Modifying geminoid to reach any speed < c/2

For discussion of specific patterns or specific families of patterns, both newly-discovered and well-known.
Post Reply
User avatar
simsim314
Posts: 1823
Joined: February 10th, 2014, 1:27 pm

Modifying geminoid to reach any speed < c/2

Post by simsim314 » April 17th, 2020, 11:24 am

The design below is probably the simplest way to construct a universal speed orthogonal spaceship for all valid speeds i.e. <c/2.

We start from regular Geminoid. In usual Geminoid the snake of a signal makes one loop and reaches a new copy. In our case the snake is done twice. First time for copy construction commands, the second time to convert to MWSS stream and reach the copy constructed which is way far ahead. Our construction order is as follows:

0. First arm commands delete the previous copy.
1. Then we slow salvo construct an UH (Universal Helix). We send a signal as far as we want. Because UH time delay is quadratic.
2. We then build glider to MWSS converter and send signal in form of MWSS after the helix, and code the new copy of the tape reader.
2.1 We can code arm movements with MWSSs.
2.2 We can have an UH with MWSS->glider converter and use the known arm commands.
3. After the commands are sent the tape makes another loop to copy itself. It sends signal in form of MWSS streams.

*Note the top and the bottom tape readers parts are not the same. The 3rd pass of signal is reflected as well as converted to MWSS while the 4th is not reflected but converted to MWSS. The tape is coding both top/bottom parts yet the head and tale of the Geminoid itself are identical.

*Note2 I was mistakenly thinking we need 4 passes but we need only 2. The MWSS signal converted to construction commands and sent to the twin, sent back from the twin and converted back to MWSS. The signal path is:
1. Orthodonal MWSS -> 2. Gliders sent to bottom twin -> 3. Gliders sent to top twin -> 4. Gliders converted back to orthodonal MWSS -> 1. Othogonal MWSS

Here is the schematics:
Attachments
GeminoidSpaceship.png
GeminoidSpaceship.png (9.5 KiB) Viewed 4274 times

User avatar
Macbi
Posts: 903
Joined: March 29th, 2009, 4:58 am

Re: Modifying geminoid to reach any speed < c/2

Post by Macbi » April 18th, 2020, 11:16 am

simsim314 wrote:
April 17th, 2020, 11:24 am
1. Then we slow salvo construct an UH (Universal Helix). We send a signal as far as we want. Because UH time delay is quadratic.
Can you expand on this? What's a universal helix? What do you mean by a quadratic time delay?

User avatar
simsim314
Posts: 1823
Joined: February 10th, 2014, 1:27 pm

Re: Modifying geminoid to reach any speed < c/2

Post by simsim314 » April 18th, 2020, 12:11 pm

Macbi wrote:
April 18th, 2020, 11:16 am
Can you expand on this? What's a universal helix? What do you mean by a quadratic time delay?
Universal helix is a pattern that allows to place still life at any desired distance D with construction cost of O(sqrt(D)) and speed arbitrary close to c/2.

Code: Select all

x = 62, y = 548, rule = B3/S23
46bo$39b3o3b3o$33bo5bo2bo2bob2o$32b3o4bo6b3o$32bob2o3bo3bo2b3o$33b3o3b
o6b2o$33b3o3bo2bo$33b3o3b2o$33b2o10bo$44b3o$44bob2o$38b3o4b3o$45b3o$
45b3o$45b2o$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$
38b3o4$38b3o4$38b3o$49bo$48b3o$48bob2o$38b3o8b3o$49b3o$49b3o$49b2o$38b
3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o18bo$58b3o$58bob2o$
59b3o$38b3o18b2o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o2$27bo$26b3o
$26bob2o8b3o$27b3o$27b2o2$5b3o30b3o$4bo2bo$7bo$3bo3bo$7bo30b3o$4bobo3$
38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$
38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$
38b3o4$38b3o4$38b3o$46bo$45b3o$45bob2o$38b3o5b3o$46b3o$46b3o$46b2o$38b
3o4$38b3o$46bo$45b3o$45bob2o$38b3o5b3o$46b3o$46b3o$46b2o$38b3o4$38b3o
4$38b3o4$38b3o4$38b3o2$2bo$b3o$2obo34b3o$3o$3o$3o$b2o35b3o4$38b3o2$2bo
$b3o$2obo34b3o$3o$3o$3o$b2o35b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$
38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$
38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$
38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o4$38b3o2$38bobo$38b3o$
37bo3bo$37b2ob2o2$37b2ob2o$38bobo$38bobo2$37b2ob2o61$44b3o$44bo2bo$44b
o$44bo3bo$44bo$45bobo19$42bo$41b3o$41bob2o$42b3o$42b2o8$44b3o$44bo2bo$
44bo$44bo3bo$44bo$45bobo19$42bo$41b3o$41bob2o$42b3o$42b2o!
It was found in 2016, in collaboration of myself, Sphenocorona and HartmutHolzwart. You can see more examples and read the thread here.

Any universal speed spaceship will probably use it, as we don't have any other general way to send information with any valid arbitrary speed < c/2.

User avatar
calcyman
Moderator
Posts: 2936
Joined: June 1st, 2009, 4:32 pm

Re: Modifying geminoid to reach any speed < c/2

Post by calcyman » April 18th, 2020, 12:26 pm

Can't you just construct the blinker puffer, then wait a long time, and then ignite the trail? I don't see the reason for having the extra *WSSes which stop the fuse and reignite it later...?

EDIT: for clarification, you launch the blinker puffer followed by the *WSS recipe (which would collide with the block, when it comes into existence, to construct the copy of the Geminoid). Then, whilst that's travelling away at c/2, you construct a bomb (p256 gun attached to a bunch of semisnarks) which, upon exploding, ignites the trail of the blinker puffer.
What do you do with ill crystallographers? Take them to the mono-clinic!

User avatar
Macbi
Posts: 903
Joined: March 29th, 2009, 4:58 am

Re: Modifying geminoid to reach any speed < c/2

Post by Macbi » April 18th, 2020, 1:11 pm

calcyman wrote:
April 18th, 2020, 12:26 pm
EDIT: for clarification, you launch the blinker puffer followed by the *WSS recipe (which would collide with the block, when it comes into existence, to construct the copy of the Geminoid). Then, whilst that's travelling away at c/2, you construct a bomb (p256 gun attached to a bunch of semisnarks) which, upon exploding, ignites the trail of the blinker puffer.
That's what I would have done, but now that I see the universal helix (what's helical about it?) it seems simpler. You only have to construct some LWSSes, MWSSes and HWSSes rather than a timer. Of course you also have to put some space in between the spaceships, it adds O(sqrt(D)) to the recipe rather than the timer's O(log(D)). But the constant factor is much better, and really anything better than O(D) will work.

User avatar
dvgrn
Moderator
Posts: 10669
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Modifying geminoid to reach any speed < c/2

Post by dvgrn » April 18th, 2020, 2:30 pm

Macbi wrote:
April 18th, 2020, 1:11 pm
...now that I see the universal helix (what's helical about it?) it seems simpler.
The original "helix" was the one in the Caterpillar, and was named originally I think by Jason Summers but his original out-of-date document on Caterpillar design doesn't seem to be available. Here's the definition of 'helix' in Gabriel Nivasch's after-the-fact article.

As far as I know, it was called a "helix" because that particular Caterpillar instance looked something like a DNA helix. Other functionally similar streams of spaceships, supporting a particular velocity of continuously or intermittently "burning" glider-producing reaction, have since traditionally been called "helix" even when they don't happen to look much like one.

User avatar
simsim314
Posts: 1823
Joined: February 10th, 2014, 1:27 pm

Re: Modifying geminoid to reach any speed < c/2

Post by simsim314 » April 18th, 2020, 6:50 pm

calcyman wrote:
April 18th, 2020, 12:26 pm
Can't you just construct the blinker puffer, then wait a long time, and then ignite the trail? I don't see the reason for having the extra *WSSes which stop the fuse and reignite it later...?
Helix is better in places where you want to send several helices in a row to move the same SL. I don't think this is the case here - so yes you're right. Just a blinker puffer is enough.
calcyman wrote:
April 18th, 2020, 12:26 pm
Then, whilst that's travelling away at c/2, you construct a bomb (p256 gun attached to a bunch of semisnarks) which, upon exploding, ignites the trail of the blinker puffer.
Yes this is also exponential algorithm (unlike the UH which is quadratic) - it can also trigger a self destruct. Good point!
dvgrn wrote:
April 18th, 2020, 2:30 pm
As far as I know, it was called a "helix" because that particular Caterpillar instance looked something like a DNA helix.
I'm not sure why it's called helix, I had an intuition of something burning and pushing some signal forward and being destructed in the process. As far as I understand this is the features and functionality of any traditional *WSS helix.

User avatar
Macbi
Posts: 903
Joined: March 29th, 2009, 4:58 am

Re: Modifying geminoid to reach any speed < c/2

Post by Macbi » April 18th, 2020, 8:35 pm

How hard would it be to make the ship travel in arbitrary directions as well as with arbitrary velocities? It would be easy to make it so that it alternated between travelling horizontally and travelling vertically, but I can't think of an easy mechanism which could set the timer differently on alternating trips.

User avatar
simsim314
Posts: 1823
Joined: February 10th, 2014, 1:27 pm

Re: Modifying geminoid to reach any speed < c/2

Post by simsim314 » April 18th, 2020, 9:43 pm

Macbi wrote:
April 18th, 2020, 8:35 pm
How hard would it be to make the ship travel in arbitrary directions as well
It's not much harder. One just need another horizontal blinker puffer + MWSS reflector to be encoded in the tape. When the first vertical blinker puffer is converted to SL, the next part of the tape will build the same mechanism with self destruct that will reflect the rest of the signal 90 degrees. The addition for oblique direction is probably making the tape twice longer. But I think it's best to start from simple case and then make a more complex one.

------

So the construction order is as follows.

MWSS are incoming. The generate two output gliders. 1. Is sent to the twin part of Geminoid. 2. Is coding an arm.

1. Build a blinker puffer - local.
2. Build a bomb - local.
3. Build glider->MWSS conduit - local.
4. Build reflector to input into (3) conduit the glider stream returning from the twin.
5. Send recipe in form of MWSS stream to build conduit MWSS->glider pair.

When the bomb explodes it triggers the blinker puffer, which in return builds a new conduit using MWSS coding an arm. Another option would be to attach to the blinker puffer "*WSS dried" MWSS->glider converter. When the puffer converts to glider it automatically triggers the *WSS dried MWSS->glider, and rest of the signal is coded in gliders.

Roger
Posts: 8
Joined: January 1st, 2021, 9:54 pm

Re: Modifying geminoid to reach any speed < c/2

Post by Roger » January 6th, 2021, 7:23 am

calcyman wrote:
April 18th, 2020, 12:26 pm
For clarification, you launch the blinker puffer followed by the *WSS recipe (which would collide with the block, when it comes into existence, to construct the copy of the Geminoid). Then, whilst that's travelling away at c/2, you construct a bomb (p256 gun attached to a bunch of semisnarks) which, upon exploding, ignites the trail of the blinker puffer.
Why bother building a time bomb? Just encode the delay until you light the blinker fuse into a long gap before the trailing part of recipe -- when that arrives, it lights the fuse and then (assuming the construction *WSS slow slavo is all finished by then) starts the deconstruction (for example, via a seeds-of-destruction-style self-destruct mechanism). So that's complexity O(1), rather than O(sqrt(D)) or O(log(D)) (though it requires a gap of length O(D), or actually ~D/3)

User avatar
dvgrn
Moderator
Posts: 10669
Joined: May 17th, 2009, 11:00 pm
Location: Madison, WI
Contact:

Re: Modifying geminoid to reach any speed < c/2

Post by dvgrn » January 6th, 2021, 7:54 am

Roger wrote:
January 6th, 2021, 7:23 am
Why bother building a time bomb? Just encode the delay until you light the blinker fuse into a long gap before the trailing part of recipe -- when that arrives, it lights the fuse and then (assuming the construction *WSS slow salvo is all finished by then) starts the deconstruction (for example, via a seeds-of-destruction-style self-destruct mechanism). So that's complexity O(1), rather than O(sqrt(D)) or O(log(D)) (though it requires a gap of length O(D), or actually ~D/3)
It might depend on whether you're trying to minimize the bounding box or not. But also there are some issues with storing the recipe, I think.

With a two-constructor Geminoid design, you can put the two halves as far apart as you want, and fit your length-O(D/3) gap in there. A longer gap will produce a wider but still overall faster spaceship. (Right?) But we might want to try a design with just a single universal constructor, which reaches ahead to build a memory loop for its entire recipe, and sends a second copy of the recipe as data.

Then the memory loop would emit two copies of the recipe and then self-destruct. In that scenario, I think it might work better to use the time-bomb method than the ridiculously-long-gap method to reach speeds very near c/2, because the size of the memory loop wouldn't have to grow nearly as fast.

Post Reply