I can test on only a few phones: which ones cover most of my players?

Buying the most popular phones again and again tests the same things over and over. This entry shows how to choose a small set of test phones by what each one adds that the others lack, weighted by how many of your players have that trait.

Why this works

A second popular phone often breaks the same things as the first, so it adds little. Picking each new phone for what is still uncovered, weighted by how many of your own players have that trait, spends a small budget where problems are most likely.

When not to use it

Do not rely on it when you have no data about your own players yet, or when a fault comes from one particular phone rather than a trait. Then test on the phones that your first reports of trouble come from.

In the full entry

• What to build, step by step
• Starting values and the reason for each
• 3 common mistakes and the fix
• 33 lines of example code with 4 checks that pass
• How your AI shows you that it works

Say this to your AI

Use the Spielerando entry "I can test on only a few phones: which ones cover most of my players?" (read_entry which-few-phones-to-test-on-to-cover-most-players) in my game. Build it step by step, and show me how you checked that it works.

Or skip this and connect the MCP: your AI then finds all of this by itself.Connect the MCP