Neil's Puzzle Building Blog
12Jul/112

New Puzzle Designs

While I've not had any time to get back to my workshop in the garage and spend time finishing off my current puzzle build, the Matrioshka (read part 1 and part 2) I have been busy. I've had a number of puzzle designs floating around in my head, and in the odd spare minute here and there I've been sketching things down and trying to make some progress. You'd be amazed how many ideas float through your head at 2am when you're wide awake.

Since a good friend poked me in the direction of creating my own puzzles, I have had a myriad of ideas floating around in my head, each one probably more crazy than the last. The problem is that having an idea in my head doesn't really do anyone any good. It was time to try to get a few things down on paper and see if there was some merit to any of them.

As you know from early posts here, my first puzzle design was cube based, and it created a nice puzzle in my opinion. (More info on that coming soon!) But I wanted something a little more interesting. I have been reading a lot about puzzle design, and Stewart Coffin's work is an excellent learning resource. He challenges you to take what he has done, and learn from it. He doesn't give the solutions to the puzzles he's designed, far from it, he challenges the reader to find them himself. Having picked up a copy of one of his books from Amazon, and it was clear that in the world of polyhedral dissections, there is a lot more scope if you move beyond the cube.

Rough Sketches of a new puzzle design

Rough Sketches of a new puzzle design

I don't want to make a copy of any of STC's designs but I did want to see whether I could create a novel dissection. My starting point was a star, as in the sketches above. My idea was to create an extruded six pointed star with an interesting interlocking set of pieces. I played for a while with this idea, but realised that the dissection would lead to some very interesting cuts to create the pieces needed, and I abandoned that for a simpler 8 pointed star, where all the angles are 45 degrees. Ok, so simpler may be relative, however I did find it easier to work with!

The sketch in the middle is a rough of the dissection, where the puzzle is 6 units wide by two units high. With those base dimensions, I had enough freedom to create an interesting assembly, without it being too complex to create. Remember I am doing this by hand, not using a computer to help!

My first challenge was to visualise how you could dissect the shape. The nice part about working with the 8 pointed star is that each of the points is a cube, and cubic dissections are fairly straight forward. Using this as a starting point, I looked at creating 4 pieces which I could create some sort of locking structure around.

The first version was fairly simple, and went part way to what I wanted. I had created a two pieces which would join together such that movement in two of three dimensions was restricted. Creating a set of two such pieces created the first version of the puzzle.

Basic Star Puzzle Solved

Basic Star Puzzle Solved

Two joined pieces of the puzzle

Two joined pieces of the puzzle

You should be able to see from the image on the right, that there is nothing to stop the two halves from sliding apart, so really this isn't a great puzzle. The reason I show it, is that it may give you an idea of the stages I went through in creating the puzzle, and if you're thinking about creating your own puzzles, then hopefully it will be of use to you too.

The next stage of evolution of the puzzle is to create a version where once assembled, the four pieces don't just slide apart on their own. Some sort of manipulation is required by the solver. The added benefit to this is that it's now more challenging to solve. To do that I started to look at ways to create a joint between the two halves I'd created. I was looking for something like a dovetail joint, and if you look at the sketches at the top of this post, my version of that is in red pen on the top of the left hand sheet.

Creating the two halves of this joint, I created a way for the two sub assemblies to be slid together to create the final shape.

Sliding Star V2

Sliding Star V2

So you may be asking why I've labelled this "Sliding Star V2". The answer is quite simple. It's the second version of the puzzle. Once I had created the basic pieces on paper, I turned to Burr Tools to model them on the computer, for two reasons which I'll get to ...

Many people turn to Burr Tools to find how to put their puzzle back together, or to create new puzzle pieces. The program has many many uses, and I'm no expert, but I am learning and getting much better the more I use it. I didn't need to use the tool to find out how to assemble the puzzle, as I already had worked out on paper that it did. What I really wanted was to verify the assembly, and check how many assemblies existed with my pieces, and also because I wanted a model file that I could export for 3D printing!

Coming back to the reason for V2, when I originally designed the dovetail ends, I envisioned the puzzle as being two halves that slid together along the yellow blue axis in my rendering. What Burr Tools showed was that one of the pieces could slide diagonally out of the assembly, and in fact didn't lock the puzzle together as I had drawn.

Here the power of the tool allowed me to see something I hadn't seen on paper, and allowed me to rework the pieces to prevent the unintended movement. Burr tools also confirmed only one solution for the V1. For the V2, when I had redrawn things, and re-entered them into the tool, I was surprised to find 2 assemblies! So my small change allows two pieces to be swapped, and still assemble the puzzle. All very interesting, and very unexpected.

With the puzzle pieces confirmed as being possible within the constraints of Burr Tools, I set about exporting the pieces, so that I could have them 3D printed. This is a very quick way to prototype puzzles, which is far more powerful that pen and paper, or even computer modelling. Seeing a puzzle in real life, and playing with the pieces can give you much more feedback that looking at it in two dimensions.

Now I'll be honest here, Burr Tools isn't the most intuitive for exporting files, but with a lot of patience, and some guidance from Derek Bosch, I was able to export the pieces in a format which would work for sending to Shapeways. (Which I have done, and I'm waiting for the results!)

Now here's the really interesting thing about the export from Burr Tools. If you export from the Solution tab, the stl files are created with location information. Which means that when you export them, they are in the position and orientation of the solved puzzle. So it means importing them into you 3D package of choice, you can render the solved puzzle, without painstakingly moving the pieces by hand. As a bonus side effect, it also shows you the tolerances between pieces, which is ideal for knowing how well things will fit when printed by Shapeways! If you're wondering, yes this is how I created the images of the puzzle above in PovRay.

Once I receive the printed parts from Shapeways, I'll update as to how things turned out.

With version two created, I turned my attention to the original goal of the project which was a co-ordinate motion version of the puzzle, using the same shape, but modified pieces to make a far more challenging puzzle to assemble. I'll give more info on that in the next post about my puzzle making progress.

Finally, I mentioned a few posts back that I had a big project I was working on. Well I'm still working on it, but no more details yet. It is still at the pen and paper stage, and there's a lot of work to tweak the designs before I take that further... Watch this space!

Filed under: Uncategorized 2 Comments


8Jul/112

Bolted Closed

Bolted Closed is a take apart puzzle from Bits and Pieces, which I came across on the Puzzle Master website and really liked the look of. The contrasting woods, and use of different metals makes this puzzle really stand out. Thanks to Puzzle Master for sending it to me to review.

Bolted Closed by Bits and Pieces

Bolted Closed by Bits and Pieces

The object of the puzzle is to remove the large ball bearing from its prison between the brass rods, wooden supports and the nut and bolts. The choice to stain the wood in the centre portion into a bright red helps to make this puzzle stand out and really makes you want to pick it up. When you do, you realise it's a fairly heavy puzzle thanks to that ball bearing which is rolling around much more freely than photographs made me think it would.

The puzzle comes very simply packaged in a white cardboard box with the name of the puzzle (listed on the box as Brass Bolt), and some warnings about not swallowing the small parts. Safety information over, I opened the box and removed the puzzle, to have the ball bearing fall out, hit the floor and roll under my sofa. Not exactly the sort of challenge I was hoping for. The main reason the puzzle arrived in the already solved state is I think mainly down to the packaging of the puzzle. With no support or padding it's all to easy for the mechanism to become free during shipping.

Recovering the ball from under the sofa, I put things back together, opened it a couple of times and thought it was almost a trivial mechanism, with no real puzzle element at all. Rather disappointed, I reset the puzzle, and put it on one of my puzzle shelves thinking it was a pretty poor puzzle. Coming back to it some months later to write the review I was able to solve the puzzle as it was intended, and it has gone up slightly in my estimation. I really shouldn't have judged it based on what may have been a fluke when I unboxed it.

In coming back to the puzzle and trying to solve it fresh, I found that it didn't open as easily as it had originally. As it turns out, when I closed the puzzle initially, I had only partially done so. The motions of putting it on the shelf and the taking it back down to photograph, had fully reset the puzzle, making it a slightly better challenge. Puzzle Master ranks this as a Level 8 - Demanding (out of 10) puzzle. I don't think I'd rank it quite as high, possibly a 7.

The mechanism itself if fairly simple, and most people should be able to open it with a bit of time spent on it. The ball rolling around helps to disguise any mechanism at play so it would be easy to miss the first move to removing the ball as a result. Rule 11 as it is fondly known does play a part, but unlike some other puzzles, the amount of force required is small, and there's no worry about hurting yourself here, unless you get your fingers stuck between the ball and the brass rods.

My biggest issue with this puzzle really comes down to how well it's made. Ok, it's not an expensive puzzle, and it does look great. Many other cheaper puzzles look cheap and this doesn't suffer that fate. If you look closely though, the fit and finish isn't all that high which lets the puzzle down slightly, but has to be expected given the price range.

The gap between the support post for the bolts and the base is large enough that you can easily see the mechanism keeping it closed, and on my version the two bolts are actually bent into a slight curve. The result being that getting things lined up to remove the ball have to be spot on or the ball will remain just trapped. I can't tell if this is by design, or not but I have to think it's unintentional. By the time you can get the ball close to being removed, the hard work is done, and the very slight bend seems incongruent with the solution.

For a beginner, this is a good puzzle. The ability to get a hint as to the mechanism will keep newer puzzlers interested and it does look good. I'd have to say, give this one a chance, and you may just enjoy the puzzle. If you're really stuck, the solution is included, but I'd leave that in the box until you've at least had a play.

Tagged as: , 2 Comments


5Jul/112

Blog Updates

After the long weekend here in the USA, it seems I didn't do much puzzling (or any to be fair), but I did start sketching out some designs for a new puzzle box that I'd like to make. Don't get too excited, it's unlikely to see the light of day any time soon as it's a big project, and will take me a long time to realise. If it does happen though, I hope it will be something rather special.

I put in a shapeways order over the weekend, for the first prototype of a 3D printed version of my Cube puzzle. This version is a little expensive still as I have kept the internal pieces with a solid wall, so it looks as close to the Lego version as possible. If it works out then I'll consider making one with mesh internal pieces to reduce the cost significantly so it could be purchased if you wanted one. At some point when I finish the Matrioshka that I'm trying to build, I'll make the cube in wood, but looks like that will have to wait till the end of the summer before I can even start on that project. Too many things to do, and not enough free time!

As far as the blog goes, thanks for all the feedback you've given me. It's appreciated, and it is good to hear from everyone. Today I upgraded to the newest version of WordPress, and I hope that will clear up some of the Errors that I know one or two people have reported seeing. I've never reproduced them myself and can't narrow it down to a problem at the server side, but I know it's an issue. If you've been affected then I'm sorry, and all I can suggest is to try another browser (Firefox seems to be the one with the issue).

I've added a new feature to allow you to rate the reviews, and would love to know what you think of that. Also if there's anything else you'd like to see added to the site, or even if you have puzzles you'd like me to review, then let me know and I'll see what I can do!

Filed under: Random Musings 2 Comments


4Jul/110

Chip by Hiroyuki Oka

Chip was created as a Christmas Present in 2007 for the Karakuri group members by Hiroyuki Oka. It's name comes from the small 'chip' of wood which is stuck to the top of the box by a magnet. This is the last of the Puzzle boxes Derek Bosch recently lent me to solve.

Chip by Hiroyuki Oka

Chip by Hiroyuki Oka

Here is what Hiroyuki Oka has to say about this puzzle box:

This is one of the Secret Boxes. At first, you need to move a device to open the box.. But you can’t see the mechanism from the outside. Maybe you can find the place of the device with the attached small wood plate.

Made from Walnut, Katsura Agathis, Rengas and Magnolia, there's a lot going on in this box. The inlay is nicely done, and adds detail to an otherwise plain box. The checker board chip made from a number of small cubes laminated together is a nice touch as this didn't have to be such an ornate part of the puzzle. The box is a slightly squashed cube measuring approximately 3.5" x 3.5" 2.75".

As the description hints, there is a hidden mechanism in this box, and the small chip attached to the box is the clue to finding it. There's nothing hidden about the fact that the chip is held in place by a magnet as its visible when you flip the piece over.

Initially, none of the panels on the box will move at all, so whatever is keeping the box locked needs to be moved before anything else will happen. Having had a few other puzzles in the past where magnets have been used, I tried all sorts of ideas like flipping the piece over to repel the magnets, hoping this might move the device and let me open the box, but I was having little success.

With a closer inspection of the box, and a lot of searching I found what I was looking for and what the description hits at. After that it was a fairly simple sequence to open the box requiring only 5 moves.

I think I spent around 20 minutes on this puzzle to open it and while I do like it, it's not one of my favourites. The walls of the box are just a little too thin for my liking, to the point that the base has a degree of flex if you press it, even gently, with your fingers. (And most people familiar with Japanese puzzle boxes will use several fingers and press gently while pushing panels to see what moves.)

A nice box, and I am grateful to Derek Bosch for lending it to me so I could solve it.

P.S. Happy 4th of July to all my friends reading in the US. American flag



3Jul/111

Free Dial by Shiro Tajima

Free Dial by Shiro Tajima was his Karakuri Group Christmas Present in 2005, and is another of the puzzle boxes Derek Bosch recently leant me.

Free Dial by Shiro Tajima

Free Dial by Shiro Tajima

The description of this simple looking box is anything but plain. I love reading the descriptions the designers give their puzzles. So often it reveals either something about the puzzle, or the designer.

Actually in this tiny box is loaded the enchanted power! Try! Let's turn the red dial in the front of the box. What happened? Probably it will be fantastic things. You might have a romance, might receive a message from somebody that lives far off in the Universe, or...?

Made from Japanese Raisin tree, Chanchin and Walnut, this box measures approximately 2 3/4" x 2 1/2" x 2 1/2". The big red button on the front is fairly appealing, but no matter how hard you push it's not going to depress. When Derek gave me this box I assumed that the slightly wider plate was the top, however the Shiro Tajima's page on the Karakuri website shows it as the bottom. Personally, I think I prefer it as the top.

Free Dial by Shiro Tajima

Free Dial upside down

I'm giving nothing away by telling you that the button will spin fairly freely, and that the plate just slides off. There's no mechanism, no lock, nothing clever here. It just slides right off.

Free Dial by Shiro Tajima

Free Dial by Shiro Tajima

Unfortunately that really doesn't help much. You're left looking at a plain top to the box, and you can see the very top of the red dial. After playing on and off with this box, I wasn't having a huge deal of success. That said, I had a fairly good day of solving puzzles, having finally cracked the Box with a tree so I had picked this back up that night, and in an 'Aha' moment, saw how to solve this one. All in all I spent around 30 minutes puzzling over this box.

It's a beautifully simple puzzle, and as with most puzzles, you have everything you need in your hands to open it, if only you can think enough outside the box to see it. In this case, thinking on top of the box may help as well. The thing I like about this puzzle is that everything is on display. There is nothing hidden, so no clever tricks or hidden mechanisms that you have to feel your way around.

The opening of the box is quite special and rather unexpected. It could easily elude you unless you pay close attention to what you see, and for that reason I really like it. This is a great little puzzle and I highly recommend playing with one if you get the chance. Given that this was a Karakuri Christmas present, there aren't that many around, but it may appear at auction occasionally, so keep your eyes open.



2Jul/110

Oskar’s Matchboxes

Oskar's Matchboxes is another puzzle designed by Oskar Van Deventer. This seems like a simple enough puzzle, where five matchboxes have their sleeves and drawers attached at interesting angles to create five unique pieces. The goal is to close all five matchboxes. Have a look here for Brian's review.

Oskar's Matchboxes Puzzle by Eric Fuller

Oskar's Matchboxes Puzzle by Eric Fuller

I recently won a copy of Oskar's Matchboxes, made by Eric Fuller in 2010, on the recent Cubic Dissection Marketplace auction. This is a beautifully made version using (as best I can tell) Mahogany and Maple. Overall, the puzzle is approximately 3.5" in size, making for a very compact version.

Matchbox Diagram

Diagram of each of the pieces

The diagram on the right shows how each of the matchboxes should be attached. The dimensions of the boxes are fairly important, they need to be on a 3:2:1 scale in order for the puzzle to work correctly. There have been a number of versions of the puzzle made, including short runs by Trevor Wood, Tom Lensch and Eric Fuller. Each of those has a different appearance as the creator can make the boxes in whatever style they desire, as long as it remains inside the dimensions listed. Recently Philos Games has started creating a mass manufactured version of the puzzle which you can buy directly or at Puzzle Master or Amazon.

When I received this puzzle, Eric's description states that there are two solutions, however I was aware that Trevor Wood claims there are three solutions. I set myself a challenge to find all three solutions or prove that there were only two. When I received the puzzle, it was already in the solved state, so I took it apart, shuffled the pieces and starting trying to put it back together. After about an hour I had the five boxes back together. This is a tough puzzle, and is very easy to get lost trying to solve it. A systematic approach will serve you best when trying to solve this, as randomly lifting pieces is unlikely to be successful.

With one solution found, I started looking for the others. After another 15 minutes, I found a way to change the location of only one piece and still solve the puzzle. Having done this, I had to take photographs of the two solutions to prove to myself that they were indeed different. The two solutions are mirror images of each other, so easy to miss that they are different. (You can click this link to see all solutions)

So two down. Could I find the third? As it happens, I had come across the third solution while trying to solve the puzzle for the first time, however it is not possible to create this third solution with Eric's version of the puzzle. At this point, you might think I'd give up and accept that it wasn't possible to create the third solution. Well, I'm not the sort of person that gives up. I wanted to create a version where all three solutions were possible. As Trevor Wood points out, the dimensions of the puzzle, and exact placement of the drawers is required to be able to create all three solutions.

One thing that made me want to create my own version was this image of the Philos Puzzle in the solved state (thanks to Brian Pletcher for the image). If you look at the image on the Puzzle Master site, you'll see that it is solved in the same way as I have it solved, but here is a completely different solution, this one much flatter than the two possible using Eric's version!

Off to the store I went to buy some matchboxes. (As a side note, you have no idea how difficult it seems to be to buy a standard matchbox in San Jose!) With a pack of 10 matchboxes in my hand I took out my tape, and started joining the sleeves and drawers together to match the diagram above, while creating an ever increasing pile of matches on my workbench.

Homemade version of Oskar's Matchboxes

Homemade version of Oskar's Matchboxes

The result speaks for itself (see the link below as I have not included the solutions by default so you can discover them for yourself!). The third solution which was not possible with the very high quality version from Eric Fuller.

If you want to see all three solutions, click here, and browse the images. Note, these won't appear if you click any of the other images so if you don't want to compare the solutions, you're not going to be exposed to them accidentally.

As an additional challenge, try putting the matches back into the boxes when you solve the puzzle. After all, they are matchboxes, and should still be able to store the matches. I went ahead and did just that, and realised that the puzzle becomes harder. Given the interesting orientation of the drawers, the order in which you close them now becomes even more important, as it;s not possible to put them together in all orders since the matches will fall out (unless you have 4 hands!)

This is a fun project, and fairly simple as long as you get matchboxes with the correct scale. Have a go, and let me know how you get on. Alternatively, pick up a copy of the Philos version and have a go at finding all three solutions yourself.