Gelatin Brian 1.3.5 (tuttminx) permutation problem/help request

[ladyolivier]

Gelatin Brain 1.3.5 (tuttminx) permutation problem/help request

So I'm working on 1.3.5 in gelatin brain, which is more or less a tuttminx. It is, however, stickered in such a way that is has permuation problems the tuttminx doesn't. The first solve I did was a face by face approach which went bust on the last three faces. I decided to tackle the most difficult part first (the permuted triangles of doom) and THEN move face by for my second attempt, a more layer-by-layer approach. I've run into the same issue both times: this puzzle, unlike the tuttminx, has 2 major permutation problems. The centers all have three colors each, and there are narrow 2 colored triangles which must line up, and must be attached to the correct sides of each center. Somehow, don't ask me how, I've gotten them all in the correct places (don't know if I could do it again!) but the last three faces (Again with the last three faces) all the 2 color triangles are upside down. So 2 questions: Am I solving this the right way, and does anyone have any algorithms for flipping the triangles without ruining everything else? (please remember when typing algorithms that I'm still quite new to this and some of y'all's [totally just revealed my heritage with that word] notation looks like greek to me. I may require further explaining to XD)

Photo with the offending triangles of doom circled:

 

[Berd]

I can see another flipped set of triangles in the top of the first image. I think...

 

[ladyolivier]

Oh, there's many more. I just didn't circle them all.

 

[qqwref]

If you move one of those purple edges around the purple face in either direction (this will take 5 moves) you will see that it goes back to its original place, but is flipped. Using that knowledge, you can flip pieces two at a time - all you have to do is bring one piece all the way around a face, then replace it and bring the second piece all the way back. There are several ways to do that so it's best to just try to figure out a move sequence yourself.

Oh yeah, and the concept of moving a piece all the way around a face to flip it shows up in a lot of dodecahedral/icosahedral puzzles, so don't think it's just a Tuttminx thing

 

[ladyolivier]

Yeah, I got the take-it-all-the-way-around thing. But it ends up making a bigger mess - fixes one thing, and makes a massive mess of the rest. I had an algorithm a second ago (which of course my brain will not remember) which was flipping things well, but I ended up chasing the last remaining flipped piece around the puzzle - I I was flipping 2 at once without wanting or meaning to.

 

[cashis]

Title says brian

 

[qqwref]

Yeah, I got the take-it-all-the-way-around thing. But it ends up making a bigger mess - fixes one thing, and makes a massive mess of the rest.

The solution to this is almost always to use a commutator or conjugate. Basically, you would do a move sequence X that messes stuff up; then do a single move or short sequence Y that only affects a few of the pieces X affected; then you undo X; and then you undo Y if necessary. By undoing X, you get to fix the mess you made initially, and the effect of Y is to give you control over the pieces that are affected by both X and Y.

An example on a 3x3x3 would be (R U R' U' R U R') D (R U' R' U R U' R') D'. The first sequence in parentheses messes up the top two layers, but by doing the inverse later, you undo all of that.

 

[ladyolivier]

Oh thank you so much, omg. It's an easy typo.