Wednesday, June 24, 2020
Just when you think that nothing new can be done with a 6-piece burr, a new design comes out. I have quite a few puzzles in this particular format. I like to see how designers transform such an innocuous looking puzzle into a serious challenge that belies its appearance. So, when Cubicdissection offered Welded Burr by William Hu, I made sure to acquire a copy. After maturing in my puzzle pile for 6 years, I figured that it had reached its peak character and I pulled it out for a go.
In no time at all, I had separated the puzzle into 3 sets of 2 pieces. Unfortunately, Eric Fuller apparently had some gluing snafus and the pairs were stuck together. It was like they were welded together…
The pieces were made from Aformosa and the ends beveled for a nice finished appearance. Each piece does look like it was made by gluing 2 standard burr pieces together as the name would suggest.
With a level of 4.2 and no rotations, you would assume that this puzzle is not difficult and you would be right. Even as an assembly puzzle, the characteristics of the pieces hint at how they are paired together.
Joining the pieces together was an interesting take on the 6-piece burr, which makes me wonder if anyone has done an exhaustive search over all possible welded 6-piece burrs to see what new possibilities can be generated. For all I know, William has already done that and Welded Burr is the result of that search. If not, it may be worthwhile for someone to take on that challenge.
Welded Burr is a great puzzle to offer to a new puzzler. With only 3 pieces, a low level of difficulty, and hints for assembly, it is very approachable for just about anyone. It’s also a good stress reliever to play with if you are having issues solving more difficult puzzles, which is why I pulled this one out of the to-do pile this week. And you thought that I only pushed the difficult puzzles into the pile.
Wednesday, June 17, 2020
What could possibly be better than a board burr? How about a board burr in a cage? Osinori Yamamoto’s Cockpit puzzle is such a challenge. With only 3 simple pieces within the frame, it is more a flight of fancy than a long-haul expedition.
My version of Cockpit was made by Brian Menold at Wood Wonders with a Canarywood Frame with Wenge Pieces. The cage is completely open on the top and bottom and has openings on the other 4 sides in the shape of a plus. Two of the pieces are identical and the other has one additional cubie. The puzzle came assembled and the objective is remove the pieces from the frame and then put them back in at some point in time.
As with many caged board burrs, the trick is to figure out the pass. This is the sequence of moves that will eventually allow one of the pieces to get pulled passed the piece that is running through it. Once you have accomplished the pass, the puzzle is pretty much solved. With a level of 16.3.2, the pass was well designed and fun to execute.
|New Email Address is |
This puzzle has been around awhile and you can read the Puzzling Times entry on Cockpit from 2013 here. It’s about time I got this one off my to-do pile.
Wednesday, June 10, 2020
Now here’s a puzzle that you can sink your teeth into and chew on for a few days. I’ve always found board burrs challenging, and then my buddy Andrew Crowell came along and added a twist. Or several twists as it turns out. While other puzzle designers were wondering how to remove rotations from their puzzle solutions, Andrew was wondering how to remove the rectilinear moves.
The first puzzle in Andrew’s new Rotating Interlocking Board (RIB) puzzle line is RIBeye. The copy that I had an opportunity to play with was beautifully made by Andrew himself from Lacewood, Purpleheart, and Wenge.
At first glance, the RIBeye pieces don’t seem so intimidating. Of the 6 pieces, there are 2 pairs of identical pieces. Not only that, but the first pair is the simplest possible board burr piece and the other is the second simplest possible board burr piece. Wow! This is going to be a piece of cake! The remaining 2 pieces are a bit different, but there are ONLY 2 of those. One of these obviously stands out since it is non-planer and has a piece sticking out to invade his neighbor’s space. You can argue that RIBeye is not technically a pure board burr, but why waste time on that when you can just enjoy the puzzle.
I always find these puzzles tricky to start. You need to figure out how the pieces are oriented with respect to each other, which is quite a challenge. Typically, for me, this entails starting with an initial guess and making changes as I go along and run into dead ends. The fact that this puzzle requires rotational moves makes it all the more difficult to identify a failed attempt because it could always be possible that I am simply failing to identify a critical rotation.
What kind of clues do the pieces offer to get started?
- The first clue is that one of the pieces has a sticker with the puzzle name on it and I’m assuming that it goes on the outside of the puzzle. Unfortunately, Andrew recognized that this would be a clue and put it on a symmetric piece that can be rotated so that the sticker can be either inside or outside. No help there.
- One of the pieces has an edge with a deeper dip in the bevel in the middle of the edge. Is this required to support a rotation? Nope.
- The most interesting thing that I noted is that the pieces were not built with the Lacewood grain exposed in the same direction. Most unusual. In fact, it seems that the pieces could almost be divided into 3 sets of pairs with the same grain orientation. Was this the case? Nope.
- A couple of splines left a small gap at the bottom of the channel. Is this revealing something even sneakier than a rotation? Nope. It might be possible to use a mixture of Wenge sawdust and glue to make a nearly invisible filler.
RIBeye took me 2 days to solve. I started by envisioning how the first piece could be removed from the final assembly and quickly settled on the “x move disassembly configuration” that I thought was being used. I then set that potential first piece aside and focused on the remaining 5.
The 2 unique pieces could not be paired together, and a lot experimentation revolved around how they could interact with each other. A lot here means hours. At the end of the first day, I hadn’t proceeded very far.
Several days later, I started day 2. Halfway through the day, I decided that I knew how to remove the second piece from the disassembly and put that one aside with the first piece. Four pieces left to go and the crux of the challenge was still left to be discovered. At 15 moves to remove the third piece, you can see why it was not particularly obvious. It may not have been obvious, but it was spectacular. It required several rotations and interesting rectilinear moves to accomplish the rotations. The effort to figure out the sequence for the third piece pretty much encompasses the remaining 2, which means that you are really trying to figure out a 27-move sequence. Awesome!
The puzzle was really well made and didn’t allow me to cheat any of the moves with unintentional rotations. However, there were several times where it was almost possible to perform an unintentional rotation, but would have required a little force to accomplish. This puzzle requires no force and if you have an opportunity to play with it, don’t be tempted to force a move.
If you like board burrs, definitely give RIBeye a try. Oh, and don't forget to start with it unassembled.
Wednesday, June 3, 2020
What happens when you increase the design space for new puzzles requiring a lot of rotations? You find something suitable to name it after like the MagellanTIC spiral galaxies.
My buddy, Andrew Crowell, is taking his goal to stump me very seriously. He has graduated from 4x4x4 Turning Interlocking Cubes (TICs) to a 5x5x5 format to provide more opportunities to frustrate hopeful puzzle solvers. The first in this new format is MagellanTIC.
I recently acquired a 3D printed copy of MagellanTIC from Andrew to enjoy. It is comprised of 6 pieces: A frame, 3 square rings, and 2 burr sticks. One of the rings is really a partial ring, but I’ll simply refer to it as ring anyway. The pieces were nicely printed in 2 different colors: speckled gray for the frame and one of the burr sticks and brilliant white for the other pieces. As with most of his 3D printed puzzles, Andrew has the name of the puzzle embedded on the frame. However, he liked this puzzle enough to embed his name as the designer on the inside of the frame as well.
The larger size of this puzzle required a larger effort to solve it. Let’s break the solving process down into steps.
1) Determining where all the pieces go in the frame - Easy. In fact, When MagellanTIC arrived, it was already assembled except for one last piece that needed to be inserted. Not one to trust a partially assembled puzzle, I took it completely apart and left the pieces to sit overnight. Sometimes puzzler designers can be sneaky buggers and if there is a right way and a wrong way to put those pieces in, you can bet that it would arrive the wrong way. My apologies to Andrew, who did send them packed in the correct orientation. Of course, without that one critical piece, it is trivial to take apart and is a great way to ship the puzzle.
2) Determining the order that the pieces have to be added to the puzzle - Easy. Although I did have the last 2 pieces to be inserted in the wrong order.
3) Accomplishing all the rotations and translations to get it all together - Anything but easy. In fact, it required a lot of effort and I spent several hours putting it together followed by a couple more hours analyzing the moves.
4) Taking it back apart - Tough. I took my first picture of MagellanTIC when it was originally assembled. As I was basking in the glory of a triumphant assembly, my wife casually mentioned that I should take it apart and take a picture of the pieces before the lighting changed. Easier said than done. It is not a trivial disassembly.
5) Keep repeating steps 3 and 4 - Interesting. The movements required by this puzzle are not easy to remember and even after putting it together and taking it apart several times, it still required concentrating on the moves to successfully execute them.
My best estimate at the level of difficulty for MagellanTIC is 18.104.22.168.3. With all the complex rotations involved, these numbers might be significantly different from your own assessment. Of course, when assembling the puzzle, you’re working the level of difficulty from right to left. The first time around, the number of moves was easily in the hundreds as I was rotating pieces unsuccessfully trying to get them in position.
When focusing on assembling a puzzle, you sometimes miss out on the how nice the movements are. It is not until you have it assembled that you can focus on the moves to accomplish the desired solution. Let’s look at each piece in the disassembly process:
15) The first piece takes 15 moves to remove it from the puzzle. The rotation required to move this piece is nicely designed and much more obvious when taking it apart than putting it together. I completely missed out on how nice it was when assembling the puzzle for the first time.
4) This one took me a bit of work to figure out how to start the insertion into the puzzle. Note that at this point of the assembly, there were already 3 other pieces in the frame. When I was disassembling the puzzle, this piece just fell out when I wasn’t looking. What originally was a tricky task during assembly, turned out to be relatively simple as demonstrated during the disassembly.
13) This is really the crux of the puzzle. I spent most of my time working through this part. It’s a little rotating nightmare. The first time that I thought that I had solved this section, I had it in upside-down. I didn’t even realize it was upside-down until I tried to add the next piece and discovered that it couldn’t reach its final resting place because of the incorrect orientation. Unfortunately, that meant that I wasn’t even close. It was much more difficult to put it in correctly. Imagine that. Personally, I found this stage a bit of a brain warping experience, and even though I can manage it quite well now, I found it difficult to count the moves with all the rotations. Unfortunately, I started working on this puzzle late at night and had to put it down until the next day while in the middle of this part. When I came back to it later the next day, I had a difficult time getting the pieces back out for another attempt.
9) Once you spend time analyzing the puzzle during the disassembly process, you discover that this part of the solution doesn’t really take hundreds of moves to accomplish. In fact, at 9 moves including rotations for only 2 pieces in the frame, it is fairly short and sweet. However, short and sweet does not mean easy and I still have to pay close attention during this part. It is nearly impossible to figure out how to accomplish this when assembling the puzzle for the first time.
3) How difficult can it be put a single piece in the frame in 3 moves. Well, during the disassembly, it’s trivial to take it out. However, during the assembly, you have no idea where it is required to be and this needs to be discovered simultaneously with the next piece. And yes, it does need to be in a very specific location.
At this point, I’ve taken the puzzle apart and reassembled it many times. Although I can perform the moves fairly proficiently, it requires a good memory and a deft touch to keep all those rotations at a minimum.
So what does the larger design space get you? Both the high number of moves to release the first piece and the complex rotations we’ve come to love in the TICs. There are also no visible voids in the assembled puzzle. If TICs are your thing, you don’t want to miss this one. I highly recommend it. If you’re adventurous, don’t forget to get it unassembled.
This is the sixth post of the monthly Andrew Crowell Rotations and Obstructions Series - Turning Interlocking Cubes (ACROSTIC). You can find the prior posts of the series here:
- January: Puzzling DNA - GeneTIC
- February: TIC, TIC, TIC - PackTIC II