Game review
Euphrosine Wants Ten Dots
7.1 / 10
- Developer
- Quinten Clause
- Publisher
- Independent
- Released
- 1 January 2016
- Platforms
- Web browser (desktop and mobile)
- Genre
- Platformer
- Monetisation
- Free, open source, no ads and no in-game purchases
- Players
- Single-player only
The short version
Euphrosine Wants Ten Dots is a short, pleasant ladybug platformer — climb trees, dodge an enemy, catch ten dice to win — that would be entirely forgettable if not for its touch control scheme, which is genuinely built rather than bolted on. Instead of on-screen buttons, holding a finger down and dragging it creates a virtual joystick wherever you touch, dragging up switches to climbing on a tree, and — the detail that sets it apart — tapping a second finger while the first is still held triggers a jump, mirroring the keyboard's arrow-keys-plus-spacebar scheme almost exactly. All three pieces were confirmed working through real simulated multi-touch, including the two-finger case, which most reviews wouldn't even think to test.
Biggest strengths
- The "hold and drag to run, drag up to climb, second finger to jump" scheme was tested end to end with real simulated touch rather than assumed from the code — a live drag was confirmed to set the run cursor AND move the player's x-position in real time, a drag up correctly triggered the climb cursor, and adding a genuine second touch point while the first was still held correctly triggered the jump flag at the same time as the held direction
- The virtual joystick appears wherever you actually touch the screen rather than pinning controls to a fixed corner, so it never fights your hand's natural position
- Touch and keyboard run through the same underlying cursor object, so neither input method is a second-class citizen — confirmed by reading the input-handling code, which checks `game.touchControl.cursors` alongside the keyboard and gamepad on every single control check
Biggest weaknesses
- The core loop — climb, collect ten dice, avoid one enemy, done — is thin even by jam-game standards, and most players will see everything the game has to offer in a single sitting
- Fairly conventional collectathon platforming underneath the control scheme; nothing about the level design, the dice-collecting goal, or the single enemy type is a new idea
- Small original audience (14 GitHub stars) for what is, execution aside, a straightforward hobby platformer
Who it's for: Anyone who wants a five-minute platformer palate cleanser, and a good reference point for what a genuinely considered touch control scheme looks like
Our recommendation: Worth playing once for the control scheme alone; don't expect a long session
7.1 / 10
There's a specific failure mode this shelf keeps running into: a game ships real touch event handlers, and they turn out to only handle aiming, or only fire from a UI element sitting outside the actual play area, or get silently switched off by a platform check nobody remembers to remove. Euphrosine Wants Ten Dots doesn't have that problem, and it's worth explaining exactly why before getting to the game itself, which is a much smaller story.
A touch scheme built to mirror the keyboard, not replace it
The desktop version's instructions read "left + right arrow keys to run, spacebar to jump, up arrow in the air to float, up + other arrow keys in a green area to climb." The touch version's instructions on the same splash screen read "hold and drag one finger to run, drag up to climb and float, while holding tap with a second finger to jump" — and reading Phaser.Plugin. TouchControl in the source confirms these aren't two separate systems bolted together. A single cursors object (up, down, left, right, space) is set either by the keyboard or by a virtual joystick that spawns at wherever your finger first touches down, and every movement check in the game reads from that same object regardless of which input produced it.
This was tested directly rather than trusted on the strength of reading the plugin code. A real touch-and-drag, held in place partway through rather than released and re-measured afterward, showed the right cursor flip to true and the player's on-screen x-position advance in step with the finger — confirming the joystick responds continuously while held, not just once at the start of a gesture. A drag angled upward instead correctly set the up cursor. And the detail most reviews wouldn't think to check: with the first finger still down, adding a second simulated touch point set cursors.space to true at the same instant cursors.up stayed true from the first finger — genuine two-finger input, working exactly as the splash screen describes it, not just plausible-looking code that was never run.
A short, pleasant platformer underneath it
The game itself doesn't ask for much beyond that control scheme. You climb a loose vertical arrangement of tree trunks and branches, drop down to catch falling dice before an enemy gets in your way, and win once ten dice are collected. It's charming — soft cloud backgrounds, a readable ladybug sprite, clean pixel-art trees — and it's over quickly. There's no scoring beyond the dice count, no difficulty escalation to speak of, and nothing about the goal or the single enemy type reaches for anything beyond "cute five-minute platformer." That's a fair trade for a personal project, but it means the reason to play this is the control scheme rather than the platforming challenge underneath it.
Pros and cons
Pros
- The "hold and drag to run, drag up to climb, second finger to jump" scheme was tested end to end with real simulated touch rather than assumed from the code — a live drag was confirmed to set the run cursor AND move the player's x-position in real time, a drag up correctly triggered the climb cursor, and adding a genuine second touch point while the first was still held correctly triggered the jump flag at the same time as the held direction
- The virtual joystick appears wherever you actually touch the screen rather than pinning controls to a fixed corner, so it never fights your hand's natural position
- Touch and keyboard run through the same underlying cursor object, so neither input method is a second-class citizen — confirmed by reading the input-handling code, which checks `game.touchControl.cursors` alongside the keyboard and gamepad on every single control check
- Clean and legible pixel art with a soft, cohesive palette; the ladybug, trees, and cloud background all read clearly against each other even at small sizes
- No analytics, no third-party trackers, and the "possible mobile gate" the install scanner flagged turned out to be Phaser's own internal feature-detection code, not an actual redirect — verified by reading it directly rather than trusting the flag
Cons
- The core loop — climb, collect ten dice, avoid one enemy, done — is thin even by jam-game standards, and most players will see everything the game has to offer in a single sitting
- Fairly conventional collectathon platforming underneath the control scheme; nothing about the level design, the dice-collecting goal, or the single enemy type is a new idea
- Small original audience (14 GitHub stars) for what is, execution aside, a straightforward hobby platformer