Sunday 27 November 2022

Amazing How Much Difference Is Caused by a Half Voxel

Bubinburr by Juno
If you have not seen it yet - the puzzles that I reviewed last week from Pelikan have gone up for sale now - there are some incredible puzzles there - go get them quick.

It's been a bit of a struggle to get something to write about this weekend - I have had to work 3 weekends in a row and after the Herculean effort of solving all of the Pelikan puzzles, my poor little brain was mush! I have so far completely failed on the last steps of Angry Walter and also failed to solve my jingly puzzles I bought months ago from Aaron. In desperation, I picked up a gorgeous six piece board burr that has been on my to solve tray in the living room for over a year.

I bought a bunch of Juno's 6 piece board burrs way back in October 2021 - I couldn't resist the 5th and 6th in the series of Grooved board burrs which I solved and reviewed (#5 here and #6 here) within a few weeks of their arrival. I then hoped to solve the Bubinburr in the days or even week afterwards. I can now hear Juno and Yukari laughing at me from the other side of the world! They must have known that that was not going to happen! Their description of it should have warned me:
"This puzzle requires some very tricky 22 moves to remove the first piece from the assembled shape. Although a few pieces tend to rotate during the solving process, there seems to be no shortcut solution using rotational movements. It can be solved using rotational movements, but it still requires about the same steps as a rectilinear movement solution, though it is tricky to count while it involves rotational movements. The number of possible assemblies of the puzzle is a very large 30,592, and finding the unique solution among them is extremely difficult.
Once assembled, the puzzle looks just like an ordinary 6 board burr, but as you can see in the images below it has been designed using a grid system of half unit lengths. It doubled the number of moves for the first piece compared to an ordinary 6 board burr that has a maximum 11 moves unique solution. 
One of the most difficult aspects of designing this type of puzzle is to ensure that the puzzle has a unique solution. To achieve this, Juno has carefully modified the shape of a couple of pieces, which are cleverly hidden in the images below."
I thought that it wouldn't be too difficult when I bought it...especially since it did not have any extra grooves to catch me out. But I quickly realised that there was something really tough to this one. The exploration was fantastic with quite a few moves possible but no really deep blind ends. I was always able to return back to the beginning and try another path. There was a weird loop in the pathway as well which I frequently seemed to head through and back to the beginning without intending that. After many hours of attempts, I had to put it down. I just could not find the missing move(s).

The puzzle stayed on my tray for over a year and I would attempt to solve it every couple of weeks without success. It "only" has 22 moves to remove the first piece but I just couldn't seem to manage it. As Juno stated, the use of half voxels made all the difference. I did not bother to look for a rotational solution and tried to keep the pieces controlled as I played. Using that half voxel cut out or added in on every piece (some are not even full thickness) effectively meant that the puzzle was built on a 12x12x12 grid instead of the usual 6x6x6. 

I am not sure what happened today - maybe it was my desperation to have something solved to write about before I spend a day working on trauma patients, maybe it was the very nice gin that I had last night, I don't know. But, this morning, I picked it up and went through my usual initial exploration and attempted my usual moves but for some reason I tried a different move - I think it may have been because I held it differently and gravity found the move for me. But suddenly there was a whole lot more room to manoeuvre than before and Aha! I had my first piece out. The rest of the disassembly was pretty straightforward and I could take the all-important photo:

Finally after over a year of trying!
I don't think I can reassemble it from memory but that is absolutely fine as making the Burrtools file is a significant part of the fun!

So what is going to be next? I have a backlog of puzzles bought from Eric and my birthday present from the present wife, Mrs S has just arrived:

Dayan Gem 10, 16 axis hexadecagon
Eitan's Edge turning Octahedron, 4-Corners cube Plus

No comments:

Post a Comment