Liquid-sorting puzzles are not a new genre. The core mechanic — pour colored liquids between tubes until each tube contains only one color — has been around for years, and there are dozens of implementations across mobile and browser platforms. If you've played one of them, you know the rhythm: identify which colors need to move, find the sequence of pours that gets them there, execute.
Decant started as exactly that kind of game. The first prototype was a clean implementation of the standard liquid-sorting mechanic, with a visual style that matched the rest of Woodyn — warm colors, minimalist UI, no ads. It was fine. It was also obviously not different enough from what already existed to be worth releasing.
The color mixing mechanic was the answer to that problem — and it nearly didn't make it in.
The Idea
The question that generated the mechanic was simple: what if some of the colors on the board weren't starting colors? What if some had to be created by combining what was already there?
The real-world analogy was obvious — primary colors mixing to form secondaries. Red and blue make purple. Blue and yellow make green. Red and yellow make orange. This is something players intuitively understand without needing it explained, which is a significant advantage for a mechanic. You can introduce it in a tutorial level and players immediately grasp not just what it does but why it makes sense.
The implementation question was how mixing would interact with the core pour mechanic. In a standard liquid-sorting game, you can only pour a color onto a matching color. If mixing creates a new color, what can that color be poured onto? What can be poured onto it?
The answer was the second key decision: mixed colors behave like any other color. A tube of purple — created by mixing red and blue — can receive more purple or be poured into an empty vessel, just like a tube that started with purple liquid in it. This matters because it means mixing is a tool for creating the color you need, not a special rule with its own exception set. The same mental model that handles basic sorting handles mixing too.
Why It Almost Got Cut
The first playtest of the mixing mechanic was not encouraging. Players understood it immediately — the intuition was right — but they found it deeply frustrating. The problem was deadlocks.
In a standard liquid-sorting game, deadlocks are recoverable in most cases. If you're stuck, you've usually made a suboptimal move somewhere that can be undone. In the mixing version, deadlocks were much more final. A wrong mix — pouring red into a blue vessel when you needed that blue elsewhere — created a purple that couldn't be unmixed. The color was permanently changed, and the puzzle was often unsolvable from that point.
The first instinct was to limit when mixing could occur — maybe require a specific "mix vessel" or only allow mixing at certain steps. But this felt like it was solving the problem by making the mechanic weaker. The mixing was interesting precisely because it could happen anywhere, in any vessel, at any time. Restricting it would reduce the complexity in both directions — fewer frustrating deadlocks, but also fewer interesting decisions.
The real fix was in puzzle design, not mechanic design. A Decant puzzle with mixing mechanics has to be constructed so that the mixing steps are part of the intended solution, not an alternative path that leads to a dead end. This means the designer has to build each puzzle starting from the solution and working backward, ensuring that wrong mixes always leave the player stuck in an obvious, recoverable way — or, better, that the puzzle structure makes wrong mixes easy to recognize before they're committed to.
This is significantly harder to design for than standard sorting puzzles. It's also why the daily puzzles in Decant are handcrafted rather than generated. An algorithm can verify that a puzzle is solvable. It can't verify that a puzzle is satisfying — that the mixing steps feel like discoveries rather than traps.
The Tutorial Problem
Once the mechanic was stable, the next challenge was teaching it. Color mixing is intuitive, but that intuition can work against you in puzzle design. Players who understand that red and blue make purple sometimes try to mix them whenever they see both colors present, rather than asking whether the mix is necessary for the solution. The mechanic invites a kind of mixing-first thinking that can lead straight into deadlocks.
The tutorial solution was to introduce mixing in a puzzle where it was the only possible move. A board with two vessels — one red, one blue — and a single empty target vessel for purple. There's nothing else to do. Pour red into blue, or blue into red, and you've created the color you need. Complete.
The next tutorial puzzle introduced mixing alongside standard sorting moves, but structured so that mixing had to happen first before the sorting sequence could proceed. By the third puzzle, mixing was just part of the vocabulary — present but not highlighted, used when needed.
Post-tutorial, the feedback was significantly better. Players still encountered deadlocks in harder puzzles, but they understood why the deadlock had happened and could identify the point where their plan went wrong. That's a solvable frustration. A deadlock you don't understand is not.
Colorblind Mode
A puzzle game built around color had an obvious accessibility gap: players who can't reliably distinguish between red and green, or blue and purple, would be blocked from playing at all. Colorblind mode — which overlays a unique symbol on each color — was added before the game launched, not after.
The symbol set went through several iterations. The first version used letters (R for red, B for blue, etc.), which tested badly — players found them distracting and reported that the letters made the puzzle feel like a word game rather than a visual one. The second version used geometric shapes — circles, triangles, squares — which tested much better. They're visually simple enough not to compete with the liquid color, but distinct enough to be unambiguous.
The mode persists between sessions. Players who need it should only have to turn it on once.
Where It Stands
Color mixing is now the thing that makes Decant different from every other liquid-sorting puzzle, and it's the reason the daily puzzle format works. The same basic board structure that would be trivial in a standard sorter becomes genuinely interesting when some of the target colors don't exist yet and have to be created from what's present.
The mechanic is also the hardest part of puzzle design. Every new daily puzzle has to be tested against the question: does the mixing feel like a discovery, or does it feel like a trap? Getting that balance right is ongoing work.
Play Decant here — the daily puzzle resets at midnight, so there's always a new one if you've already solved today's. Or read how Canopy was built if you want the other side of the Woodyn development story.