Hi Everyone-

I found Twisty because I was looking for someone to help me solve a rubik-like puzzle I wrote called Triplex. I am not a cuber, so even though I wrote it, I couldn't solve it.

Well, it has a solution. In the week it took Twisty to check me out, the forum at speedsolving solved it. In fact, I think they solved it in less than four hours!

I am still working on Triplex. But, it you want to see it, it can be found at www.kidpaint.net

It does not have a leader board yet (I'm working on that this weekend), but it's definitely playable. AND, it's not a push-over. I STILL have problems solving it.... even after being shown how.

My best time so far is over twelve minutes.

- george

Nice!

Also, the hexagonal non-virtual equivalent to this mechanism is the Trio puzzle.




Pantazis

George, What solution have you used for your solves? Was it similar to Jaap's solution for the Turnstile, or maybe intuition?

Hi George,

Nice little puzzle, I like the finish .
I would put a left click for an anticlockwise turn and a right click for clockwise. It would match other simulators better, the type of clicks drove me mad (I didn't have far to go though).
Cheers,
Burgo.

I like this puzzle, and the solution is clever.



I agree, I am used to Gelatinbrain's applet, and then this was very confusing. I spent a few seconds trying to work out whether to hold shift with my click, and then got it wrong anyway .

My fourth solve I managed to get 1minute and 7 seconds. If I can work out this clicking I might be able to get it a bit faster.

Is it likely we will be seeing any other versions, with different colour schemes, number of circles or anything?

Nice work!

-Mark-

I liked that puzzle alot! I didnt completely solve it but I was really close.
Quite intuitive and fun! thanks

Goetz-

I don't have a solution. My approach is to solve RED, Magenta, White and Yellow then move it out of the way in order to solve blue and green as if it were turnstile. But, mostly it's just dumb luck.

I don't think turnstile is the same as the blue green of Triplex, because I can't seem to follow Jaap's turnstile solution for Triplex. Either they're not the same, or I'm too dense.

However, don't follow my lead. My best time is pathetic. I think I'll stick to coding.

- george

Pantazis-

Is Trio == Triplex??? Are they the same puzzle?

- george

Hi George, yes actually they are! (for some reason I thought the triplex
was based on a pentagon. Both Trio and Triplex are based on hexagons).

Only the color configuration differs, which does not change too many
parameters when solving it.




Pantazis

Uhallgeo,

Very nice puzzle. Congratulations on some nice ActionScript coding.

My intuitive solution is not very elegant or efficient, but I see this as "fourteen minutes well wasted".

Skarabajo

Skarabajo, you just spoiled the ending for everybody .
So I will spoil the middle
George,
Highlight this spoiler text with your mouse:
>[ I solved all the edges first using (R'LRL'),
then saved the central Y corners for last, all corners can then be placed with Jaap's (R'LRL')x3 = 2+2 corner swap.
Now obviously with so many same coloured pieces some of the larger outer areas can be solved first with intuition.
For the last central Y, you just have to change the puzzle orientation as required.
If you end up with the situation where white needs to be swapped with the centre only:
Make the white corner between L & R, call [(R'LRL')x3]=x and perform: U x 2U' x 2U x U' ]
This might not be a super fast method but it might help make it simple for you.
Cheers,
Burgo.

Hi George, thank you for sharing this nice puzzle.
Welcome to our little community!

This is how I have solved it. (Please highlight the hidden spoiler below, if you want to read it.):
[

1. I solve all blue, green and light blue pieces (everything but the U circle) in the L and R circle intuitively. This not very hard.
2. Swap two pairs of triangles until all triangles are correct
I use a certain sequence of turns to swap two pairs of triangles (called "corners" by Burgo because they remind us of corners of a 3D puzzle.
Notation L and R are clockwise turns of the L and R circles. L´and R´ are anticlockwise turns.
Diagram 2. shows three differently grey pieces: very dark grey = unchanged by the following sequence, they may be correct or not; middle grey: triangles to be swapped; light grey = rectangles I do not care yet ("edges" in Burgos text)
I use a sequence of 6 moves R´ U´ L´ U L R to do a 2-2 swap at a time.
3. I permute 3 rectangles as shown in diagram 3.
It is a clockwise 3-cycle. All the dark grey pieces remain unchanged after the sequence:
R´ U´ L´ U L R R´ U´ L´ U L R
You will recognize that it repeats the sequence from 2. twice.
We write this as (R´ U´ L´ U L R)x2 (The sequence in brackets is performed twice)
You can make it a 10 move sequence recognizing that the R at the end of the first and the R´ at the begiining of second performing result in a noop.]
You see the win win situation of being a member here.
You share your puzzle with us and we share our solutions - if you want.
This is the most valuable thing about the twistypuzzle forum: sharing.

It's not so hard to solve the red blue green part but the rest white purple yellow are confusing me. Strange. It wouldn't be so hard? Just like a cube with U, L, R face turning should be.......Still working on it. Oh yeah I don't want to see the solutions yet

Ha this is nice. The crazy animation when you solve was fun too. I needed 11 minutes to solve it. I tried to cycle corners like on a 3x3x3 but couldn't do it. edges worked though. I did it kind of intuitively but wasn't really on top of things. Konrad's solution is great. I haven't even highlighted the spoiler but it's based on algos that will always work so I bet it's speedsolve capable.

Everyone-

THANKS!!! What a great bunch of posts.

It still amazes me that you guys solved this little puzzle so easily. I only wish I had completed the 'High Score' module this weekend. It seems I'm not a speedy coder OR a speedy puzzle solver.

Looking at your solutions, I have decided to put the 'SOLVE' button on hold. Programming a solution presents a tough problem. After the 'High Score' module I think I'll add fluff (more animations and a character that will give hints).

But, I gotta tell you guys.... this is fun. I think I'll also add a two circle (Turnstile) and a four circle version.

- george

Thanks Konrad, for sharing your method too.
I kinda forgot to say that I worked my way around one colour at a time also.
I only solved it once (because the clicking format was frustrating for me, I really hope George might amend that for us , or give an option to change settings). I have actually never solved the trio type 2D puzzles until this one, so it was fun for me.
Another suggestion might be to have a click and drag `rotate function` for the entire puzzle (where the little `solved state diagram` rotates with it). This would be good for beginners to visualise it (and quirky).

I will definately return to have a few more solves, especially when he adds other puzzles. My kids will probably like your colouring pages too.
And my 5yo will like the puzzle too .

Cheers,
Burgo.

Very cool, i wrote something similar in c++ for my freshman game at school. it wasnt fun because i wasnt good enough to code complex puzzles but it had a similar mechanic. And i based it off of a very similar puzzle i saw here!

Everyone...

I sorta crawled into a hole to write the Leader Board for Triplex. I uploaded it today. So, now if you solve Triplex your time is recorded and displayed....if it's in the top ten.

- george

Everyone-

For those who solved Triplex. I've posted the next version (with one more wheel) called Quadplex. It's posted on www.kidpaint.net

Good Luck!

- george