[u]In short:[/u] OCTOPICOM is an entry-level logic game about processing light signals. The game follows the style of Zachtronics, but is much simpler and shallower. Optional optimization challenges are more demanding and account for more than half of my playtime. I found some enjoyment in the game, but since it's marred by bad UI, controls and design decisions I'm having a hard time recommending OCTOPICON at full price. If you like these kinds of games and can stomach a somewhat rocky user experience it's fine to pick up on sale.
[u]In long:[/u] As with many light-based games, [b][u]gameplay[/u][/b] revolves around additive colour mixing. You start with a pre-built emitter for white light, some inputs applying colours and outputs. Using mirrors, colour filters and switches you are tasked with transforming the input into the required output. For how many parts crowd your toolbox, your selection of useful components is surprisingly limited. In the end 2 kinds of transistor-like gates (default open with off-switch, default closed with on-switch) and a delay/storage part will do 98% of the heavy lifting in terms of computational logic.
The game is cycle-based, but their number is set for each level (usually 64) and distances don't matter. This means you can only optimize for area/components used - speed or efficiency are not a thing the game measures. More often than not you are also not providing general solutions (say for randomized inputs), but rather solutions for one single carefully crafted input sequence, even though it typically covers edge cases to make levels more interesting.
In terms of [b][u]level design and difficulty[/u][/b] I'm honestly not too impressed. It took me a grand total of 6.5 hours to solve the game's 42 levels. 16 qualify as tutorial, 16 are fairly easy, and just the last 10 have some moderately advanced challenges. As a veteran of the genre many solutions came quite naturally and immediately to me, usually not even requiring a test run. In fact, much of my initial playtime was spent figuring out the sometimes convoluted rules from their terrible descriptions and the sample outputs, but once you grok those, the solutions were usually easy to implement.
The difficulty curve is very gentle and beating a level always gives you access to the next two (a great way to prevent getting stuck on a single level), making this a good entry-level title into the world of computational logic games. Most importantly, it shares the trademark aspect of having some satisfying and elegant solutions. Discovering a way to dramatically simplify a convoluted solution always feels nice. This is why the (optional, only for achievements) optimiziaton goals for the levels turned this game around for me. After beating the game in 6.5 hours I was a lot more negative on it than after spending another 9 hours optimizing all the levels. Sure, suddenly being space-confined often just boiled down to rearranging spread-out components. But when it forced me to really think about mechanics and to find new and simpler approaches, it added some of the spice and enjoyable magic of iterative design I had been missing so far. Two levels even ended up having some interesting non-obvious conceptual challenges.
The [b][u]presentation[/u][/b] is very lackluster, below being merely functional. This is not even "developer art", it's just straight-up "no art". Menu buttons are coloured rectangles. The components are grey on grey to the point where it inhibits visual clarity and leads to errors. It is hard to tell the orientation of the most common components, switches (black rectangles with 3 tiny nubs) and mirrors (half light, half dark grey). At least you can freely zoom in to fix problems.
Sound effects boil down to 7 electronic beeps. The background music consists of 8 royalty-free tracks. They are not strictly bad, but just way too animated for a logic game. I turned off the music maybe 2 minutes into the game because I couldn't think straight.
The [b][u]UI and controls[/u][/b] are generally poor. Tooltips and level tasks are not very clear and sometimes misleading. Expected outputs are not shown by default, but require clicking a strange "G" button. Some crucial features are never tutorialized at all (like merging beams). Optimization challenges are only shown on the final screen of a level and there is also no indicator whether you've completed them outside of said final screen. There are multiple save slots per level but no in-game way to transfer solutions between them (and thus no way to fork and work iteratively).
There are hotkeys for most things but they are not rebindable and some of them are fairly unintuitive. The mouse wheel rotating components goes the "wrong" way for my taste and there is no way to change it. Every component you place is automatically deselected unless you hold Shift to place multiples, which gets old fast (especially considering RMB to clear your selection is so accessible anyway).
But not everything is bad - there is a good undo/redo feature and I also like that you can manually step through cycles. It took some getting used to, but I ended up liking the box-select feature with CTRL. If only the levels were bigger so you could take full advantage of this feature during rapid prototyping (the emitter being in the center of the map but only shining one way doesn't help). Reading through the patchnotes the game was in a much worse state at release and some user feedback has already taken off some of the edge (apparently it used to be that most levels showed no expected solution and you were stuck trying to interpret the badly-worded instructions), but there is still a lot more polish this game could have used.
On the [b][u]technical side[/u][/b] the best I can say about the game is that it has *very* low resource demands and passable settings. You can separately lower the volumes of sound effects and music. Windowed mode lets you drag the window to the right size, which is then remembered. There's also 2 half-baked simulation settings and a colour-blind mode, which just crudely adds letters to every beam or square of colour.
So far this is not great, but I wouldn't mind it as long as the game was perfectly stable. Unfortunately I once had a glitch where it kept showing deleted components, and twice the game crashed on me (no progress lost, it even saved the move that crashed it). Both problems were related to moving groups of objects.
In [b][u]conclusion[/u][/b] this is a decent title if you can temper your expectations. It provides a weekend afternoon of light puzzling if you don't mind fighting the interface a bit, and some more if you get into the optimiziaton challenges. This is neither a very long, nor a very deep game, and it is presented crudely and poorly to boot. Even for a 2-man studio this is quite rough stuff. Still, given that I ended up having fun finding elegant solutions, at the usual sale discount of -50% the game feels adequately priced to me. If you are having doubts you might want to check out the demo (either here on Steam or from the game's itch.io page where you can also buy a DRM-free version).
