How to play Saltfire — guide, tips and FAQ
Everything you need to play Saltfire, how the score works, and why par is the number that matters.
What is Saltfire?
Saltfire is a small daily puzzle about combining things. Every day names one target — Obsidian, Lantern, Rumour — and you have to reach it starting from four primordial elements: Salt, Ember, Dew and Breath. Tap two elements together and they fuse into a third, if the pair takes at all. Ember and Dew make Steam. Salt and Ember make Ash. From there the tree opens out into stone and weather, growing things, made things, and a good number of stranger ideas.
It is free, there is nothing to install and nothing to sign up for, and it takes about five minutes. Everyone in the world gets the same target on the same day, which is the whole point: your score is directly comparable with anyone else’s.
How do you play a daily fusion puzzle?
- Read the target. The day opens by naming what you are making and showing its par — the smallest number of merges the puzzle can be solved in.
- Tap two elements. Your tray starts with the four primordials and nothing else. Pick one, pick a second, and they drop into the well.
- Watch what condenses. A pair that works produces a new element and adds it to your tray. A pair that does not is a dud — it still tells you something, and it is marked so you never try it twice by accident.
- Chain upward. Everything you make is now usable as an ingredient. Most targets are four to eight merges deep.
- Reach the target. The moment it appears your score locks and you get a card: merges taken, par, and how far over you were.
- Keep playing anyway. The workbench stays open after you solve. Every further merge is free, costs you nothing, and anything new you find is kept permanently in your cabinet.
The scoring rule in one line: a pair you have not tried today costs one step whether it works or not; a pair you have already tried costs nothing and can never cost you again. So guessing is not free, but forgetting is.
What does par mean here?
Par is the smallest number of merges the day can be solved in, and it is a computed number rather than an authored guess. Before a puzzle ships, a breadth-first search walks the entire recipe tree outward from Salt, Ember, Dew and Breath, level by level, and the first level at which the day’s target appears is the floor. The build then replays the chain that search found against the game’s own merge rules, and if the replay spends one step more or one step fewer than the search claimed, the build fails and that day never runs. So par is the provable optimum, not an estimate someone eyeballed.
Two things follow from that, and both matter while you are playing. First, par counts distinct pairs, not branches of a tree: an intermediate you make once and then reach for twice costs one step, not two, because it is already sitting in your tray. The short way through a day almost always reuses something. Second, par assumes a perfect run by someone who already knows the tree, and nobody plays that cold. Two or three over par is a good day. Matching par is the best round that exists — you cannot go under it.
How do you get better at Saltfire?
- Build the obvious middle first. Ash, Brine, Dust, Steam, Mist and Stone show up in an enormous number of chains. Two or three of them early is rarely wasted.
- Read the target as a recipe. Most element names tell you their own ingredients if you say them out loud. Obsidian is cooled lava. A Lantern is a flame you can carry. Work backwards from the noun.
- Self-merges are real. An element combined with itself is a legal move and several early elements only come out that way.
- Duds are information, not failure. Par assumes a perfect run; nobody plays a perfect run cold. Two or three over par is a good day.
- Most things have more than one recipe. If a route stalls, the element you want is usually reachable from something else already in your tray.
- Yesterday’s knowledge is the whole skill. You always start from the same four elements — what carries over is what you remember.
What can you do when you are stuck?
Nothing in Saltfire is ever bought, unlocked or gated — every bit of help is free and sitting on the screen already. Three things are always there for you.
- The ledger. Every pair you have tried today is written down, in order, with what it produced. Duds are struck through. Tap any line to open that pair again — it costs nothing, because you already paid for it.
- The tier tag. Under the target name is how deep it sits: a tier-2 element is two or three merges from raw Salt, a tier-5 one is a long way up. It tells you whether to keep building or to start combining what you have.
- The nudge. After about a dozen merges the ledger quietly names one element that appears somewhere in a shortest solution. One element, never the route — enough to unstick you, not enough to finish the puzzle for you.
And if a day genuinely beats you, it is not lost. The moment it closes it moves to the archive with its whole solution written out and an explanation of why that particular chain is the short one. That is the honest place to look up an answer, and reading a few of them is the fastest way to get better.
What else do people ask?
Is Saltfire free?
Yes, completely, and there is nothing to buy. The site carries a small number of ads and that is the whole business model.
Do I need an account?
No. There is no login and no sign-up, and we hold no personal data about you. Your cabinet, your streak and today’s progress live in your own browser. The one thing that reaches our server is the number of merges a finished run took, counted anonymously into that day’s totals with no identifier of any kind attached to it.
Why do I start from four elements every day when I have discovered hundreds?
Because par would mean nothing otherwise. If a veteran began the day holding two hundred elements, no two scores would be comparable. The cabinet is a museum of everything you have ever found, not a head start.
What exactly is par?
The smallest number of merges the day can be solved in. It is computed before the puzzle ships and then proved by actually replaying the chain against the game’s own rules, so it is a real number and not an estimate. You cannot beat par — matching it is the best possible round.
Can I see the answer?
Not for today. Once a day has passed it moves to the archive with its full solution and an explanation of why that chain is the short one.
Does a dud really cost a step?
Yes — a new pair costs one step whether or not it takes. Trying the same pair again costs nothing, so you are never punished for a misclick or a bad memory.
How do I keep a streak?
Solve on consecutive days. Miss one and the streak starts again; your best is kept either way.
Is this a serious tool?
No. It is a game, for fun, and nothing in it is advice about anything.