Sierpinski Network

[qqwref]

Here's a puzzle from Meep's site:
http://meep.cubing.net/html5/sierpinski.html
There is also a higher-order version for crazy people:
http://meep.cubing.net/html5/bigsierpinski.html

In this topic we can discuss methods, times, tricks, and so on.

I think I have something vaguely resembling a method now. I got a very lucky 9.43 single on level 3 (the highest level on the main puzzle) and a 199.83 first solve on level 6.

EDIT: 6.73 lololsingle on level 3

 

[Ickenicke]

Funny puzzlez. But it is hard to find a good method!

 

[ben1996123]

First attempt at level 6: 17:28.70.

 

[Ickenicke]

How do I get a totally white puzzle to black at the easiest puzzle? Fast?

 

[y235]

click on every point one time

 

[aronpm]

How do I get a totally white puzzle to black at the easiest puzzle? Fast?

Click every node

 

[mr. giggums]

Level 3
Lolsingle: 5.10
NLsingle: 5.60
Could easily be fast if I were more acurate.

EDIT: 2.17 lucky 4 clicks
EDIT2: It was Level 2 not Level 3.

 

[Ickenicke]

click on every point one time

Thank you, I found out that by myself

 

[y235]

Interesr Fact: If solving the Sierpinski Network optimally, you don't have any node pressed twice, and the order of the nodes that you press doesn't matter.
Another Fact: A Sierpinski Network of any order can be fully changed from white to black by pressing all the nodes one time.

 

[Ezy Ryder]

Just got three 0.03 singles on the basic level. It's fun.

 

[emolover]

293.73 level six. It is so unresponsive.

 

[Stefan]

I wish the background was grey or yellow or whatever, would make white nodes easier to see.

 

[mrpotatoman14]

3.2 average on lvl.2 I'm really slow at clicking accurately on the nodes.

 

[Meep]

My best on level 3 was 11.00 without having some really easy solve (10 clicks). I got like 500s on level 6 but I was half paying attention lol. I made a general grid-based light's out network generating thing where you draw the puzzle in a text file but it's an exe (Windows only ): ).

http://meep.cubing.net/html5/lightsoutnetwork.zip

Other puzzles I made that may be of interest:
http://meep.cubing.net/html5/meepuilibrium.html
http://meep.cubing.net/html5/separ4te.html

 

[qqwref]

Level 3
Lolsingle: 5.10
NLsingle: 5.60
Could easily be fast if I were more acurate.

EDIT: 2.17 lucky 4 clicks

4 clicks on level 3 wtf o_0
And your other times are ridiculous too, what's your method?

Also: 174.73 level 6. Oddly enough I can't get anywhere near 1/3 of this time on level 5.
And 5.80 lucky level 3, plus some sub20s (lol) with parity.

 

[Carrot]

EDIT: I'm lol..

Level 2:
1.30 single

Level 3:
21.53

 

[emolover]

4 clicks on level 3 wtf o_0
And your other times are ridiculous too, what's your method?

Also: 174.73 level 6. Oddly enough I can't get anywhere near 1/3 of this time on level 5.
And 5.80 lucky level 3, plus some sub20s (lol) with parity.

Can you tell me your parity? Mine is stupidly long.

 

[Itchy Cacti]

:Lololol .43 2-click solve
Edit: 1-click solve-.5:fp
Edit 2: 1-click .03

 

[qqwref]

Here it is:

I've discovered a more 'mathematical' explanation of parity. Look at the vertices that form a corner of one of the smallest (6-node) triangles; we'll call these corners. It's easy to see how to toggle the colors of two corners. But, if you click them all, you get a pattern that looks locally solvable but actually affects an odd number of corner colors. So it'd be hard to actually predict as you go. I'm pretty sure that every parity algorithm must press all of the corners and then use the other pieces to fix the changes that were made. In fact, if you look at the trapezoidy things on mine, the two non-corner presses fix two corners, so it's easy to see how it works.

PS:
7.60, (21.67[p]), 8.57, (6.53), 7.70 => 7.96
8.93, 20.37[p], 9.50, 7.90, 11.23, 9.90, (21.83[p]), 7.60, 21.67[p], 8.57, (6.53), 7.70 => 11.34

 

[Meep]

Also: 174.73 level 6. Oddly enough I can't get anywhere near 1/3 of this time on level 5.
And 5.80 lucky level 3, plus some sub20s (lol) with parity.

I think it's the way I made the timer. If your computer's performance slows down unfortunately the timer would too (Which is why I capped it at n=6, cause my work computer starts to slow down at n=7 though my home computer does at n=6). It might be best stackmatting the higher order ones.

Also yeah, what's your parity fix? I always just restart if I get parity 'cause my fix is stupidly long. Also, is there a way to determine parity at the start? I was looking for patterns as to whether I should start at a certain corner etc.

Edit: Ninja'd on parity