This puzzle is my first puzzle with multiple solutions. Tricky stack cutting was required.

Originally Carnival of Harlequin by Miro was going to be Puzzle #833. I then had a cutting disaster with that puzzle after almost a week of work, and ended up going ahead with Octastar as #833. Octastar too give me a lot of trouble - I had several major technical problems trying to make Octastar which required that I start over several times. Due to the string of problems, there was no 100 Puzzles Project puzzle in January 2010.

As it happened (from my Journal archives)

January 25, 2010

Monday 11:34 pm - Here are some design notes for the new #833 (replacing the pushed back Carnival of Harlequin), Octastar.

January 27, 2010

Wednesday 5:19 pm - I was at the shop today doing some experimental cutting for the new #833, Octastar. I thought I could get away with double cutting just two boards. I learned that the logistics for cutting this little puzzle will be quite a bit more complicated. Looks like each of the eight external triangles will have to be cut as four pairs of boards, then eight interior sections somehow knit together.

There is no way I can get the puzzle done by the end of January. Thus, I've pushed the 100 Puzzles Project back a month, it is now scheduled to culminate in May of 2010. The last puzzle of the Project and most spectacular puzzle I may ever make, may take an extra month. So at a deliberate pace, I'll get Octastar done by the end of the first week in February, launch the auction, and then go right into cutting #98 (the Carnival of Harlequin remake). I should easily get that one ready for an early March auction launch and then the big #99 for an early April launch. The rest of April and likely part of May will be spent working on #100.

I'm now thinking that I'm going to put some kind of geometric pattern print on the wood as an image, instead of painting the pieces, a design yet to be determined. There are a LOT of design issues yet to be worked out.

January 28, 2010

Thursday 6:04 pm - No puzzle work today. The new print for #834, Carnival of Harlequin arrived, looks perfect. Tomorrow I'll resume design work for #833, Octastar, and prepare the board for #834. I'll drop off the board and the print for #834 for professional mounting early next week.

January 29, 2010

Friday 4:18 pm - Today I worked extensively on the design for #833, Octastar. Making things more complicated, I've decided that I do want a print mounted on the wood. The alternative is to paint the puzzle after it is cut, something I want to avoid due to the nature of the design. I designed an ingenious pattern which works with what I want the puzzle to do; the execution will be involved as I will have to mount multiple small boards and then carefully choreograph the stack cutting in order to come up with one puzzle. I have designed the set of print templates required for this. This weekend I have a tango workshop so I'm not going to have much time for the puzzle. As time becomes available I'm going to cut a full prototype, to see what problems come up, if any, and then later, in any case, I will cut the official puzzle, as the prototype's images won't be be high resolution nor permanently mounted. The prototype will not be numbered. I've changed my mind again regarding cutting style, it will be Swirl Curl, not knob. I'm current projecting 120 pieces, this number could vary quite a bit. If I can pull this puzzle off, it might be very cool!

January 30, 2010

Saturday 7:00 pm - Today I prepared a board for #834, Carnival of Harlequin and eight (!) boards for what will become the prototype puzzle for #833, Octastar. I expect to figure out "manufacturing efficiencies" during the creation of the Octastar prototype. Tomorrow I'll mount image sections on the eight boards which will eventual become the prototype; on Monday I'll drop of the board and print for Carnival of Harlequin at the service bureau I use for mounting large prints. I may get some cutting in on Monday for the prototype, much more likely on Tuesday.=

February 2, 2010

Tuesday 6:34 pm - Went to the shop today and did some work on the prototype for #833 - the thing is complicated, still haven't worked everything out. Tomorrow I'm going to "digitally" work on the puzzle in Photoshop to try and solve some problems. More later. I expect to increase the size of the puzzle.

February 3, 2010

Wednesday 7:41 pm - I have now worked out a strategy for how to cut #833, Octastar. I can do it using just two boards with some very precise repositioning between each stack cut - a total of five stack cuts. The method for the first prototype, now abandoned, involved eight boards and was going to be way too complicated to complete. The first prototype process was useful as it revealed some flaws in my planned cutting paths, now corrected. I now plan to cut a second prototype before I start the official puzzle. I need to redo the Octastar imagery as the level of precision turned out not to be good enough during the first prototype cutting. I'll prepare the second prototype boards tomorrow, print the revised imagery onto production paper and mount the prints for a full dress rehearsal. I'm increasing the size of the puzzle and it will have around 200 pieces instead of around 120.

February 4, 2010

Thursday 11:20 pm - Today I designed two high precision images along with some stack cutting guidelines around the images, and printed them. I went to the shop and prepared the two boards required for this puzzle, #833, Octastar - the second prototype. I plan to mount the prints tomorrow morning. Today I dropped off the board and print for #834, the remake of Carnival of Harlequin for professional mounting. This job will be done by Monday at the latest.

This weekend I plan to cut #833 - the second prototype. While I'm calling this version the second prototype, if the cutting "works" and there are no problems, then it will be the official #833 puzzle. The boards and prints are to my highest standards. The puzzle is going to take a very, very high amount of precision to cut correctly - I will reveal the image only after the official puzzle is cut.

February 6, 2010

Saturday 12:22 am - Yesterday both glue mounts for the second prototype of #833, Octastar came out looking perfect. I also picked up the professional mount for #834, the remake of Carnival of Harlequin, it too, looks perfect. I'm almost tempted to move up Carnival as the next puzzle again, but will focus on Octastar. Today I have a mid-day appointment to have my car's windshield replaced due to rock damage, so this will greatly disrupt my puzzle making plans - I may get some work in, if not, then Sunday.

Saturday 6:04 pm - Today I spent about 3 1/2 hours at the shop, and did the five "stack" cuts. I spent a lot of time using the guidelines around the images to get the boards aligned as precisely as possible, then nailing them together, doing the cutting, going through this process for each of the five cuts. The work turned out to be perfect - I mean the resulting "cross board fit" is so precise for each of the five cuts, the images aligned perfectly, the result exceeds my best hopes! This is going to be a spectacular little puzzle.

Tomorrow I'll return to the shop and do the edges and then cut the individual pieces for each of the stack cut hunks, completing the puzzle. So this is officially puzzle #833, Octastar, the second prototype exercise being a total success. Tomorrow I'll reveal all, but here is a preview:

February 7, 2010

Sunday 1:36 pm - I cut the various pointed sections. And..... well what can I say? The sections fit perfectly but the puzzle doesn't work! Yeah..... big design error in the cutting strategy. Executed perfectly, but the wrong plan. So..... yes, this puzzle was "prototype #2" after all. Now abandoned. I now have a full understanding of the situation and will make another go at it. Is the 100 Puzzles Project jinxed? No. I will have no further pictures or commentary until this puzzle is done, in fact this is my last journal entry until the puzzle is done. When the puzzle is finished I'll tell all, here.

February 10, 2010

Wednesday 5:04 pm - Octastar has been delayed again. Meanwhile Octastar Prototype #3 is in progress.

On February 8th, I redesigned the cutting alignment guides for Octastar; I also designed a template for a hasty art piece for this Friday's Love, No Love art show at Mixed Media.

On February 9th, I mounted the two images for Octastar; I also printed the template for the Love, No Love piece and cut out the work (15 heart outlines of various sizes) and spray painted them.

Today I was at the shop from early this morning, working on Octastar. I did the five stack cuts resulting in 10 sections and then cut the edges of the 8 star points, so I could see how well the alignment was and how the puzzle really works. The puzzle works very, very nicely! But the image alignment is flawed. This flaw, plus the bottom stack cut board getting caught on the saw's table causing me to nearly lose a piece in the yellow area, plus a problem trimming the yellow star point, make the puzzle, again, not acceptable for the auction. However, I do plan to cut the full puzzle and it will be an official puzzle #833, Octastar Prototype and will go into my own collection.

Here are some pictures of the 10 sections of puzzle #833, Octastar Prototype.

I think this puzzle, when done to my satisfaction will be of great interest to puzzlers and collectors.

I now need to focus on getting the work for the Mixed Media art show done. This weekend will be taken up by a dance workshop. Next week I'll finish the cutting of #833 Octastar Prototype, then begin the journey for another attempt at Octastar. I have several ideas that came to me while cutting this version, which I think will make next one easier, decreasing the chances of errors. If the next puzzle fails, I will put Octastar aside and cut #834 Carnival of Harlequin - although this is a big puzzle, it has already been professionally mounted, cutting should be routine. Then in March I'd get back to Octastar.

February 11, 2010

Thursday 4:54 pm - Completed a rush-rush art piece Expressions of the Heart for Mixed Media's Love, No Love art how. No photo yet. Then two hours to help a friend who is moving out of town to get to the train station, seven bags, a logistical exercise that took patience to get done. I helped carry the bags onto the train, then they announced "doors closing in one minute", I had to scramble down some steps to a lower level of the rail car, and the doors closed before I could get off, then the train started to move! I had to run down two coach car lengths, find a conductor and get the train stopped, which I managed to do. Whew! I apparently should not have been allowed on the train in the first place. Didn't see any porters.

I think I've figured out a way to make Octastar from one board, this will eliminate one of the stack cuts. Tomorrow I'm going to do a quickie experiment to test this and I plan to complete #833, Octastar Prototype.

February 12, 2010

Friday 5:02 pm - I was at the shop for a couple hours today, experimenting with cutting Octastar using a single board - I cut quickly (that is without regard to aesthetics) and only part of the test puzzle. Works fine. I expect to cut the real Octastar on Monday. Dance workshop this weekend. Have to rush now to Mixed Media to help set up for tonight's reception which I'm going to miss for the most part. I've decided not to cut the Octastar Prototype puzzle after all, instead will cut another formal Octastar, for myself, later this year after the 100 Puzzles Project is done. Will update the Puzzle Queue to reflect this later.

February 15, 2010

Monday 10:45 pm - I have pulled all pictures of Octastar - I'll have new pictures when the puzzle is finished.

Monday 6:14 pm - Yesterday I printed the "final" print with cutting guidance for Octastar and today I mounted it. Tomorrow is "cut Octastar day"; I plan to cut #833, Octastar my schedule has been cleared.

I'm not quite sure how I want to auction Octastar. My 100 Puzzles Project rules say that the price set by the auction will be the minimum price for subsequent puzzles of the same image and size. But with the recent high prices, the price for this puzzle may be too high for other people to purchase it, should I want to sell copies in the future. An exception to this rule is that I can sell "mini" version at a lower price, for example I have done this with my Geometric Pattern Puzzles. The question is, "how much smaller is mini?". The problem is that this puzzle is already not that large. No decision reached at this time. I'm currently expecting to auction the puzzle starting this Friday.

February 16, 2010

Tuesday 7:22 pm - MAJOR all day marathon at the shop today - finally there is a puzzle #833! After the January disaster with Carnival of Harlequin and repeated attempts to cut Octastar, Octastar is done. Not perfect - there are issues, but for my first real trick puzzle, an impressive effort. The puzzle is still at the shop, needs to be sanded and a few non-serious voids fixed. I upped the projected piece count from 200 to 250, will have the final count later. Pictures tomorrow.

February 17, 2010

Wednesday 5:03 pm - Introducing Octastar - a brilliantly designed puzzle (if I may so say) with multiple solutions!

There are other solutions, but they are in one of the five unique shapes shown here with other color combinations.

Today I inspected the pieces, repaired a void and now I'm going to try to put the puzzle together! Preliminary piece count: 253.

I will be launching the auction for this puzzle on Friday. For the first time in the history of the 100 Puzzles Project, I'm going to make an exception to my pricing policy. Regardless of what this puzzle goes for on eBay, I reserve the right to make copies of this puzzle and sell them at a price lower than what it goes for on eBay if the eBay price is stratospheric. This will be the only exception, the remaining three puzzles in the Project will adhere to my standard policy. Because I am semi-retired from puzzle making, going on to almost fully retired, and due to the weeks I spent designing and figuring out how to make this puzzle, I've set the price for copies at [EDIT February 18th - I am not decided on the price]. At this time I am not accepting orders, and might not ever. I do plan to make a copy for myself after the Project is completed, and will be bringing to the Puzzle Parley this August - not for sale, but for play!

I'll have more to say about this puzzle when I launch the auction on Friday.

February 19, 2010

Friday 2:36 pm - Auction launched. Final piece count 254.

February 22, 2010

Monday 8:29 am - I mentioned there are five general shapes in which Octastar may be solved. This morning I realized there are six; I've updated the auction description.

There have been many geometric puzzles over the years, decades, even centuries, for example the very famous Tangram Puzzle. And of course there have been many jigsaw puzzles over time. I'm wondering through - I don't ever recall an interlocking jigsaw puzzle designed to go into different geometric shapes via the trick of stack cutting. Is mine the first? Feedback on this subject is welcome.

February 23, 2010

Tuesday 12:33 am - With Octastar now already bid up to $775, that puts it well beyond what I think most people would be willing to purchase it for as an ongoing Custom Puzzle Craft product. I've received more email regarding this puzzle than perhaps any other, so chances are quite good that I'll come up with something similar but not the same to offer sometime in the future. No idea when at this time.

February 24, 2010

Wednesday 6:17 pm - No bids today for the Auction (100 Puzzles Project #97) so far. Auction ends in two days.

February 26, 2010

Friday 7:11 am - Auction ends today.

February 27, 2010

Saturday 12:37 am The Auction is over, I continue to deeply thank you for bidding!

Saturday 2:25 pm - A tsunami advisory was issued for the San Diego area after the big earthquake in Chile. Just slight tidal variations were predicted, no warnings were issued. The tsunami time was predicted for 12:02 PM. I went to my favorite observation spot in Coronado. The surf was small with the average wave reaching an old concrete structure at 11:50AM. 12 Noon - water was the furthest away from the structure. Water started to come around the structure at 12:03 PM; peaked around 12:12 PM (water rising maybe 6 inches); 12:19 PM - back to normal. Virtually no one else on the beach had any idea anything was supposed to happen or did happen.

From the auction description:

The 100 Puzzles Project is nearing completion! Note: Due to major problems with what was to be a 1300 piece puzzle for January, the project was delayed a month. The cutting of the January puzzle will be restarted soon, the auction is planned for March.

This month features my first "multiple solution" puzzle, Octastar. I've been wanting to make a trick puzzle for a long time and I finally got around to it! This trick puzzle uses the puzzle cutting technique of "stack cutting" whereby two sections of the puzzle are "stacked" and fastened on top of each other and then cut, eventually resulting in puzzle pieces which may be placed in more than one location. Usually, stack cutting is a "frustration technique" for the puzzle assembler, as the wrong placement of pieces will lead to a situation whereby eventually the conundrum has to be recognized and backed out to continue with completion of the puzzle. Stack cutting may also be used to create puzzles with more than one solution. In designing Octastar I came up with a solid color geometric design which was particularly suited for expressing strikingly different solutions. My original plan was to have a single color. In hindsight, this would have been a lot simpler to cut, as some very precise alignments and cutting would not have been necessary, and in hindsight, a single color (or on the other hand a much more intricate pattern) would have been much, much more difficult to assemble as the simple shapes and colors I used ended up defeating most of the error inducing properties of clever stack cutting. I consider this puzzle to be fairly easy to assemble.

Overall I consider the puzzle's design to be brilliant - I'm very pleased with the way the idea works

This puzzle was cut in my Swirl Curl cutting style with 254 pieces. There are no figurals. This puzzle does not include my standard signature piece, rather I selected a relatively big piece and signed the back of it the usual way and for the Project selected a second piece and marked the back indicating the puzzle belongs to the 100 Puzzles Project.

Caveats and a request

This puzzle is my first completed "stack cutting" puzzle. I went through about a half dozen attempts to make this puzzle - various prototypes and experiments over the course of several weeks, none of development puzzles were completed. Some were disasters, other abandoned after lessons were learned. All of the development works will be destroyed. My first attempt involved eight boards! I then narrowed down the cutting solution to just two boards, and then finally to a single board! As this is my very first completed "stack cutting" puzzle, there are problems with it as might be expected. I encountered problems while cutting the red-green stack cut, the resulting cut was quite a bit rougher than the others. When the puzzle is solved in the square configuration, the yellow and orange edge has a small jog at the transition point.

As the 100 Puzzles Project draws to a close, recent puzzles have been bid up to much higher prices than normal. Please do not bid this one up to a super high level! I'm making this unusual request because my rules of the 100 Puzzles Project state that I cannot sell a puzzle in the same size or larger at a price lower than what the auction goes for! Also this is a fairly small puzzle, with just 254 pieces. I've already received several requests to purchase a puzzle with this design and in the coming "mostly retired" years of puzzle making I would like to make a few more and be able to sell them for a reasonable price. The subsequent Octastar puzzles would likely have higher quality in execution too. In my journal I said I would make an exception to the pricing rule, but I have now decided there will be no exceptions within the 100 Puzzles Project; I will run the project to its conclusion as originally designed. This puzzle has spawned further ideas - so in the coming years I will likely offer variants of one kind or another. Another possibility is to license the design to other cutters.

Note: I eventually plan to post on Octastar's puzzle page detailed photos of the stack cutting process.

100 Puzzles Project - Multiple Solutions Puzzles - History

There is no prior history, this puzzle is the first one!