Lightspeed wire

[idonteven]

So, while clicking around on Lifewiki, I found a page called "Lightspeed wire". Then I began to read, and something caught my eye:
"A lightspeed wire would be a wick which burns nondestructively at the speed of light..."
Would be? I found it to be odd, articles like this don't usually word things like that.
Then I made the pattern in the image, and... it worked out as well as you would expect :P
So, i took it as a challenge to make it stable, and after some experimentation, I made some progress!

Code: Select all

#CXRLE Pos=-39,-6
x = 74, y = 11, rule = Life
2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b
2o$2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b
2o2b2o2$2b70o$bobo68bo$obo4b65obo$bo2bo67bo$2b70o2$2b2o2b2o2b2o2b2o2b
2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o$2b2o2b2o2b2o2b
2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o2b2o!

I mean, it explodes when it reaches the end, but it's something! :)

[dvgrn]

So, while clicking around on Lifewiki, I found a page called "Lightspeed wire". Then I began to read, and something caught my eye:
"A lightspeed wire would be a wick which burns nondestructively at the speed of light..."
Would be? I found it to be odd, articles like this don't usually word things like that.

Yes, the wording is a little odd and should probably be updated. The example is copied from the Life Lexicon entry, which hasn't been updated since Jason Summers found a working example of a lightspeed wire, in 2003 -- a chain of beehives rather than a single connected still life. It's complicated, though: you have to burn the wick ten times, or once in each direction, before it recovers completely -- and some of the ends get burned off and have to be rebuilt.

The latest version of this structure is Jormungant's high-bandwidth telegraph. See also the "Lightspeed signals in Life" link from the lightspeed-wire wiki article.

[idonteven]

While messing around with the signal, I made this happen:

Code: Select all

#CXRLE Pos=0,0
x = 63, y = 13, rule = B3/S23
bo2bo2bo2bo2bo2bo2bo2bo2bo2bo2bo2bo2bo2bo2bo2bo2bo2bo2bo2bo2bo$b61o2$b
61o$o2bo58bo$b2o4b55o$4bo53bo$b55o4b2o$o58bo2bo$b61o2$b61o$bo2bo2bo2bo
2bo2bo2bo2bo2bo2bo2bo2bo2bo2bo2bo2bo2bo2bo2bo2bo2bo!

Just thought i'd say that. :)

[Shaos]

Cool! Any progress with that?

[Hunting]

Cool! Any progress with that?

It's not too good to necropost like this.

Plus, there are thousands of stabilizations for that wire...

P. S. As far as I know, there's no known LW->X or X->LW converters. People are not looking for it that hard though... But who will want to use it when you have the beehive wire?

[dvgrn]

As far as I know, there's no known LW->X or X->LW converters. People are not looking for it that hard though... But who will want to use it when you have the beehive wire?

Well, the beehive wire is a horrible kludgey way to send lightspeed signals. The beehives shift sideways and forward/back after every use, so you have to send ten signals to get back to where you started (and rebuild long chunks of wire that get used up after every cycle). A "real" lightspeed wire like this that recovers properly after every use would be great to have.

Unfortunately there doesn't seem to be a good approach for finding stable inputs or outputs for these zebra-stripes wires. Dean Hickerson's 'drifter' program from two decades ago can still be run, on somewhat larger areas of unknown cells than was possible in the 1990s. But recent runs have not turned up any new interfaces. Possibly a modern effort to parallelize and distribute a search among thousands of CPUs might turn something up.

Searches like this tend to get hundreds of times more CPU-intensive every time you add a few cells, though, so unless there happens to be a converter fairly close to the search space that we've been able to reach already, we might still not be able to find anything. There aren't a lot of "close calls" with converters for these orthogonal lightspeed signals, so it doesn't look too hopeful that a solution is within reach. Never know until you try, though!

[Shaos]

Possibly a modern effort to parallelize and distribute a search among thousands of CPUs might turn something up.

I though about something distributed as well

P.S. Look here to continue discussion: viewtopic.php?f=2&t=4430