Mobile Game Optimization: A Board Built for One Hand
Most of our traffic arrives on a phone and the board was drawn for a mouse. This is the mobile game optimization list: touch targets, board scaling, what happens to a live match when a call comes in, and the things we decided not to build.
Most of our traffic arrives on a phone, and the board was drawn for a mouse. Mobile game optimization is now the largest item on our engineering list: how a touch interface board game handles a fingertip wider than a grid cell, how the board scales in portrait, and what happens to a live match when a call comes in. Here is the full list, including the parts we decided not to build. Open the play page on your phone first and the rest will read differently.
Where the board breaks on a phone
A fifteen-line board sits at a comfortable 640 pixels square on a desktop browser. On a 6.1 inch screen each cell drops to roughly 21 pixels, and a fingertip cannot tell which intersection it is over. Pinch zoom recovers half of that and costs you the whole-board view, which is what gomoku needs. It is the hardest trade-off in the mobile work.
Touch targets on a touch interface board game
The target we are building to is 44 effective pixels, wider than the grid itself. When a tap lands between two intersections the closer one wins instead of the move being rejected. After a stone is placed there is a 300 millisecond window to undo it, aimed squarely at mis-taps.
- 44 pixel effective targets, wider than the line spacing
- Nearest intersection wins, so no tap is wasted
- 300ms undo window after a stone is placed
- Press and hold to preview the point, so your finger stops covering it
Responsive game layout
Portrait is the default: board on top, information and controls below. Turn the phone sideways and the board shifts left, with the right column taking the clock and the hints. Notches and the bottom gesture bar are handled with safe-area insets, so a system gesture never eats your move.
Mobile chess usability, and why xiangqi is harder
Xiangqi is a harder problem than gomoku. Each piece carries a character that turns to mush at phone scale, and the movement hints for cannons and chariots have to be drawn on a small board with no room for a written explanation. Our approach is to highlight the legal destination points and move the text into a second panel. It is in development, and it is the slowest part of the mobile chess usability work.
What we are not doing
- No forced landscape. One-handed phone use is the common case, and landscape needs two hands
- No animation beyond the haptic toggle. Things moving on the board get in the way of reading the shape
- No mobile-only paid tier. It is the same free game on every screen
- No background polling for the sake of the leaderboard. Switch away and the match pauses
FAQ
Is the board too small to read on a phone?
No. In portrait the board fills the width, and you can pinch into a section and release to return to the full view. While zoomed, placement still resolves to the nearest intersection.
What happens to my match if a call comes in?
The game is held on the server. As long as you do not start another match on the same account, you can pick it up where you left it.
Do I need to install an app?
No. Open the browser and play. There is nothing to install and no signup required.
When will xiangqi and go work on mobile?
Gomoku works now. Xiangqi is in development with the mobile movement hints as part of that work, and go is last in the build order.
Related
- Performance Work: A Smoother Board
2026-10-19
- Go Roadmap: Why It Comes Last
2026-10-15
- From 1200 to 1800: A Rank-Up Roadmap
2026-10-13