the following is a new creation of mine in the context of Composite Puzzles. Unlike the previous installments of this series, it constitutes a single large puzzle which consists of several smaller components. I encountered the concept – an assortment of smaller puzzles embedded in a larger grid – many years ago. Constructions like this one have appeared frequently in puzzle events, in particular WPC team rounds; the details of how the individual puzzles are pieced together have varied, of course.
Since this puzzle series is directed at more experienced solvers, I have shortened some of the instructions in the PDF file. (You will easily find more extensive wordings of the rules on other puzzle sites.) Here is today’s set: City of Circles
I have often been critical of other authors’ works, and courtesy demands that I point the same degree of criticism at myself. It turns out that this puzzle construct has a considerable number of flaws, and I will explain below what I think they are. In case you want to come to an independent conclusion, you may want to solve the puzzles first before reading on.
The first question we should ask ourselves is whether the puzzle is supposed to be solved by a single solver or by a team of solvers. This may not sound like an important matter, but the two targets are rather divergent. Team puzzles should have a larger number of starting points, ideally different in nature and distributed over various parts, so that the team members can independently make progress in the first stage of solving. This is less important for puzzles designed to be solved by a single person, which should focus on a structured overall solving path.
When I created today’s set, I did not have a clear purpose in mind. As a result, the outcome is neither the one nor the other. It does not have separate significant starting points, but it does not have a clear opening into the solution either. In fact, the first part of the solution path is very rough and pretty much the opposite of elegant. The best thing one can say is that intuition is quite likely to lead to good results.
The locations of the smaller grids are not fixed. Such a situation is not unusual, but it requires a certain amount of care on the part of the author. The global rules must somehow bind the puzzle components together. Personally, I like to see a lot of interactions between the parts, both between the individual puzzles and between local and global elements. In this case, there is no major interaction; the global instructions feel more like an amendment than like an integral part of the construction.
To be honest, I am not sure to what extent the conditions regarding the loop through the unused cells are required. The numbers outside the grid serve to approach the grid positions in the initial stage; however, many parts of the solution may be unique without the exterior clues, which can be considered a bad sign. It also means that, on the whole, the solving path can be broken down in two parts: locating the individual grids and filling them. There are hardly any solving steps which go back and forth.
Another negative aspect is that the individual puzzles are not too great, either. Actually there are two different tasks which must be analyzed here: identifying the puzzle type for each grid and solving the puzzles thereafter. And in this case, it turns out that both have serious shortcomings.
The selection of puzzle types is one problem. I did not have much choice when it comes to puzzle styles which include black and white circles, and some of them are actually interesting. However, having two puzzle types which only deal with circles of one color is a major drawback. (Plus, the interpretation of black circles as part of Battleships seems rather artificial.) For many parts of the grid one can guess which puzzle they accomodate, even without any circles on grid lines or vertices for the Galaxies. As a side note, I have an idea about a similar construction with other symbols, but I will save it for another occasion.
Due to the overall size, most of the individual puzzles are easy or even trivial. (Virtually all the puzzle types involved here need a certain size to become exiciting, and hardly any of them reach this size.) This leaves the question where the actual challenge of the puzzle lies. Most of the energy goes into locating the six grids; there is some crude logic behind it, but it is probably much simpler to try out some constellations and see where they lead.
Finally, there is an issue notation-wise. Some of the puzzle types are Dissection Puzzles, and marking borders inside the respective grids could lead to confusion as opposed to the drawing of the six puzzle grids – not in this particular puzzle, I guess, but in general. It would be preferable to have puzzle types which avoid this problem per se, but once again, my choices were limited.
So, why am I publishing this at all? One reason is that puzzle solvers have different tastes, and I hope that some may enjoy this composite puzzle despite the various flaws I have mentioned above – especially if they are intuitive solvers who like to think outside the box. The other reason is that I want to share my general views regarding quality puzzles, even on occasions when I am not able to live up to my own standards, so that I can improve on my work in the future.