Finding Your Way Through the Labyrinth - Daedalus

Aaargh!  There’s a mushroom growing in my box of puzzles.  Wait a minute - It’s a cubic mushroom.  Aaargh!  One of the puzzles got all moldy.  Ewww!

Luckily, this doesn’t happen often.  However, when it does, it’s no big deal.  Although it looks like mold, it’s from the wax used and the bloom can simply be wiped off.  With a name like Daedalus, I expected that there might be wax issues.  When I received the puzzle, sensitive to Daedalus’ plight, I put it where the sun don’t shine.  That’s right, it’s been in the maze of my basement all this time.  I should mention that once the pieces were wiped down, they were easier to move.  It would have made discovering the required moves simpler without the waxy buildup.

Daedalus was designed by Gregory Benedetti and has a difficulty level of 23.9.8.3.3.1.2.  My copy was made in Padauk by Maurice Vigouroux.  It looks like a 3x3x3 cubic dissection, but that is misleading.  The internal labyrinth, supported by a complex array of pins, slots, and moving pieces (yes, the labyrinth is changing as you are solving the puzzle), really make this a 9x9x9 cubic dissection.  It also requires rotations, so BurrTools is not going to help you.

Cubes, pins, slots, rotations - what’s not to like?  The puzzle consists of 8 pieces with one large frame/cage piece and 7 other smaller pieces.  All the pieces have an assortment of pins and/or slots.  I’m assuming that the 2 little pieces are the Minotaur and Theseus.

Daedalus is definitely a disassembly puzzle.  Starting with the puzzle assembled, you have no indication of the labyrinth’s complexity, which needs to be discovered as you progress through the journey.  During this process, I had the Minotaur and Theseus chasing their tails in rotational hell for quite a while.  Once you have figured out what to do with those 2, the rest is straightforward.  However, figuring out what to do with those 2 pieces is a challenge.

Not Quite Right!

Being able to see the internal structure of all the pieces, I found assembling the puzzle to be easier than the disassembly.  Knowing that none of the pins and slots is visible on the outside of the cube is useful in identifying the location and orientation of the pieces.   Oh, and if the puzzle is made with the wood grain aligned like mine, you can also make use of that information as well.  I failed to note that the first time I reassembled the puzzle and although I successfully constructed a cube, one of the pieces was misaligned.  That gave me the opportunity to enjoy the entire process again.

Daedalus is a fantastic puzzle and Cubicdissection is planning on releasing a batch in the near future.  I highly recommend this one and suggest that you purchase one quickly when they become available before they disappear.

Puzzling DNA - GeneTIC

We are still working on discovering the secrets of the DNA puzzle, but where in there is the puzzle gene?  Would looking for the puzzle gene in the DNA puzzle be a meta puzzle?

Great-Grandpa's Puzzle

As for myself, although my parents are very intelligent, puzzles just aren’t their thing.  So where does that puzzle gene come from?  It turns out that one of my great-grandfathers was a doctor who enjoyed working on puzzles.  I’m hoping that I don’t have to wait for my great-grandchildren to carry the torch.

Speaking of the torch, my wife can’t understand why anyone would display firewood in the china cabinet and occasionally reminds me of the bonfire that will be held when I go.  I frequently have to remind her that she shouldn’t burn through the family fortune all at once. (Upon reading this, my wife pointed out that she was more than willing to have firewood in the china cabinet and was actually the one who made room for it.) 

These thoughts were sparked by Andrew Crowell’s new puzzle, GeneTIC. The puzzle has a difficulty rating of 7.5.5.2.5 (This is my best guess from counting the moves by hand, but I’m sure that Andrew has an official count generated from his program that will be released in a future puzzle update.  Counting rotational moves is not always straightforward).  This is the latest addition to the puzzles Andrew has labeled “Hardest”.  However, in my opinion, this puzzle should be labelled excellent instead of hardest based on the feeling that I had after completing it.  Not difficult, but very enjoyable.

Each of Andrews TICs is a treat and I think of them as puzzle candy.  Each one provides an enjoyable experience from first inspection to determine where the pieces need to reside up through determining the gyrations that each piece requires to get there.  I recommend GeneTIC as an exemplar of this process.

As with many of Andrew’s Turning Interlocking Cube (TIC) designs, there is a big cage/frame piece that holds several smaller pieces together in a cube.  GeneTIC has 5 of these smaller pieces, with the smallest being what I am coming to consider the classic “Crowell TIC piece”.  These innocuous looking little pieces move, rotate, and provide a lot of fun without falling out of the puzzle.

Being a TIC (not you, the puzzle), you know that there is a rotational move required somewhere.  For GeneTIC, 4 of the pieces require rotations and the single piece that is not rotated is the odd one out.  In other words, TIC on steroids, and in a 4x4x4 cubic dissection format!  As a puzzle designer myself, struggling to conceive TIC designs, I can only marvel at the prodigious output of the Crowell TIC production engine in both quality and quantity.

You’ve undoubtedly noticed from the photos that my version of GeneTIC was 3D printed.  As far as I know, all of Andrew’s puzzles have a unique color scheme and GeneTIC can be identified as the puzzle with the white frame and green pieces.  Recently hampered by budgetary constraints, I was forced to make some difficult puzzle acquisition decisions.  I’ve noticed several puzzlers have adopted a buy, solve, sell approach, but I’ve never been able to give up a puzzle and this strategy wouldn’t work too well for me.  My two options were to buy all 14 of Andrew’s hardest TIC’s as 3D printed puzzles or try to select 3 or 4 of them from the list.  This time I decided to collect them all in the 3D printed format.  Of course, the 3d printed versions are nowhere near as nice as the finely crafted works of art made with exotic woods, but my desire to have all the hardest designs ended up being the major consideration.  I’ve seen several comments on the Internet from people lamenting that puzzles have become too expensive to fit in their budget.  3D printing is part of the solution to that dilemma.

So, my recent order from Andrew consisted of all 14 of the hardest and 2 of the medium, 3D printed puzzles.  When they were ready, Andrew emailed me a photo of the 16 puzzles assembled.  However, they arrived as a jumble of puzzle pieces.  Nice guy, that Andrew Crowell!  In all fairness, I requested to receive them unassembled so that I could fully enjoy the discovery process.

Pre-Shipped vs. Received

With 16 of these puzzles at the beginning of this year, I decided to do an Andrew Crowell TIC series with a dedicated entry each month.  GeneTIC is the January post and the first for 2020.  My biggest challenge will be to space them out over the year and keep from doing them all at once now.

An Impediment to Rolling Along - Wheel Lock

A lot of times, OK, most of the time, I don’t pay a lot of attention to the name of a puzzle while solving it.  It can result in merrily rolling along until it bites you in the lug nuts.  This was one of those times.

Wheel Lock was designed by Tzy Hun Chein and made by Brian Menold at Wood Wonders.  It consists of a frame with 4 pieces and has a level of difficulty of 15.18.14.4.

The grain of the Tamboti frame is gorgeous.  It has a nice weight to it and a lovely smooth finish.  It also has a very interesting bevel on the outside edges.  The bevel on each edge is deepest in the center and becomes shallower as it approaches each end until there is only a little smoothing at the corners where 3 edges intersect.  I’ve never noticed this on a puzzle before and don’t recall Brian making any references to this technique on his site.  I don’t know how it’s done, but the result is fantastic.  My best guess would be to use a sanding mop to get this effect.

The frame has an open slot on 2 of the sides.  Decoration or functionally required?  They are just wide enough for the ends of the pieces to slide into.  I’m assuming that these slots will be used to accept the ends of the 4 pieces as they are moved out of the way of other pieces traveling on the solution path.

Wheel Lock was one of several puzzles that I bought from Wood Wonders in August 2017.  When they arrived, I immediately took them apart to make a nice selection of disassembled puzzles adorning the back of my china cabinet shelves in the dining room.  As much as I like the random look of puzzle pieces hanging out of boxes, now a couple of years later, I caved in and sacrificed the random disarray of Wheel Lock pieces to the puzzle gods by reassembling them into an orderly packed configuration.

It would have been easy to insert the 4 pieces within the frame if it weren’t for those little cubes that were sticking out from all sides within the frame.  Since all the pieces have the same simple shape, the complexity of this puzzle is in the pattern of those cubes.

When you look at the pieces put together, they can spiral clockwise or counterclockwise.  Of course, if you turn them over, the spiraling gets reversed.  For each possible spiraling direction, each piece can be oriented in one of 4 possible ways unless prohibited by the internal cubes.  If they were all possible, this would lead to 512 different combinations to try.

Sometimes you get lucky, sometimes you need to cold start your brain and think about it.  This time I felt that I got lucky.  I started with the spiraling direction that had a corner where the piece could only be inserted 1 way.  Everything just rolled along from there until I had all the pieces back inside in a reasonable amount of time.  It didn’t seem like a 15.18.14.4 effort, but they never do.  Moves get eaten up pretty quickly as pieces are moved within the frame.

It was then that I looked at the box that it came in.  One nice thing about Wood Wonder puzzles is that the boxes have a sticker on them with information about the puzzle including name, designer, moves required, and a picture of the puzzle.  Brian also sometimes adds the type of wood used to make the puzzle as well.  It was at this point that I realized that the 2 toned pieces in the picture made a nice symmetrical pattern in stark contrast to my avant-garde solution.  Duh!  Does that look like a wheel?  Phooey!  Now I had to update the picture on the box.

Needless to say, I was happy for an excuse to spend some more time with this puzzle, and not having solved it seemed like a good enough excuse.  It turns out that there are 8 solutions if you ignore the color pattern, which contributed to my quick success.  Adhering to the color pattern, there is a unique solution.

To achieve that nice wheel pattern, the 4 pieces are made from 2 contrasting wood types: Plum and Mulberry.  Since the pattern on the ends of the pieces is important, this required additional work to avoid the ends of the connecting pieces from becoming visible.  Brian also took extra care with the joinery to avoid butt joints that would result in weaker pieces.  At first glance, they look the same, but if you pay attention to the coloring, one is indeed different. 

Wheel Pattern? - Not So Much!

At the beginning of Wheel Lock take 2, it was obvious that the pieces could only be oriented with each other in one way.  The only question is whether they went in right side up or upside down (i.e., spiraling clockwise or counterclockwise with respect to the frame orientation).  In either orientation, it is obvious where the unique piece goes.  For the frame orientation that I was using, I decided to try the clockwise piece orientation first, which turned out to be the correct choice.

Adding the pieces to the frame is a process of determining the order that the pieces need to be added and the movements required to get them in place.  To start this process, I looked for the best set of 2 pieces that was the most difficult to put in the frame together.  It took me some time to identify these 2 pieces and determine how to get them in the frame.  Once the interaction of those 2 pieces was understood, adding the third and fourth pieces was an effort, but since there were already 2 pieces in the frame, movement options were greatly limited.  It turns out that the third piece that I added needed to go in last and the last piece that I added needed to go in first.  Although one of the pieces had a tendency to rotate within the frame, the solution only requires rectilinear moves.

In the past, I used to look at burr puzzles that required a color pattern to have a unique solution as flawed designs, a cheat added by designers to solve the problem of multiple solutions.  However, after thinking about it recently, if there is a color pattern, nobody in their right mind (I knew I would find a way to exclude myself) would spend time searching the non-pattern problem space.  I found Wheel Lock to be a superb design and thoroughly enjoyed the solving process - twice.  My only disappointment was the limited use of the slots in the fame.

Happy New Year!

Today is the first day of the new year and I hope that you have all resolved to do more puzzling this year.  With any luck, you have received many new brain twisters over the holidays to get you started for 2020.

The subject of this first post of the new year is named Happy New Year and designed by Johan van de Konijnenberg.  It has a unique level 16.2.4.3.4 solution and was released by Cubicdissection in July 2013.  Wow! That was more than 6 years ago and should give you some insight into the appalling state of my puzzle backlog.

Unlike the ultimate Penultimate Burr Box Set challenge puzzle described in the Christmas post (A Christmas Present For You – Ultimate Penultimate Burr Box Set Challenge), Happy New Year was specifically designed to be a challenge.  You immediately expect it will be more challenging with 16 moves to take out the first piece instead of 1.  As intimidated as I was by the ultimate Penultimate Burr Box Set challenge, I was even more intimidated by Happy New Year.  Normally, I don’t find puzzles that require 16 moves to take out the first piece daunting but this one had more than 8 pieces, which compounds the level of difficulty.  I figured that it would take me several days to put back together once I had it disassembled.

Happy New Year and Sweet Sixteen

With the New Year deadline looming, I took it apart, let the pieces sit idle for a while, reassembled it, and wrote the post all in plenty of time.  After the euphoria of finishing so far ahead of the deadline wore off, I realized that I had the wrong puzzle.  As it turns out, I had just worked on Sweet Sixteen.  You really can’t blame me for mistaking one for the other.  Although the solutions are very different, they are both high-level burr puzzles released by Cubicdissection in July 2013, they both use the same 3 types of woods, the pieces are the same size with the same finish, they have nearly the same number of pieces (11 vs 10), and they both require 16 moves to take out the first piece.

Armed with the correct puzzle and a diminishing span of time to the deadline, I started disassembling Happy New Year.  The puzzle is not difficult to take apart but I was surprised that it didn’t start falling apart after the first 4 pieces were removed.  The remaining 6 pieces hold together quit well and come apart in an interesting manner.  Once it was all apart, I scrambled the pieces and left them alone for a while before tackling the assembly on another day.

The assembly was very interesting and followed a logical progression.  The piece locations can be determined without too much difficulty, making the crux of the assembly determining how to get the pieces where they belong.  It turned out that it was fairly easy to get 8 of the pieces together, with only 2 remaining to be added.  Of course, it turned out that the remaining 2 pieces were not the last 2 pieces that needed to be added.  They were part of the stable 6-piece core, which made holding all those pieces a fiddly balancing act without them.  I was able to disassemble the 8 pieces until I only had 4 remaining and could see how to add the 2 pieces to make the 6-piece core.  Adding the remaining pieces was relatively simple after that.

As you would expect from Cubicdissection, Happy New Year is well made and the tolerances allow the pieces to be easily moved.  The puzzle is very attractive and made from Maple (2 pieces), Cherry (4 pieces), and Walnut (4 pieces).  The pieces have a very small bevel on the ends and none along the length of the pieces.

This was a great design, and I thought that the 6-piece core made it really interesting to solve.  The use of the 3 types of woods keeps the level of difficulty from getting too difficult and allows the piece locations to be determined without a lot of effort.  If you’re interested in higher level burrs than the 6-piece burrs, this would be a good one to try.

A Christmas Present For You – Ultimate Penultimate Burr Box Set Challenge

 

Merry Christmas!  Hopefully, Santa left you many puzzles under the tree.  And if you don’t celebrate Christmas, a Happy Whatever You Celebrate to you!

It’s been 6 months since the Penultimate Burr Box Set was released by Cubicdissection.  I figured that most people have already solved all 898 of the puzzle challenges by now and that the sets were starting to collect a layer of dust.  Instead of generating hundreds of additional 6-piece burr challenges, I decided to determine if I could find a burr puzzle that required more than 6 pieces.  In essence, I was looking for the ultimate Penultimate Burr Box Set challenge.

The picture above shows the result of that search.  This shape can be constructed with 2 different sets of pieces, each with a unique solution and a 1.2.2.2 level of difficulty.  However, the only difference between the 2 sets is the substitution of one piece and the substituted piece is only different by 1 cubie.  In effect, there is really only one puzzle.  The pieces used are deferred to the bottom of the post in case you wanted to figure out which 8 of the 27 pieces are required to build the shape as part of the solving process.  For the sane readers, just keep reading.

To come up with this bonus, I experimented with several shapes using BurrTools until I found this one.  Here is a timeline of my experience testing it:

• 0 minutes: With 8 pieces, this looks like a daunting task.  Complex Burrs with more than 6 pieces sometimes use different colors for pieces based on their orientation to help the solving process.  Of course, all these pieces are of the same type of wood and nothing special has been done to aid the solving process.  Then again, this puzzle wasn’t specifically designed to be difficult by an evil (I use this word in the most affectionate way possible) puzzle designer.
• 5 minutes: OK, I think that I know which pieces are the 4 that go together, which I will refer to as frame pieces, and the 4 that cross each other inside it, which I will refer to as cross pieces.  I already have 7 pieces in place.  How hard can it be to add 1 more?  I’ll probably have to shuffle some pieces around, but it shouldn’t be a big deal.  I’m also pretty sure what the last piece to be added is.
• 10 minutes: I now have 7 pieces where it looks like the last piece would fit inside.  However, there are 48 possible assemblies and only 1 can be constructed.  After some fiddling around, this one doesn’t seem to be the lucky one.

• 

  Wrong Orientation!

  15 minutes: I’ve got them all together!  Wait a minute.  They aren’t in the correct orientation.  However, I’m pleased to have found another symmetric target shape.  I decided to check out my clever new shape in BurrTools.  In seconds, BurrTools started laughing and informed me that there at 92 different ways to construct that shape with those pieces.  92 different ways!  If you dropped the pieces on the table they would probably fall into that shape.  OK, so let’s call that the easy, warm-up objective.
• 1 hour: Uh, maybe all this trial and error isn’t the way to go.  I keep doing the same thing over and over.  Time to break down and start thinking about it.  There are 48 different ways to put the 4 frame pieces together and then there are 24,576 ways to place the 4 cross pieces inside, resulting in 1,179,648 different combinations and that is assuming that I was correct in the division of pieces.  Given these numbers, most puzzlers take the lazy way out and start thinking about how to solve the problem.  It seems that 2 of the cross pieces have to work together and this appears to lock up the 4 frame pieces where you can’t add the other 2 cross pieces.
• 1 hour, 15 minutes:  Solved!  For me, the trick was figuring out what the second to last piece to be added was and arranging all the other pieces so that this move could be accomplished.

Is this a good puzzle?  Yes and no.  It was a lot of fun to work on and well worth doing.  If you have the Penultimate Burr Box Set, definitely pull it back out, dust it off, and enjoy this ultimate challenge.  It’s a great puzzle!  If you don’t have the set and someone offers to sell you this puzzle, don’t buy it.  It’s a terrible puzzle!  This puzzle was not specifically designed to be a clever 8-piece burr puzzle and would be scoffed at by any discerning puzzle collector.  It’s fine as a puzzle challenge in a large burr set, but not good enough to justify as a standalone puzzle.

Any excuse to pull out the Penultimate Burr Box Set is a good one.  The set is beautiful and well crafted.  The pieces are spot on and a pleasure to play with.  Cubicdissection did a fantastic job making these sets and I always enjoy spending time with it.

The pieces required to make the ultimate Penultimate Burr Box Set challenge puzzle are: 5, 6, 7, 8, 9, 10, 15, and 12 (or 17 instead of 12).

The Marriage of a Great Design with Great Craftsmanship ‒ Bouquet

When I saw Bouquet, my heart beat a little faster and said “buy, Buy, BUY”.  My wallet said “Whoa Cowboy, puzzles don’t grow on trees you know.  Do you really need another Christoph Lohe puzzle or another puzzle made by Brian Menold?”  What a stupid question!  Of course I do!  And puzzles do come from trees, so there.

I first saw Bouquet at this year’s Rochester Puzzle Picnic (A Decade of Puzzling - RPP 2019).  I only spent a few minutes with it as the event was wrapping up and failed to purchase a copy from Brian before he left.  I’m fairly certain that there aren’t many puzzlers out there that have not regretted passing on an opportunity to acquire a puzzle.  After returning home, I realized what a horrible mistake I made and immediately contacted Brian to acquire a copy.  Luckily, there was one still available.

I really enjoy Christoph Lohe’s designs and have twice in the past selected one of his designs for my top 3 puzzle acquisitions of the year.  Tulip made by Cubicdissection in 2015 and Mimicry made by Pelikan in 2016.  This year it looks like it will include Bouquet made by Brian Menold at Wood Wonders.

Not only is Bouquet a beautiful puzzle, it’s an excellent puzzle.  Brian made several different versions of wood combinations and the one that I chose was made with a Wenge frame and Maple and Paduak pieces.  I thought that this color combination enhanced the image of a flower.  I know that the name is Bouquet but it reminds me more of a corsage with the 2 colors used for the pieces.

This puzzle has a 23.3.5.3.3 level of difficulty.  Taking it apart is not difficult and was a lot of fun figuring out how to manipulate the pieces and remove them from the frame.  It’s amazing how such an open frame can hold the pieces for so many moves.  There is a lot of empty space in the center of the puzzle allowing for a lot of movement of the pieces.  As a general rule, I try not to pay too much attention when taking a burr apart so that I can enjoy it more when putting it back together.

To thoroughly enjoy the assembly process, I left the pieces sitting around for a couple of months before attacking the reassembly.  When I pulled them off the shelf, my initial reaction was that maybe I made a mistake.  I wasn’t sure how the pieces sat in the frame and I had to resist the urge to look at an online picture of the puzzle.  As daunting as it initially looked, it didn’t take long to figure out.  Now all that was left was to figure out which piece go where, what order to add them, and how to move them.  Oh, and its going to take about 23 moves to add that last piece in.

The first thing to realize is that 3 of the pieces are right-handed and the other 3 are left-handed.  This is easy to track in the 2-color version since all the right-handed pieces are one color and all the left-handed pieces are the other color.  For versions that use 6 different types of woods, you have to work a little harder to keep track of them.

For some puzzles with a frame, it is easiest to determine how the pieces are oriented with each other before inserting them within the frame.  This is not one of those.  For me, the best way to tackle this puzzle was to add pieces to the frame one at a time using the frame to hold them in place.  As each new piece gets added, you have to figure out how to manipulate the other pieces to get the new piece in.  Along the way, you may have to reorient a piece or switch the locations of some pieces.  This process is not as bad as you may think since 2 of the Maple pieces are identical and the 3 Paduak pieces are similar to each other. 

I was able to get 5 pieces in fairly quickly, but the last one eluded me.  I had already discovered how the pieces could be manipulated to that last one through the heart of the puzzle if I could only get it started.  I was even able to pull another piece out and test that process.  I just couldn’t figure out how to get both those pieces in at the same time.  Obviously, neither of these pieces were the last to go in and it took some more experimenting to determine which piece that was.  It turned out that the last piece to be added was the one that I put in third and the second to last one was the one I put in first.

This puzzle is another masterful design from Christoph Lohe and I thoroughly enjoyed the solving process.  It generated a great feeling of satisfaction and confirmed that some of the pistons in my old clunker of a brain are still firing.  I’m really glad to add this masterpiece from Wood Wonders to my collection.

Half A Dozen Rhombic Dodecahedrons - Cluster Buster

For puzzlers used to working with cubic dissections, it may be a bit daunting to be presented with a puzzle based on rhombic dodecahedrons.  Rhombic what?  To keep it simple, a rhombus is a diamond (all 4 sides of equal length), dodeca means 12, and hedron means a shape bounded by some number of planes.  So, a rhombic dodecahedron is basically a 12-sided object with diamond shaped sides.  The rhombic dodecahedron also has the nice property that if you have a lot of them (at least a cluster), they will pack together with no empty space between them.

Cluster Buster and Truncated Cluster Buster were designed by Stewart Coffin in 1973 and described in his book, Puzzle Craft, 1985.  Both consist of a cluster of 6 rhombic dodecahedrons that has been dissected into 6 identical pieces.  The truncated version has the 6 outer tips cut off to provide a square face on the outward facing side of each rhombic dodecahedron.  The difference between the two is cosmetic and both operate the same way.  The two objectives are to take it apart and to put it back together.  You might be thinking, Duh!, but these challenges are quite a bit different.

Cluster Buster has 1 sliding axis and you need to position your fingers in very precise positions to pull it apart.  This is more difficult than you would imagine with a well-made version of this puzzle.  There is an alternate method of disassembly, but I wouldn’t recommend it for a nice version of this puzzle.

Getting it back together is an entirely different matter.  It isn’t difficult to determine how the pieces should be oriented for reassembly but trying to hold 6 pieces with two hands makes this an interesting dexterity challenge.  In addition, the pieces have to be perfectly aligned to get them reassembled.  A lot of Coffin designs seem to have this theme.

My version of Truncated Cluster Buster is a 3D printed version from Pacific Puzzleworks that was a gift from John Rausch of Puzzle World.  For a 3D printed puzzle, this one is well made.  However, it is quite a bit easier in both the disassembly and assembly compared to a well-made version in wood.  The rounded edges of the pieces provide the ability to get a hold of the edges instead of having to only used the faces when puzzling it apart.  That combined with the frictionless surfaces of the PLA plastic allow the puzzle to be quickly disassembled.  Similarly, the rounded edges make assembly a bit easier since you don’t have to be as exact in aligning the pieces before putting them together. 

The 3D printed version is a very nice puzzle, but if you want the full experience, I recommend finding one of high quality made in wood to play with.  However, if you’re planning on buying one, a nice one made in wood is going to cost you a lot more than a 3D printed one.  The one from Pacific Puzzleworks is available on Etsy here.