we are: 371528 bits to form the salvo ... time to prepare the starting block and final (no)stack configuration. (unzip; > test.rle; set hashLife step 2^20 ...) Or this monster atachment ... And to test the salvo, put x = 7, y = 13, rule = B3/S23 2$2b2o$2b2o4$2b3o$2bo$3bo! at starting glider. A...

Could you please post the RLE of all of the well-separated slow-salvo gliders emitted by your FixedBottom compiler? That way, I can use my tools to find where it first departs from the correct construction. Okay, so there are exactly 2 mistakes out of 4444 (so your construction is almost perfect --...

x = 119537665, y = 120061953, rule = B3/S23 Here's the comparison RLE showing your construction (red) versus the correct construction (green). Hopefully this helps with debugging: x = 568109, y = 568770, rule = DoubleB3S23 2C$2C62$59.2C$58.2C$60.C125$176.2A$175.2A2B$176.BC$178.B126$309.A$ 308.ACB$3...

I am using ... 0 [?] (g0%1)[366] (g0%1)[376] (g0%1)[373] (g1%2)[373] (g0%2)[372] (g1%2)[368] (g0%2)[363] (g0%1)[362] (g0%1)[351] (g0%1)[348](g0%1)[353] (g0%1)[346] (g0%1)[344] (g0%1)[339](g0%1)[340](g0%1)[343](g0%1)[350] (g0%1)[372] (g0%1)[378] (g0%1)[393] (g1%2)[392] (g0%2)[400](g0%2)[401](g1%2)[4...

I should check if it is agnosticised input problem or the emission failed ... Hmmm, the order of gliders at this time differs from the original salvo ... did I make simillar manual edit problem as with the WHITE portion ?!? Which salvo are you using? The 'corrected levels.txt' in this post definite...

I've replaced Alan Hensel's decimal counter seed in the repo with the spacefiller seed, and run the semilator script to completion. Here's the pattern right before the exciting stuff happens (runnable in LifeViewer, but faster in Golly). #C [[ MAXGRIDSIZE 14 ]] x = 8609, y = 5562, rule = B3/S23 3252...

AlbertArmStain wrote: ↑

July 25th, 2023, 11:47 am

What type of abstract machine is the UCC? I imagine it's a random-access stored-program machine

It's effectively a mixture of a multitape Turing machine (except with one-sided tapes) and a counter machine. If you have the SQ unit then it combines it with the functionality of a 2D Turing machine.

However, some reworking of the southern blob (with help from kissat) gives another 10 factors of 121: Now increased by another 34 factors of 121 to 1.1038 * 10^1046, or: int('306'+'377400'*162+'38'+'78'*13+'56',11) Here's the diehard: x = 109, y = 91, rule = LifeSuper 18.8Q20.12Q4.5Q2.Q15.2M15.Q$19...

Alright, I give up. Loaf arithmetic completely stumps me. I can't even figure out that repeated 78. Anyway, another improvement thanks to calcyman's suggestion: Unfortunately that's not an improvement because the delay is increased after the first fuse has been triggered, rather than before. Howeve...

EDIT: Another improvement: I think the number of loaves is now 22662402577642174435947431184969555898745678704977348354995743290524822022958475465270887245043507354335465420518734145139890075276898963492965858467297318244534656657984050264469288282685909876678155404742349527299630131197888721314978...

calcyman wrote: ↑

May 16th, 2023, 4:19 pm

calcyman wrote: ↑

May 11th, 2023, 5:43 pm

At 22620 cells, this should be constructible using around 50 000 slow gliders, or 400 000 ECCA bits.

It looks like it's around 52 000 gliders.

With pslmake, it's now 40 545 gliders, which will translate to around 300 000 ECCA bits.