Hardware Analysis
In "Hardware Analysis" you can inspect user and match distributions across hardware and quality-setting dimensions, and explore user activity. This document covers Hardware Data Analysis, Quality Analysis, and Retention Analysis.
1. Hardware Data Analysis
1) Hardware Distribution Overview
Filter the hardware query by time range and version. On Android you can additionally filter by vendor and runtime environment. The Distribution Overview shows the overall distribution across hardware dimensions (you choose which). On Android, the typical charts include:
- Pie charts — distribution by RAM, vendor, device, OS, CPU, CPU cores, GPU, ROM, 32/64-bit, available ROM, graphics API version, custom device tier, suspected emulator, etc.
- Bar / trend charts — daily-flow trend of suspected emulators. A toggle controls whether emulator-related data is shown; the hover tooltip displays the date and value.
You can switch the metric base between match count and device count.
2) Data Details
In Data Details you can review data across (but not limited to) RAM, vendor, device, OS, CPU, CPU cores, GPU, ROM, 32/64-bit, available ROM, graphics API version, custom device tier, suspected emulator, and DAU.
You can use Group & Drill Down for deeper analysis. On Android the typical drill-down relations are:
| Current dimension | Supported drill-down dimensions |
|---|---|
| RAM | OS version, device |
| 32/64-bit | OS version, device |
| OS | Device |
| CPU | Device |
| ROM | Device |
| Available ROM | Device |
| Custom device tier | Device |
| CPU cores | CPU, device |
| GPU | Graphics API version, device |
| Graphics API version | GPU, device |
| Other dimensions | No drill-down |
2. Quality Analysis
1) Top Quality-Setting Distributions
Configure the desired filters: time range, version, scene, device, vendor, etc.
In the Top Quality module you can review the bar chart for each grouping dimension, including match count, user count, match-count share, and user-count share.
By default the Top 10 quality entries by match count are shown; drag the bottom range to view more. You can pick the grouping dimension—by default all sub-dimensions of quality are selected, but you can select specific ones (depending on the product).
2) Result Data Table
The result table shows specific data for the selected filters: custom quality category, match count, match-count share, user count, user-count share, etc.
Each dimension has its own column; all sub-dimensions are selected by default. Changing the sub-dimension selection refreshes the table.
3. Retention Analysis
1) Daily Trend
Configure the dimensions and metrics to observe. Common filters include time, version, device, and an emulator toggle.
The Daily Trend module shows the daily activity trend, the period-over-period change, and a daily line chart.
2) Data Details
The Data Details module shows DAU, indexed by date. You can group and drill down by version or device.
