BenchmarkXPRT Blog banner

Category: Android

Kick the tires. Look under the hood.

Today, we released the source code for the MobileXPRT 2013 CP to the community (community login required). If you already have an Android development environment set up, building the benchmark is very straightforward. If you do not have a build environment, it’s not difficult to create one. All the necessary software is free, and the build documentation includes detailed instructions for creating the build environment on Windows 8.

Of course, you can also build Android applications using Linux or Mac OS X. The instructions include links to the information you need to set up those environments as well.

If you have suggestions about how we could make the application better, or you write some code you would like to submit for inclusion in the benchmark, please contact us at benchmarkxprtsupport@principledtechologies.com.

Bill wrote about a presentation he gave to Mobile PC Extended Battery Life Working Group (EBL WG) in the post Presentation XPRT. We’ve put the slides from that presentation up on Slideshare. Check them out!

Eric

Comment on this post in the forums

It’s finally here!

As the community members know, we released the MobileXPRT 2013 community preview yesterday. As I said in last week’s blog A look into the future, you are free to publish results from this community preview.

We hope you’ll submit results to be published in the MobileXPRT database. You’ll find the file containing your results in the MobileXPRT/results directory on the SD card of your device. The results directory will contain a time-stamped directory for each run, for example ALL28May2013112522. You will find a file named mobilexprt_results.xml in the appropriate time-stamped directory. To submit a result, just e-mail the mobilexprt_results.xml file from your run to benchmarkxprtsupport@principledtechnologies.com.

The benchmark includes a user manual in the UI. In addition to the benchmark itself, you can also find installation instructions and a design overview in the member’s area.

We will make the source code for the community preview available in the next few days.

Given the diversity and changing nature of Android devices, we expect that you may run into some issues. As with any of our community previews, if you encounter any issues, please let us know so we can continue to improve the benchmark as we prepare it for release.

We’re very excited about the release of the MobileXPRT 2013 community preview. If you’re not a member, join and find out what all the excitement is about!

Eric

Comment on this post in the forums

A look into the future

This week, we’ve been testing builds of the upcoming MobileXPRT benchmark. So far, things have been looking really good.

MobileXPRT is designed for testing Android-based devices, from smart phones to tablets. Like TouchXPRT, MobileXPRT uses native code to implement real-world use cases. Mobile XPRT offers two categories of tests: Performance tests and User Experience (UX) tests.

The five Performance tests are Apply Photo Effects, Create Photo Collages, Create Slideshow, Encrypt Personal Content, and Detect Faces to Organize Photos.

The five UX tests are List Scroll, Grid Scroll, Gallery Scroll, Browser Scroll, and Zoom and Pinch.

The benchmark provides a score for each individual test, as well as a rollup score for each category.

Sometime next week, we plan to release the MobileXPRT 2013 Community Preview. The CP, as its name makes clear, is not the final MobileXPRT release. It is, though, a useful tool for beginning to measure Android device performance. It is also a great way for everyone in the community to see the current state of our thinking and to provide feedback—rather than reading a design spec, you can actually run this version of the tool and see what you think!

As with TouchXPRT, we will allow publication of scores from the community preview. We will also be releasing the source code  to all community members. If you’re curious about not just what we’re doing but how we’re doing it, you can find out. If you have suggestions about how to do it better, we’re all ears.

As a reminder, the community previews are available only to members of the BenchmarkXPRT Development Community. If you haven’t joined yet, come join us!

Eric

Comment on this post in the forums

Three names, two hosts

As Bill mentioned a couple of weeks ago in The Name Game, we’ve been considering changing the name of PhoneXPRT. The rationale for this is that the tests in PhoneXPRT are useful for a range of devices, from phones to tablets. We asked for your opinions about the name. After getting your input and talking amongst ourselves, we are considering three possibilities:

  • Leaving the name unchanged.
  • Changing the name to MobileXPRT. While this would convey the scope of the benchmark, some people thought the name might be too general.
  • Changing the name to TouchXPRT for Android. While there is some similarity between the tests in TouchXPRT and the ones we are developing for this benchmark, the two benchmarks would not initially be comparable.

Let us know what you think. We hope to settle the name question soon.

As we mentioned in Loose Ends, some users in mainland China are reporting extremely slow download times when running WebXPRT. We have set up a trial host for WebXPRT in Singapore to see if this improves the situation. Preliminary, US-based tests have seen no significant difference in scores when running from the Singapore host.

If you are in China and want to try running WebXPRT from the new host, you will find it at http://54.251.252.204/webxprt/. Please let us know your experience.

Eric

Comment on this post in the forums

Rapid Evolution

As Bill mentioned last week, we are considering changing the name of PhoneXPRT, not only because of the interest in using the benchmark scenarios on Android based tablets, but also because the line between phones and tablets has become blurred.

Devices that are too big to be a phone and too small to be a proper tablet are everywhere. PC Magazine says the first true “phablet” – possibly the ugliest portmanteau in the history of technology – was the AT&T EO 440 in 1993. However, it was more 8 years before Samsung had the first really successful phablet, the Galaxy Note. Now, less than 2 years later, there are rumors that Samsung may kill the Note in favor of the Samsung Galaxy Mega.

Obviously, this is one of the fastest evolving areas in tech. This rapid evolution has given us an almost bewildering array for devices, from small phones such as the Sony Xperia Mini, which a child can hold in one hand, to the ASUS Transformer AiO P1801, which has a whopping 18.4” screen! All this speed and diversity obviously pose challenges for the new benchmark, but it makes the work very exciting as well!

We have received comments about the name and we really appreciate those. If you have any thoughts, let us know. We hope to make a decision about whether to change the name soon.

Eric

P.S. I should note: The Transformer AiOP1801 also runs Windows 8, which means it’s a great candidate for TouchXPRT as well.

 

Comment on this post in the forums

The name game

There’s a lot going on in the world of the XPRTs. We’re working on HDXPRT 2013, writing white papers, building up our results database, and thinking ahead to the next versions of TouchXPRT and WebXPRT. At the same time, we are giving quite a bit of attention to PhoneXPRT.

As we said in the PhoneXPRT press release last month, PhoneXPRT will use the same kinds of realistic scenarios the other XPRT benchmarks do. These scenarios include tasks that people perform on phones as well as other mobile devices.

We’ve gone back and forth on the seemingly simple question of how to define what a phone is. At one point, defining a phone as a mobile device that you put to your ear seemed to suffice. As phones grow larger and tablets smaller, that doesn’t really hold up.

We also have a decent bit of interest in using the benchmark on Android-based tablets as well as phones. That seems like a good idea to us, as we are all for getting the most out of any benchmark.

One issue, however, is that PhoneXPRT is not a great name for a benchmark that may be commonly used on devices other than phones. So we’ve started thinking about what else we could call it.

As always, we look to the community. Do you think calling the benchmark PhoneXPRT would limit its usefulness for benchmarking tablets? Do you have any ideas for more inclusive names? We really need your feedback here and look forward to getting it. Please send suggestions to benchmarkxprtsupport@principledtechnologies.com or post your comments in the forum. Thanks!

Bill

Comment on this post in the forums

Check out the other XPRTs:

Forgot your password?