Online since 2002. Over 3300 puzzles, 2600 worldwide members, and 270,000 messages.

TwistyPuzzles.com Forum
 It is currently Thu Dec 05, 2013 7:36 am

 All times are UTC - 5 hours

 Page 2 of 2 [ 78 posts ] Go to page Previous  1, 2
 Print view Previous topic | Next topic
Author Message
 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Mon Dec 10, 2012 2:03 pm

Joined: Thu Dec 31, 2009 8:54 pm
Location: Bay Area, California
bmenrigh wrote:
Hi Konrad, a 4-cycle isn't possible (odd permutation). A 5-cycle would be possible.
I was mislead by this remark
rline wrote:
...I found a pure triangle sequence which is 14 moves long. It's a [4,3] commutator. Only issue is that it cycles triangles across 4 faces, which is not ideal when really 3 is the most I want. Hopefully I can alter it.
@rline, what did you mean by "across four faces"?
I missed that. I assume he meant 3 faces and wants 2 but I'd like clarification.

With regard to cycling 3 triangles on 2 faces, it's trivial to prove that on this puzzle that can't be done with a minimal-length commutator. Setup moves (nested conjugates sequence) must be added.

bmenrigh wrote:
I reverse-engineered rline's [4,3] sequence from his picture. It is
GB: [ACD', AEF, ACD, AEF', ABC, DCG, ABC', AEF, ACD', AEF', ACD, ABC, DCG', ABC']
KVTD: [FR', FL, FR, FL', UFR, FDR, UFR', FL, FR', FL', FR, UFR, FDR', UFR']
How on earth can you do this reverse-engineering? You are just looking at rline's picture and you can tell what he did? Wow!
"reverse-engineered" is probably a bit strong because it implies some sort of sophisticated thought process. From his picture there can only be one possibility and it doesn't really take turning the puzzle to see it. I'll try to make a video this evening explaining the logical process for determining the sequence since the myriad of possibilities are reduced to one and explaining how should be illuminating.

_________________
Prior to using my real name I posted under the account named bmenrigh.

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Mon Dec 10, 2012 3:49 pm

Joined: Fri Feb 18, 2011 5:49 pm
Location: New Jersey
I don't have this puzzle yet (should within a month or so), and I haven't actually solved 1.2.11 yet either (though I have solved some related puzzles)... but here's my input for solving this puzzle, which probably overlaps completely with algorithms/strategies already posted by everyone else... I'm sure the algs I have are basically identical to Brandon's, so this is just my outline.

Solving the physical puzzle:
1. Solve as much as is comfortable with intuitive block-building (varies depending on the mood, but can do more than half the puzzle's pieces this way)
2. Edges [1,1]
3. X-centers (petals) [1,1]
4. Centers [3,1]
5. +centers (triangles) [5,1]

Solving for fewest moves on GB, before using the [3,1] centers alg to solve the centers, first use it as a paired double-3-cycle of the +centers to solve about half of the remaining +centers before using it for the centers. Finish the remaining ~15 +centers like normal with the [5,1]. This is similar to the fewest-moves strategy for 1.1.5.

Depending on how easy it is to efficiently solve the +centers with the paired-3-cycle, there is the option to fully solve them with this, and then solve the last few centers last with a pure [7,1] 3-cycle: [BAF',CAB,DCG',DGK',EKJ,DGK,DCG,CAB',BAF,CAB,CGD',DGK',EKJ',DGK,DCG,CAB']
Using a 16-move alg for them isn't that bad because the worst case only requires using this alg 3 times. I would have to actually work my way through a solve to know which alternative is better for move-count. I suspect that the setup-moves required to fully solve the +centers this way would be overly cumbersome and nullify any possible benefits, similarly to the case with 1.1.5, but maybe not since this is a somewhat shallower puzzle...

Can't wait until I have the actual puzzle in my hands =)

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Mon Dec 10, 2012 5:41 pm

Joined: Mon Feb 28, 2011 4:54 am

Quote:
How on earth can you do this reverse-engineering? You are just looking at rline's picture and you can tell what he did? Wow!

Yes i was pretty impressed by this as well!

Quote:
Hi rline, your 14 move sequence is a 4-cycle, right?

No, it's a 3-cycle, like I said.

Quote:
Have you ever looked at Brandon's [5,1] (or more detailed [[1:[1:1]], 1]):

No, deliberately. I wanted to find my own solution without being influenced by anyone else.

Quote:
I find Brandon's sequence easy to memorize and shortest should be short enough.

I haven't tried his sequence, I'm sure it's easy to memorise. But I'm sure you'll agree also that if you find a sequence yourself, which you like, and which you find easy to memorise (which this one certainly is), then an extra couple of moves is irrelevant.

Quote:
Notice that the X Y X' Y' in rline's construction works not because it is a commutator but because the first three moves are a conjugate. The last AEF' / FL' move has nothing to do with the ABC, DCG, ABC' / UFR, FDR, UFR' portion and can be left off.

Thanks for explaining this, Brandon. I still don't follow the GB notation (and don't really want to, either...brain too full) but I should try it without those extra moves.

Quote:
@rline, what did you mean by "across four faces"?

Yeah, sorry. I meant that once the bottom half of the puzzle is done, then if I look at the top face, then around that face, there are 5 other faces. And going from left to right, there are at most 3 unsolved faces in a row. My original 3-cycle where I mentioned the across 4 faces involved a 4th solved face.

The cycle i came up with later, and which I showed in the picture (the one Brandon reverse-engineered) was a slightly different cycle which only used 3 faces. I was much happier with this.

Quote:
I missed that. I assume he meant 3 faces and wants 2 but I'd like clarification.

Sorry gents. To clarify: my original "I found a 3-cycle across 4 faces" was different to the final version I found, shown in the picture and which I used to solve the puzzle.

Quote:
Would you like to share your [4,3]?

Sure. I'll share it as I would have shared it, not using GB notation, but using your (Konrad) proposed notation in the thread above.

(UL' UR UL UR') (UFL L UFL') (UR UL' UR' UL) (UFL L' UFL)

Brandon has said that a couple of moves are irrelevant. Fair enough. I found it by doing an edge piece series, and then the 3 moves in the middle are moves to isolate the triangle and replace it with a new one. You can see why I like it: it cycles triangles only on the unsolved faces, and of course it has the standard edge piece series as its backbone (even if the last move is irrelevant).

Having solved the puzzle, I can't ever imagine solving it by doing all edges first, then all petals, then all triangles. I think it would be way too annoying. I mean, it's doable, but it's far far more interesting building it up.

Many people have expressed how they can't wait to get the puzzle. I can confirm it's an absolutely beautiful puzzle, it's a wonderful solving experience. But there are still issues with catching, popping, etc. I don't want to take anything away from mf8 as I think it's fantastic they've released it. I would just love to be able to deal with these turning issues.

_________________

Blogger Twisty Puzzling - Simple Solutions for Puzzling Twisties

Tutorials: Hexaminx Blockbuilding Tutorial | Curvy Starminx

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Mon Dec 10, 2012 9:39 pm

Joined: Thu Dec 31, 2009 8:54 pm
Location: Bay Area, California
rline wrote:

Quote:
How on earth can you do this reverse-engineering? You are just looking at rline's picture and you can tell what he did? Wow!

Yes i was pretty impressed by this as well!
I created a video explaining this process. My apologies for the poor audio and video quality. YouTube always sucks at transcoding my videos because they use Theora + Vorbis in the OGV container.

_________________
Prior to using my real name I posted under the account named bmenrigh.

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Wed Dec 12, 2012 10:42 am

Joined: Thu Sep 17, 2009 6:07 am
Location: Germany, Bavaria
Today, I started my first solve and I have to say, this is really hard, especially hard to turn. I had said
rline wrote:
...So Konrad, I take it by this that you're saying it will be quite a challenging solve? Will you be using commutators for most things?
Yes to both questions (probably more tedious than really hard, but very easy to screw it up...
quoted from the first post of Brandon in this topic.
I'll stick with my first opinion, but I want to add one thing: Finding a plan and finding the sequences is easier than solving the puzzle in reality.

I admire rline tremendously, because he has done half of the triangles by intuition. I gave up doing this very soon. I found the easy part hard enough without caring for triangles. The big question is: Solving the triangles as you do the first half doesn't come for free. How many moves and thinking is reqired for this?
Maybe, doing them last doesn't cost much more?

I'm certainly less visually gifted than Andrea, Brandon, rline, Burgo, DKwan, schuma, ...[EDIT: I edited the list for "ladies first"], so, I needed my time to get used to the movement of the petals on this puzzle. Fortunately, the reversing of setup moves is not so hard on this puzzle than on others (e.g. the Pentultimate). Finding the setups was more of an issue for me.
Here I am:

Because, I'll be travelling soon, I'll leave it at this state until end of the year. (I'm behind schedule for my travel preparations already.)

Congratulations rline!

(Unfortunately, I share the opinion of all the others about turning. Lubrication and breaking in didn't help much.
On the bright side: No pops. The triangular stickers are fixed now by superglue.)

When I return at the end of the year, I'll try the hard part.

_________________

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Wed Dec 12, 2012 3:19 pm

Joined: Mon Feb 28, 2011 4:54 am

Quote:
The big question is: Solving the triangles as you do the first half doesn't come for free. How many moves and thinking is reqired for this? Maybe, doing them last doesn't cost much more?

Trust me: doing them last costs way more! It's much much easier doing as many as possible first, so that all the 1st half ones are done. I'm quite sure I could convince you that it's worth it. I'm currently in process of making a video or four to show my method. The 2nd video (the one solving all of the lower half) will be the longest one I've made I think. Once you see how the triangles move it becomes easier. I've found that even with only having to 3-cycle the triangles at the end, it's still very hard. I almost think there might be a need for multiple similar 3-cycles to account for the many differrent positions the end triangles can be in.

Oh, and many thanks for your encouragement

_________________

Blogger Twisty Puzzling - Simple Solutions for Puzzling Twisties

Tutorials: Hexaminx Blockbuilding Tutorial | Curvy Starminx

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Wed Dec 12, 2012 4:14 pm

Joined: Fri Feb 18, 2011 5:49 pm
Location: New Jersey
The big question is: Solving the triangles as you do the first half doesn't come for free. How many moves and thinking is reqired for this?
Maybe, doing them last doesn't cost much more?

Just to back-up rline, block-building for half the puzzle should definitely be the way to go for fewest-moves for solving on GB. Requiring a [5,1] to solve pure at the end means if you solve 2 at a time with 0 setup moves, you will require 6 moves per piece. This is a generously low estimate of moves-per-piece for a solve with commutators because setups will certainly not be 0. When you solve the pieces of the first half with block-building methods, 6 moves is more than enough to properly align a single additional piece (with some thought), especially because of the duplicates in this case and how much "room" you have to work with on the other half of the puzzle.

For a physical solve, the best way to leverage the "half-puzzle-first" approach will be slightly different than for GB fewest-moves solving. It probably works best (for time) to have some simple/short conjugates that can be used for inserting small groupings, and then just getting used to how to setup the pieces for those cases. For example, one possible procedure would be to pair a petal+triangle group on the incomplete half of the puzzle, then use a simple [1:1] conjugate to insert the pair appropriately between some already-solved edges. This is comparable to F2L for a 3x3 I think.

I'm sure rline's coming videos will demonstrate this in more detail.

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Fri Dec 14, 2012 5:14 am

Joined: Thu Sep 17, 2009 6:07 am
Location: Germany, Bavaria
I could steal some time and have solved it from the state shown above.
The hardest thing is to plan the setups. After some time, my brain got accustomed a bit to see how pieces will move. This was my very first "Icosahedron" type of puzzle.

Five triangles to go (you need to enlarge the pictures, if you are really interested):

The last two:

Regarding the strategy "triangles last", I was definitely not talking about fewest move count, I was talking about my time solving a puzzle completely unknown before. I can talk only about myself and my given visual capabilities and "intuition":

I looked for the necessary commutators on Gelatinbrain 2.1.1. (OK, I improved the triangle 3-cycle by using Brandon's [5,1]; my sequence was much longer.)

With this given knowledge, I could do the first part (everything less the triangles) straight forward.
If I do half of the triangles intuitively during this first phase, I have to do a lot of thinking how to build the blocks and have to invent much more. So, this first "easy phase" would have consumed much more time. The question for me is, how would my balance would have looked like? (In my second solve it could be already quite different, as I have now improved my "intuition" a lot, havong learned so much about the puzzle.)

BTW, by accident a few triangles will be solved anyway and at the beginning finding an opportunity for a 3-cycle with very few setup moves is easy. It becomes harder for the last twenty or so. In this phase we are on the same trip, though.

P.S. Turning has improved considerably. I had no serious catches.
I'll just continue turning it, maybe, give the puzzle a new round of lubrication, but will not try anything else, so far.

P.S. #2 I have lubricated the puzzle again and it is getting better and better.
My advice is: Patience pays off. There are many opportunities to turn it while your brain does other things.
@kitslam: If you read this, solving it with one hand may be even beyond your amazing capabilities.

_________________

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Fri Dec 14, 2012 11:29 pm

Joined: Mon Feb 28, 2011 4:54 am
Well I've finally finished doing my videos for this puzzle. My dayjob commitments kept getting in the way but they're done now. These videos are a little unusual just from their length. They come in at over 2 hours. The 2nd video is 1 second short of an hour, though I've put some links in it to navigate to different parts of the video if desired. I hope at least one person out there watches the whole thing I decided to basically do an entire solve, apart from the last video of the triangles, where I've done a number of examples but not every single one. If anyone's struggling with building up the whole lower half I'm fairly sure watching that second video will make things clear.

(I should point out that I interchange the words "face" and "vertex" more than I should so ignore that. Also, around the 55 minute mark I say that I've done something wrong but then realise that it's right and all is well - I hope you can forgive me, since there are no cuts in the video and by the end I was ready for a nap )

I've loved figuring this one out. I keep saying it but I'll say it again: this is a beautiful puzzle, aesthetically and also as far as solving goes. The turning is average but I would strongly recommend that anyone who's not sure whether to risk it or not does indeed "risk it". Even with the not-wonderful turning it's still worth the money.

Written tutorial: http://twistypuzzling.blogspot.com/2012 ... edron.html

Part 1 - Bottom Face
Part 2 - Lower Half
Part 3 - Upper Half Edges
Part 4 - Upper Half Petals
Part 5 - Centers and Triangles

_________________

Blogger Twisty Puzzling - Simple Solutions for Puzzling Twisties

Tutorials: Hexaminx Blockbuilding Tutorial | Curvy Starminx

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Sat Dec 15, 2012 3:57 am

Joined: Thu Sep 17, 2009 6:07 am
Location: Germany, Bavaria
Wow rline!
I just had a quick glance at your written tutorial. (As I said above, my time for puzzles is more limited these days than usual.)
This is an impressive work!

If you think back to your comment that you have lost one hour stickering it, how many more hours have you "lost" since then? And how much fun has it created? (We forget about the stiff turning, that was annoying sometimes for sure.)

I share your opinion completely, that this is a wonderful puzzle, so beautiful, such a nice challenge!

I'm looking forward to have a look at your videos around the end of the year.

P.S. I have looked at your part1. Very clear and easy to follow. Listening to your language is quite relaxing for me, as a non-native speaker. The same is valid for Burgo's videos. You Aussies are quite compatible with my limited capabilities getting sense out of spoken English

P.S #2: As far as I know, there is not such a thing as a "standard colour" scheme for Dodecahedra puzzles.
Even the 12 colours vary. E.g. when I get custom made stickers for custom Dodecahedra, some colours are similar, some are not. Whenever I sticker a dodecahedrom myself, I use the a personal scheme (shown in my photos above) and try to stick as close as possible to existing puzzles in my collection.

_________________

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Sat Dec 15, 2012 4:28 am

Joined: Mon Feb 28, 2011 4:54 am
If you think back to your comment that you have lost one hour stickering it, how many more hours have you "lost" since then? And how much fun has it created?

Well, making the videos was 2-3, figuring it all out was probably 15 hours worth of solving, lost time due to turning issues was at least one hour ( ), stickering, lubing, tensioning, breaking in, adjusting probably around 3. So far, totalling 21 hours. Paid \$42. Comes out to \$2 per hour of use. Pretty good value!

And the fun? Well you can't put a price on that

I genuinely hope it can all be a help to others as well.

_________________

Blogger Twisty Puzzling - Simple Solutions for Puzzling Twisties

Tutorials: Hexaminx Blockbuilding Tutorial | Curvy Starminx

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Sat Dec 15, 2012 5:45 am

Joined: Mon Feb 27, 2012 10:57 am
Location: In my study drooling over my puzzle hoard - Precioussssss!

That is an unbelievable piece of work Rline!

I am really pleased to see that you manage to use the basic ideas from the Marshall technique very creatively - this gives me hope that I should be able to get quite far by myself. Thanks to your discussion above and Brandon's video I sort of understand the basics of commutators (although as soon as it gets beyond the basics I'm lost in the notation!) so I might even be able to do the final small triangles!

I plan to use just your outline initially as my "battle plan" and then see whether I can solve it myself! After this I will, as Konrad says, listen to your dulcet Aussie tones on the videos to hear your own explanations. As always, I will rip the YouTube videos for my own personal collection for offline viewing.

Sooo many puzzles arrived recently and sooo little time
Twisties arrived or in transit: Traiphum's Megamorphinx, Hawk's 4x4x2 & 5x5x3, Starminx I and II, Classical Tuttminx & 3x5x7 cuboid (more from Traiphum ordered for next year!)
Metal puzzles in transit: Wil Strijbos' Lotus, Hanayama 2 rings teddy, Constantin's 4 stones labyrinth
Wood puzzles in transit: rather an embarrassing number of burrs and cubes coming from all over the world - as far away as Australia.

_________________
Kevin

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Sat Dec 15, 2012 6:32 am

Joined: Tue Feb 08, 2011 3:17 am
Location: Australia
Hi Friends,

Unbelievable, 2 pages already, within a week (and a tutorial video)! I had a twist of the Bauhinia today and then a read of the posts so far, thanks for everyoneâ€™s contributions, it was an interesting read.

I didnâ€™t build any full faces like Rline, I couldnâ€™t see an efficient method, but I did do the `petals` of the first face and those directly surrounding them as I went, and I built the edges from the bottom up. Then I solved by piece type, petals > centres > and last the centre points.

The difference I think is the way that I solved the centre points. My 3cycle went like this:
[FLâ€™ (UR UFLâ€™ URâ€™ UFL) FL (UR UFLâ€™ URâ€™ UFL)] x2
I found it fairly flexible and easy to set up for visually. If you perform the sequence youâ€™ll see itâ€™s flexibility visually.

The puzzle itself still needs breaking in more, and I applied a liberal amount of lube which helped a lot, but I have given my forearms a bit of a workout! I think after it is fully broken in the turning will be much better. I didnâ€™t have any issues with popping or catching, but itâ€™s pretty sharp every now and then.

Cheers,
Burgo.

_________________
1st 3x3 solve Oct 2010 (Even though I lived through the 80s).
PB 3x3 55sec Jan 2011 (When I was a kid 1:30 was speedcubing so I'm stoked).
1st 3x3 Earth (nemesis) solve Jan 2011 My You Tube (Now has ALLCrazy 3X3 Planets with Reduction)

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Sat Dec 15, 2012 9:28 am

Joined: Wed Apr 13, 2011 8:37 am
Location: Germany
Hi guys,
is the bauhinia less hard than ft starminx ?

Cheers,
Andrea

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Sat Dec 15, 2012 10:31 am

Joined: Thu Sep 17, 2009 6:07 am
Location: Germany, Bavaria
Hey Andrea,
my personal opinion: I needed some time to get used to the movement of pieces on the Bauhinia in case more than one layer (vertex) needs to be turned for setup moves.
The overall piece counts are 162 on the Bauhinia, 102 on the FT Starminx.

The number of harder parts (triangles on the Bauhinia, star points on the Starminx) is identical (60).

So, the main difference for me was the fact that the reversing of setup moves was less dificult and risky on the Bauhinia than on the Starminx.

All in all, maybe, the difficulty of both is not much different.
I'm interested what others will say.

In any case, the Bauhinia is worth the money.

Yes, it is a bit stiff, especially at the beginning, but this a good training for your arm muscles.
If you happen to be a rock climber (I was one thirty years ago), you will welcome this

_________________

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Sat Dec 15, 2012 11:22 am

Joined: Fri Jul 22, 2011 8:12 am
Location: Paris
rline has done an excellent tutorial on this cube.
Great job, as usual !
I agree with Konrad, the solve on the bauhunia is pretty similar to the ft starminx, but set-up moves (for triangles) seems to be less disturbing. On the other way, you have to be more clever solving petals and edges than with the ft starminx, because for this last puzzle, one use pyraminx crystal algs.

But, I have not received my puzzle, so...

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Sun Dec 16, 2012 12:05 am

Joined: Tue Feb 08, 2011 3:17 am
Location: Australia
Hi Rline,

I just watched your Bauhinia videos.. What an exceptional piece of work ! I particularly enjoyed the way you built the lower 1/2 of the puzzle. Really interesting, and it reminded me of solving the Crazy Earth in my avitar.

I also really enjoyed seeing you build the top Petals, and doing the Centres with your adapted CPS sequence. It's just really refreshing to watch someone elses method.

I liked the `Live Action` sticker replacement! Isn't that cheating though?

Thanks for the videos,
Awesome!
Cheers,
Burgo.

_________________
1st 3x3 solve Oct 2010 (Even though I lived through the 80s).
PB 3x3 55sec Jan 2011 (When I was a kid 1:30 was speedcubing so I'm stoked).
1st 3x3 Earth (nemesis) solve Jan 2011 My You Tube (Now has ALLCrazy 3X3 Planets with Reduction)

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Sun Dec 16, 2012 4:11 am

Joined: Mon Feb 28, 2011 4:54 am
First, thankyou everyone for your kind words about the videos. I'm very glad they're helpful. I especially value Burgo's comments because he's an Ausse too he's such a fantastic solver.

Burgo wrote:
I liked the `Live Action` sticker replacement! Isn't that cheating though?

Hmmm, well now I know for sure that you watched the whole lot I did mention there were no cuts in the videos.

Burgo wrote:
The difference I think is the way that I solved the centre points. My 3cycle went like this:
[FLâ€™ (UR UFLâ€™ URâ€™ UFL) FL (UR UFLâ€™ URâ€™ UFL)] x2
I found it fairly flexible and easy to set up for visually. If you perform the sequence youâ€™ll see itâ€™s flexibility visually.

Burgo this is an outstanding sequence! So simple, so symmetrical and cycling triangles on 3 adjoining faces. Unbelievable as always... I wish I'd come up with this one! (I'm still happy with mine cause I found it myself, but yours is wonderful).

Burgo wrote:
I think after it is fully broken in the turning will be much better.

Unfortunately I don't think so anymore. I've done a lot of solving with it and it improved somewhat with heaps of lube but it seems to have reached a point where it won't improve any more, and in fact seems occasionally to be slightly worse.

P.S. I have looked at your part1. Very clear and easy to follow. Listening to your language is quite relaxing for me, as a non-native speaker. The same is valid for Burgo's videos. You Aussies are quite compatible with my limited capabilities getting sense out of spoken English

Yeah we Aussies know how to make people relaxed. You can't trust those Americans when it comes to clear speech

_________________

Blogger Twisty Puzzling - Simple Solutions for Puzzling Twisties

Tutorials: Hexaminx Blockbuilding Tutorial | Curvy Starminx

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Thu Jan 03, 2013 5:24 am

Joined: Wed Apr 13, 2011 8:37 am
Location: Germany
Hi friends,

Brandon: Thank you for your video with the commutator.

Rline: Thank for your great solution videos. It must a lot of work. I prefer solution to solve a puzzle area by area. Youre great solver.

The advantage is the color scheme. The "original" color scheme has the advantage that opposite colors have similar color values and different brightness and saturation.

Light yellow in opposite of yellow, light blue in opposite of blue etc.
So you must only memorize 6 colors. Front blue up white left red, back green is like original Rubiks Cube. You must only memorize the yellow and purple (back/right).
The crazy megaminxes with colored plastic are colored in this way.

schuma: thank you for the program. It' very nice. Nice idea to make mobius projections with riemann spheres.
I used this program to explore commutators and conjugates.
I don't know which gelatinbrain is near to the bauhinia the 1.2.3 or the 1.2.11 ?

The mobius/riemann is very near to the real bauhinia, I think.

I used a different notation only for me at home, to write down sequences:
Because the puzzle is icosahedral ( edges like alexander star)

Code:

2+

2

1+                           3+
1                    3

5          4
5+             4+

The inner pentagon I number clockwise from 1 to 5.
The adjacent 5 outer corners I used the same number with a "+".
In this way you must not think obout geometric directions. And the main corners have only one symbol.

Example the triangle commutator
(1' 3 1 , 5 1+ 5') =
1', 3 , 1, 5, 1+, 5',
1', 3' , 1, 5, 1+', 5'

Cheers,
Andrea

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Thu Jan 03, 2013 5:40 am

Joined: Thu Sep 17, 2009 6:07 am
Location: Germany, Bavaria
Sorry, am I supposed to see a diagram together with your code?
I see only numbers and as it stands I cannot interpret them.

_________________

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Thu Jan 03, 2013 7:25 am

Joined: Wed Apr 13, 2011 8:37 am
Location: Germany
Sorry, am I supposed to see a diagram together with your code?
I see only numbers and as it stands I cannot interpret them.

On face turning hexaedron (Rubik's Cube) it's easier to use symbols 'R' 'L' 'B' etc.
On Rex Cube (face turning octahedron) it's possible to number the corners (vertices).
On face turning dodecahedron (Megaminx) perhaps it's easier to use your notation.
On face turning icosahedron ( vertex turning dodecahedron) perhaps it' easier to use numbers ?

In gelatinbrain the dodecahedron axis get letters 'A' 'B' 'C' ...
Perhaps it's easier to give numbers instead.

The bauhinia has 10 turnable corners in one half sphere.
Give the inner 5 corners the numbers 1 to 5.
Instead of giving the numbers 6 to 10 to the outer vertices I gave the the number 1+ to 5+ to the outer vertices.
The reason is that it's easier to see that a 2+ is adjacent and outside of the 2.
Otherwise it's not easy to see that a 7 is near the 2 or a 'G' is near the 'B'

So a number denotes a clockwise turn. Anticlockwise turns have a ' .

Attachment:

bauhinia1.jpg [ 84.62 KiB | Viewed 2352 times ]

Make a 1 ' move. This looks like this picture:
Attachment:

bauhinia2.jpg [ 82.08 KiB | Viewed 2352 times ]

An outside turn looks like this:
Attachment:

bauhinia3.jpg [ 80.99 KiB | Viewed 2352 times ]

I hope this help to understand my notation.

I was able to use this notation to transfer sequences from the different simulators to the real Bauhinia Dodecahedron.
The intention of this notation was : make it easier to associate symbols with vertices.

Cheers,
Andrea

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Thu Jan 03, 2013 11:45 pm

Joined: Thu Dec 31, 2009 8:54 pm
Location: Bay Area, California
Burgo wrote:
The difference I think is the way that I solved the centre points. My 3cycle went like this:
[FLâ€™ (UR UFLâ€™ URâ€™ UFL) FL (UR UFLâ€™ URâ€™ UFL)] x2
I found it fairly flexible and easy to set up for visually. If you perform the sequence youâ€™ll see itâ€™s flexibility visually.

Hi Burgo,

I was intrigued by the form of your sequence. You aren't using a standard commutator so I decided to look into why your sequence works. Your use of a setup move on the [1,1] commutator and then applying the commutator again so that the net effect is a 3-cycle + 2-2 swap is creative although rather inefficient. I was going to type a long description of your sequence and provide an alternative but I figured a video would be better received so that's what I did: http://youtu.be/1cMEBeH5szo

Perhaps you can chime in on the thought process you used for finding the sequence? It's very likely that my analysis of your sequence is not the mental process you used for creating it. I'd very much like to understand your thought process because I'm sure there are subtle insights there that would be useful for a great many other puzzles.

_________________
Prior to using my real name I posted under the account named bmenrigh.

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Fri Jan 04, 2013 4:04 am

Joined: Wed Apr 13, 2011 8:37 am
Location: Germany
Hi Burgo,

thank you for sharing your sequence. Very usefull , this. In my notation
3 ( 1+', 2, 1+, 2' )
3' ( 1+', 2, 1+, 2' )

Cheers,

Andrea

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Fri Jan 04, 2013 8:05 am

Joined: Tue Feb 08, 2011 3:17 am
Location: Australia
Hi Brandon,

Thanks for your video, itâ€™s interesting seeing how someone else views what I do. Sometimes you have to see how things work during a solve to understand why it might be appealing to use, or how I came up with it. I often come up with things during solving. A small example of seeing this would be putting the pieces `back` by: [FLâ€™ (UFLâ€™ UR UFL URâ€™) FL (UFLâ€™ UR UFL URâ€™)] x2 because that is how I am `seeing it` when I am inventing it, and I will explain the moves:

I generally hold the puzzle with the corner UFR directly in front.
I will start by mentioning my 2 Petal 3cycles (which everyone else would use too): Wide: (UR UFLâ€™ URâ€™ UFL) and Narrow (URâ€™ UFL UR UFLâ€™). The Point 3cycle will work on the wide and the narrow versions: the setup for the narrow is UFR.

My sequence is an alternate way to use these 3cycles of groups, it came about because I noticed that doing a vertex twist broke the group by a single Point. I knew how to make the Commutator sequences that could be developed here, but just I found them visually confusing.. for some reason I donâ€™t like lifting something up, doing something and putting it back, and then reversing that after.

The setup FLâ€™ (it doesnâ€™t matter if itâ€™s FL or FLâ€™) because the intention is to `store` all but one Point of the `whole moving group` (consisting of 3 Points, 1 Centre and 1 Petal) . It also introduces a surrogate subgroup.

The 3cycle (UFLâ€™ UR UFL URâ€™) moves the Point that matches the `stored group` into position [it can be reversed by exchanging (UR UFLâ€™ URâ€™ UFL) `left up or right up`]. FL completes the group and stores the next subgroup.

Repeating (UFLâ€™ UR UFL URâ€™) moves the next Point into position and FLâ€™ completes the group.
And the cycle continues once more (UFLâ€™ UR UFL URâ€™) FL.

The last (UFLâ€™ UR UFL URâ€™) just returns the completed groups.

It works a bit like a Sune.

Of course, when you are in a solving situation you often place 2 points and not 3, so remembering the group that will not be matched correctly will help with visualizing the flow of the sequence. I just like `seeing it all working` during the sequence, I find it reassuring as Iâ€™m plugging along, the move count for me is less important.

@Andrea, your nomenclature makes sense, but personally I prefer letters to numbers in visualizing.

Cheers,
Burgo.

_________________
1st 3x3 solve Oct 2010 (Even though I lived through the 80s).
PB 3x3 55sec Jan 2011 (When I was a kid 1:30 was speedcubing so I'm stoked).
1st 3x3 Earth (nemesis) solve Jan 2011 My You Tube (Now has ALLCrazy 3X3 Planets with Reduction)

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Sat Jan 05, 2013 7:00 am

Joined: Wed Apr 13, 2011 8:37 am
Location: Germany
So, the main difference for me was the fact that the reversing of setup moves was less dificult and risky on the Bauhinia than on the Starminx.

All in all, maybe, the difficulty of both is not much different.
I'm interested what others will say.

I solved my Bauhinia. I need two days. Some fails. I need perhaps 7 hours to solve it.
Yes it's easier undo the setup move, but it's more tricky and difficult to find setup moves to solve the triangles.
Turning is more hard. But my ft starminx had more catches.

Perhaps it' a little bit harder than the ft starminx.

I used the solution method from rline. But I replaced Brandons (3,3) commutator with Burgo's sequence. I decided which sequence needs less setup moves. In 99% Burgo's sequence needs less setup moves. Burgo's sequence is easier to turn. The number of moves doesn't matter.
Thanks Burgo.

Cheers,
Andrea

OT: omG, my rainbow nautilus, I haven't an idea to solve it.

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Sat Jan 05, 2013 11:19 am

Joined: Thu Sep 17, 2009 6:07 am
Location: Germany, Bavaria
@Andrea,
looking backwards, the first solve of the Bauhinia was quite a bit easier for me than the first solve of the FT Starminx two years ago (it was Tom's Mini Starminx).
The main reason is that the reversing of setup moves seems to be quite easy on the Bauhinia. I did not screw up once doing this. (Once I made a failure doing a triangle 3-cycle, because I watched a TV movie in parallel , but never did I fail to reverse a setup sequence.) During my first FT Starminx I screwed up several times.
In both cases I had the complete theory of my solving strategy in my head before the scramble.
The only problem was to translate theory into prctice!
Probably it counted, too, that my first Mini Starminx was a misprint and did not turn so well. (OK, at the beginning it was terrible. Not Tom's fault, though)
Just my personal experience, certainly somebody else can view it differently.

@Andrea's notation: It certainly makes sense (your second post with the pictures made it perfectly clear Thank you!). For all 20 vertex layers you would need to count to 10 and it is not obvious to me how you would do this.
Probably, everybody will like what he/she invented. I prefer something that is known to the readers, when it comes to sharing sequences.
The Gelatinbrain notation is around for a long time and has the big advantage that you can cut and paste sequences into the program.
Therefore, I said that if I ever post a sequence here, I would provide it twice, in Gelatinbrain's and in Konrad's notation (as Brandon was so nice to call it in his sequence utility ).
This is an advantage too, that Brandon's utility translates so nicely from Gelatinbrain's to Konrad's notation.
I may have said this before about standards (and this is based on a long history of looking at computer standards): One accepted standard is better than two excellent competing standards.
In a short sentence: I'll stick with the letter notation.

@Burgo's sequence: I found both interesting, Brandon's video and Burgo's reply.
Burgo wrote:
Hi Brandon,

Thanks for your video, itâ€™s interesting seeing how someone else views what I do.
I think Brandon just made an analysis why it works and I found his video easy to follow.
He showed a [3,3] commutator that achieves the same 3-cycle in 12 moves:
[CBH, AFB', CBH', CHG', CBH', CHG, CBH, AFB, CBH', CHG', CBH, CHG] or [[CBH:AFB'],[CHG':CBH']] or [UR, UFL', UR', R', UR', R, UR, UFL, UR', R', UR, R] or [[UR: UFL'],[R':UR']
Brandon did not say that he wanted to describe Burgo's thought process, on the contrary:
bmenrigh wrote:
...
Hi Burgo,
....
Perhaps you can chime in on the thought process you used for finding the sequence? It's very likely that my analysis of your sequence is not the mental process you used for creating it. I'd very much like to understand your thought process because I'm sure there are subtle insights there that would be useful for a great many other puzzles.
While I congratulate Burgo for finding his sequence â€“ it is certainly much better than my first 30 move sequence â€“ I'm surprised that several people prefer it over Brandon's [5,1] described in his first post of this thread.
I understand Burgo himself, if he sticks with his own sequence. I do not follow other persons who will take something foreign anyway. I prefer Brandon's [5,1]
1. it is shorter by 8 moves
2. it is very easy to remember and very easy to understand why it works
3. the 3-cycle is between three different vertex layers. In my view, this allows even more flexibility for the setup moves.

@Nautilus: I had hoped that you can help me, if I scramble it and got lost. Maybe, I have just to wait a few hours more? (Moderator hat on: If you want to post any details, please, make a new thread. ) For now I have many other things to do.

_________________

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Sat Jan 05, 2013 5:06 pm

Joined: Mon Feb 28, 2011 4:54 am
Quote:
I understand Burgo himself, if he sticks with his own sequence. I do not follow other persons who will take something foreign anyway. I prefer Brandon's [5,1]
1. it is shorter by 8 moves
2. it is very easy to remember and very easy to understand why it works
3. the 3-cycle is between three different vertex layers. In my view, this allows even more flexibility for the setup moves.

I can chime in here. I found my own sequence, then looked at Brandon's. I liked mine better because I found it and because it "felt" better to me. Mine was longer than Brandon's. Then (after making my videos) I tried Burgo's and loved it. I wish I'd come up with it. To me, his sequence, although "longer" was much much easier to remember and perform than mine or Brandon's. I think a lot of the time the factor of "what feels right" plays a much larger role than we might think.

_________________

Blogger Twisty Puzzling - Simple Solutions for Puzzling Twisties

Tutorials: Hexaminx Blockbuilding Tutorial | Curvy Starminx

Top

 Post subject: Re: How to solve the Rex / Bauhinia DodecahedronPosted: Mon Jan 07, 2013 5:03 am

Joined: Fri Jul 22, 2011 8:12 am
Location: Paris
I also tought that bauhinia is easier to solve than ft starminx, but this is longer... I have solved the bauhinia at the first attempt, altough I have messed up a lot of time the starminx at the end. Just because, the set-up move is easier to undo than on a ft starminx.
I have used rline sequence, because this is easy to remember, to understand and to apply.

Top

 Display posts from previous: All posts1 day7 days2 weeks1 month3 months6 months1 year Sort by AuthorPost timeSubject AscendingDescending
 Page 2 of 2 [ 78 posts ] Go to page Previous  1, 2

 All times are UTC - 5 hours