Use cases~5 min read
Test an Android Screen Size Without Owning the Device
How to test an Android app on a screen size or foldable device you don't own, using TaloTrace's cloud-hosted Android emulators and custom screen geometries.
TL;DR
Yes, for screen sizes. TaloTrace, a QA testing platform for web, iOS and Android apps, runs your Android app on cloud-hosted Android emulators and serves custom screen geometries on the fly, so you buy no hardware. The caveat is that the devices are virtual and physical devices are not offered. Fold and unfold posture changes are not something TaloTrace documents today, so ask about them directly.
Key takeaways
Virtual devices: TaloTrace runs Android apps on cloud-hosted emulators, with no hardware for you to provision or attach.
Custom geometries: on Android, custom screen geometries are served on the fly.
Device matrix: the same scenarios can run across several device or OS configurations in one submission, with results grouped for comparison.
Evidence: every run has a screen recording, and each finding carries the time window where the defect shows.
Limits: no physical devices on any platform, and alternate matrix profiles are not available to everyone.
What Are You Actually Trying to Do Without the Device?
You can test an Android app on a screen size you do not own by running it on a virtual device instead of a physical one. That matters because your device lab only holds the phones you could afford to buy, and layouts may behave differently on screen shapes you do not own. Google's Android Developers Blog publishes a developer guide for Pixel Fold and Pixel Tablet that covers adapting apps to larger and foldable screens.
The job is narrow: see how your app behaves on a screen size you cannot put in your hand, and get evidence you can act on. That is a coverage question, so the useful answer is what runs, where, and what comes back.
How Does TaloTrace Run Your App on a Screen You Do Not Own?
TaloTrace has three cloud-hosted, virtual execution planes: a browser for web apps, an Android emulator for Android apps, and an Apple iOS Simulator for iOS apps. The plane is chosen automatically from the app's platform. There is nothing to provision or attach.
On Android, custom screen geometries are served on the fly. When you start a run, from the app or through the API, you choose the device profile. Which alternate profiles you can use depends on your matrix access, so confirm that when you get in touch.
For a deeper look at the Android side, see our post on AI QA testing for Android apps.
How Do You Cover Several Screen Sizes in One Go?
The device matrix lets you run the same scenarios across several device or OS configurations in one submission. Results are grouped so you can compare behaviour between configurations.
Each device runs on its own lane, and lanes execute independently. One slow scenario does not hold up the others, and a matrix submission makes progress across configurations at the same time rather than one after another.
You write no test scripts, per size or otherwise. TaloTrace works through your app like this:
- Sight-based navigation: it navigates by looking at the screen, not brittle element IDs, so it keeps working when the UI changes.
- Interaction range: taps, typing, scrolling, swiping, long-press, double-tap, drag and pinch.
- No maintenance scripts: there are no test scripts to write or maintain.
Where Do the Scenarios Come From?
You point TaloTrace at your app and, optionally, describe the flow you care most about in plain language, such as "complete a checkout". It explores the running app, plans the user journeys worth testing and drives each one to a verifiable outcome. The journeys it completes become replayable scenarios.
Scenarios can also be authored from a chat description, or written by hand. A goal counts as done only when a machine-checkable predicate proves the outcome, so a journey is not reported as a pass on an assumption.
If you want the full flow, the how it works page walks through it.
What Do You Get Back From Each Run?
A Run produces a set of Findings, and you get a Trace of each completed run. Each Finding is backed by a recording, TaloTrace's reasoning and step-by-step reproduction, including the Evidence panel with TaloTrace's step-by-step trajectory.
- Screen recording: captured for every run, with the time window where each defect shows.
- Per-scenario verdicts: sitting alongside the recording.
- Visual analysis: a vision model analyses the recording for visual and functional anomalies, not only hard failures.
- Severity: five levels, Critical (P0) to Trivial (P4), and you can supply your own severity guidance for TaloTrace to follow when rating findings.
Repeated observations of the same defect collapse into a single issue within a run. Findings live in TaloTrace's built-in issue view.
What Should You Look for on a Screen You Cannot Hold?
On an unfamiliar screen size, look for layout problems before logic problems. A virtual device with a different geometry is a quick way to expose them.
- Clipped or truncated text: labels and buttons that no longer fit.
- Overlapping controls: elements that collide when the width shrinks or grows.
- Off-screen actions: a primary button pushed below the fold or out of reach.
- Broken scrolling: content that cannot be reached on a shorter screen.
TaloTrace's vision model analyses the screen recording for visual and functional anomalies on screen, not only hard failures, and each Finding carries the time window where the defect shows.
What Makes TaloTrace Different for This Job?
TaloTrace separates finding a bug from reporting it. It drives your app and records what looks wrong, then independently verifies each finding before it reaches you. Nothing you see is the raw, unchecked output of the step that produced it.
Results also start hidden. Findings become visible only when a reviewer approves them, or when the project is explicitly configured to auto-approve. Low-confidence findings are held separately and are released only by an explicit per-finding decision.
You can start runs on demand or on a daily or weekly schedule, which always runs the latest finalised build. Our post on an autonomous mobile app testing tool covers the wider picture.
Where Does This Stop?
Naming the edges is what makes the rest believable, so here they are.
| Platform | Screen sizes | Physical devices |
|---|---|---|
| Android emulator | Custom screen geometries served on the fly | Not offered |
| iOS Simulator | Each requested device type served by a simulator of that exact type; arbitrary custom sizes not offered | Not offered |
- Virtual only: physical devices are not offered on any platform. An emulator is not a substitute for checks that depend on real hardware.
- Foldables: custom Android screen geometries are supported. Fold and unfold posture changes, hinge behaviour and similar foldable-specific states are not something TaloTrace documents today, so do not plan around them. Ask about those directly.
- iOS: each requested device type is served by a simulator of that exact type, and arbitrary custom screen sizes are not offered.
- Matrix access: matrix testing is gated. Alternate device profiles are not available to everyone, so confirm what you can use before you plan around it.
Findings are also not exported to Jira, Linear or GitHub. They stay in TaloTrace's built-in issue view.
How Do You Get Started?
Beta is open, and access is granted rather than automatic. Tell us which screen sizes you are missing and we can talk through what is available to you.
Frequently asked questions
Can TaloTrace test my Android app on a physical foldable?
No. TaloTrace runs on cloud-hosted virtual devices only, and physical devices are not offered on any platform. If your release depends on behaviour only real hardware shows, you still need that hardware for that check.
Does TaloTrace test fold and unfold transitions?
Not as something TaloTrace documents today. What TaloTrace supports on Android is serving custom screen geometries on the fly. Ask about posture changes specifically when you get in touch.
Can I run the same scenarios across several screen sizes at once?
Yes, through the device matrix. You submit the same scenarios across several device or OS configurations in one submission. Each device runs on its own lane, so configurations make progress at the same time.
Do I have to write test scripts for each screen size?
No. TaloTrace explores your app, plans the user journeys worth testing and turns the journeys it completes into replayable scenarios. It navigates by looking at the screen, not by element IDs.
Is the device matrix available to everyone?
No. Matrix testing is gated, and alternate device profiles are not available to everyone. Confirm which profiles you can use when you get in touch.
Does this work for iOS screen sizes too?
Only partly. On iOS, each requested device type is served by a simulator of that exact type, and arbitrary custom screen sizes are not offered. Custom geometries are an Android capability.
