Better Together: Real Devices, Emulators & Simulators for Mobile Testing

Sauce AI for Test Authoring: Move from intent to executing in moment.|xBack to ResourcesBlogPosted September 8, 2021

Better Together: Real Devices, Emulators & amp; Simulators for Mobile Testing

Using emulators & amp; simulators and existent device together in an machine-controlled testing environment, allows QA squad to get the most out of their nomadic testing efforts.

quote

Ten age ago I saw a Selenium test that could tick on a capital & quot; I & quot; that was one pel extensive. I was not capable of clicking on it myself. Around that time, a lot of people began using phrases like & quot;End to end is further than you think. & quot; Simulators and emulators, after all, could not simulate warmth, pathetic memory management, low power, and other job.

Then again, that was ten years ago.

The teams I work with today tend to have a stronger focus on thepipeline of speech work, along with eight, sixteen, or perhaps thirty-two tests at the same time. This kind of work isn & # x27; t precisely a outstanding fit for physical devices secure into a cable and running to a laptop. Managing that many literal device in the cloud is no walkover either.

In other words, there may be a time and a place for simulators, emulators, and physical devices. The challenge is which tool to use where.

Let & # x27; s verbalise about it.

Scaling your tests

A single machine to perform build is certainly a way to part; it was howExtreme Programming begin. That takes a individual machine and allows it to run examination on one earpiece.

Most mod engineering processes require to run exam on every new commit pushed to the branch, and possibly on every ramification. Even with just one branch, if the squad append ten minutes of new mechanization per dash, that phone is travel to be bogged downwardly within a twosome of month. Plus, you likely don & # x27; t want to run on the same device every time -- you & # x27; d like to shift thing up, extend on a tablet, or an older twist, or a different operating system.

Enter Simulators and Emulators.

The property for emulators

For autonomous testing across multiple user personas, check out SUSATest — it explores your app like 10 different real users.

Emulators re-create the entire twist, downwards to the chip, in software. You can conceive of it nigh as virtualization, as it is potential to run Windows on a Macintosh laptop with a tool like VMWare -- even rearwards when the computers ran on an entirely different Central Processing Unit (CPU). Emulators don & # x27; t have heat or battery problems. They may perform at a different speed than the existent thing. Still, when it comes to test of normal functions under normal weather, an emulator is going to do the same thing as the real device. Which is strange, given their reputation for being different.

Because tests on copycat are software-defined, multiple tests can be run on tens of emulators at the click of a push without having to manually make each aper for the tests. Further, automation is easier with emulators as the tests can be fulfil without manual intervention, and be contain remotely.

The problem with emulators is speed and memory. That extra tier of processing, like virtual machines, leads to a lag. Yes, you can run emulators in the cloud, but when it comes to full hurrying ahead, most people in reality use simulators, the ape ’ s less realistic, but easier to run cousin.

The place for simulators

Simulators create the appearance of the device in software. They do not re-create the device all the way down to the CPU, but instead mime the look and feel. Apple, for example, has been madly combine their mobile operating system, iOS, with their background, OSX, for years. The Safari browser in the two is essentially the like. Through the exploiter of share codification libraries, Apple can easy put a desktop simulator in their XCode coding tool. Plus, it pass much more quickly than the emulator.

With a smaller memory footprint than an emulator, easier to & quot; spin up & quot; and down, and easier to run in parallel, simulator seem like the quick fix for a Uninterrupted Delivery Pipeline or yet a simple test run. Until you consider that the simulators are the least like the existent thing. Simulators represent what the instrument provider expects the device will do … not what it will actually do.

As it turns out, you want to use all three options.

And physical devices as well

If any one thing I & # x27; ve done will be remembered in ten more geezerhood, then it will probably be for championing theequilibrate breakfast strategy. That is to say, it takes respective things. Yes, test by paw with physical devices. Sit on a bus and go through different cellular web. Leave the phone on with the coating fighting and see how fast the battery drains or how much network you use. Feel the covering, viscerally, in your paw. Test every feature with a existent device by hand. Right alongside you can guide some small subset of those experimentation and draw them in so much detail they can be executed by a figurer on every regression run.

Best of all, you can run real device in the cloud. Several companies, Sauce Labs include, hold ways to connect to real device, to either run tryout automation, or, perhaps, to have a human look at a device/operating system combination that you don & # x27; t have available. Done as part of human exploration, real devices in the cloud can be cheap and easy.

Running tests in the cloud has benefits too, in that you can run a outstanding number, you can run several at the same clip, and you can get a picture of examination run results. The downside of this, in general, is higher cost and dull test results.

Putting it all together

To get the virtually out of their mobile testing efforts, organizations need both existent devices and aper and simulator. One pick promises more accurate test results, while the other delivers greater legerity.

Screen Shot 2021-09-08 at 2.40.58 PM

Coded correctly, you can modify the eccentric of gimmick at runtime for any test run. A general exam run could use simulators (easy, tight, cheap), perhaps the all-night run uses emulators, and the occasional regression or & quot; what if & quot; or & quot; big browser ascent & quot; run could use existent devices. Create a scheme with a balanced breakfast, and you may find you can have speed, cost, and coverage, all at the like time.

It & # x27; s a great clip to be in testing.

Published:
Sep 8, 2021
Share this post
Copy Share Link
LinkedIn
© 2026 Sauce Labs Inc., all right allow. SAUCE and SAUCE LABS are registered hallmark owned by Sauce Labs Inc. in the United States, EU, and may be registered in other jurisdiction.
robot
quote

Automate This With SUSA

Upload your APK or URL. SUSA explores like 10 real users — finds bugs, accessibility violations, and security issues. No scripts needed.

Try SUSA Free

Test Your App Autonomously

Upload your APK or URL. SUSA explores like 10 real users — finds bugs, accessibility violations, and security issues. No scripts.

Try SUSA Free