At first we would like to thank BAPCo for supplying us an iPAD tablet in order to test and review their latest TabletMark v3 software.

 

 

 

About BAPCo:

 

“A non-profit consortium, BAPCo's charter is to develop and distribute a set of objective performance benchmarks based on popular computer applications and industry standard operating systems. Governments all over the world use BAPCo benchmarks to assist in product purchasing decisions.”

 

Product Features:

 

Support for Microsoft® Windows® 8.1, Android, and iOS

Touch capable user interface

Ability to execute multiple iterations

Web & Email Scenario

Photo & Video Sharing Scenario

Video Playback Scenario

Battery Life Test

Automated Runs

 

System Requirements:

 

All platforms require 7″ or greater screen size

 

Windows 8.1 or 8.1 RT

 

1.3 GHz dual core processor (AMD, ARM or Intel)

2 GB system memory

 

iOS 7.1.2 or newer

 

A6X processor or newer

1 GB system memory

 

Android 4.3 or newer

 

Dual core processor (ARM or Intel)

1 GB system memory (500 MB free memory)

 

TabletMark v3 is the latest version of the benchmarking software from BAPCo and it aims to measure the user experience for light productivity and multimedia. As we have seen with the latest PCMark software from Futuremark, we have multiple test scenarios:

 

-Web and Email

-Photo and Video sharing

-Video Playback

 

The software is available for Windows 8.1 and 8.1 RT, Android 4.3+ but also for iOS 7+ OS versions. As extra requirements we would remind 2GB of RAM and at least 2GB of available storage space. Tablets with screen diagonals of 7+ inches are accepted and the minimal supported resolution is 1024x768. The application also supports the latest 64-bit ARM and x86 CPUs.

At the moment of the review, the iOS version of the benchmark is not yet available to the public via App Store, but it should appear there later on.

After downloading the software, we will see the GUI of TabletMark, which includes the currently installed version, the selection of test categories, the name of the running project which we can fully customize and also the number of iterations. From the same area we can select if we will perform a conditioning run or a battery rundown. To the right we have the Results tab, one Help tab but also the About tab:

 

 

(click to enlarge)

The results tab is split into multiple columns like Project Name, running Date, TabletMark rating, Battery Life Rating but also Submission Status:

 

 

(click to enlarge)

Let’s get a bit in more detail regarding the included test scenarios; first we have Web and Email, which includes simulated Web Browsing with 34 page loads and includes HTML 5, CSS, JavaScript technologies but also video tags.

 

 

 

 

 

The Email test scenario includes working with a 70MB mailbox file (like Outlook PST), DEFLATE compressed, AES CBC 128 encrypted and the JPG attachments are BASE64 encoded. We also have an editing scenario of small text files:

 

 

 

Next, we have Photo and Video Sharing testing scenario; here TabletMark will simulate Photo Editing environment with 8x 14 megapixel photos, 3x 7 megapixel HDR photos (3 exposures each) with native C code, shared among platforms. The next stage implies applying filters like Brightness, Contrast, Saturation and Tint:

 

 

 

Video Editing simulation is also available at this stage and the application works with 1920x1080 H.264 8000 Kbps video with AAC or MP3 128 Kbps audio. Two video trans-coding projects are worked on:

 

-40MB clip made up of 4 other clips

-13MB clip made up of 2 other clips with B&W filter

 

 

 

Lastly, we have Video Playback which does not affect the final calculated score and implies 3x 1-minute 1920x1080 MP4/AVC clips (which have about 60MB each) which are looped three times:

 

 

 

After the test runs have been completed, we will be presented with a full-disclosure report, with the possibility of PDF and XML export and online submission. An online results database is also in development at the moment. Each scenario score is an aggregate of all measured operations relative to the reference machine which receives a score of 1000 (Lenovo ThinkPad 10, Windows 8.1 32-bit):

 

 

(click to enlarge)

If we are planning on doing a battery life test, we’ll have to enable the Battery Rundown checkbox, adjust the brightness of the tablet to the setting we feel most comfortable and leave it running. TabletMark will attempt to run the maximum possible number of iterations and afterwards it will display the Battery Live result; as far as we have seen with our iPAD Air 2 sample, the overall screen brightness seems to influence quite a lot the final battery score, so we would suggest the manufacturer to set a fixed value instead right before starting the tests, in order to be able to perform an easier comparison as soon as the online database will be available:

 

 

(click to enlarge)

To be more precise, before performing a battery life test, we need to have the specific tablet charged to >97% before the test can begin; the 50-minute script will be looped continuously until the battery dies with the following stages:

 

1. WE = Web Browsing, Email, Notes

2. 3 minute idle (screen on)

3. PV = Photo Editing, Video Editing

4. 3 minute idle

5. Video playback (9 minutes)

6. Idle to 50 minute mark (about 8 minutes)

7. Repeat

 

The new TabletMark benchmark from BAPCo is one interesting software each hardware enthusiast should check out, which is offered free of charge and thanks to the community the database will grow consistently in order to have comparisons between lots of tablet models which are now present on the market from different manufacturers.

Unlike some other benchmark which model a single task by reflecting only one aspect of system performance or benchmarks which are built around synthetic workloads, which are not always accurate, TabletMark aims to use a variety of simulated workloads and data sets chosen by the application experts in an effort to reflect much closer the performance a regular user may expect when performing similar activities.

Remember our E-CEROS Revolution 2 we have tested a while back? Well, we have got the chance to run the Android version of the TabletMark v3 which is currently available on Play Store; while we were able to run the full test successfully, the result section was not able to produce an Overall Performance Rating score and TabletMark has displayed the following error: “Found outputs not within expected range”.

 

 

(click to enlarge)

TabletMark uses for output validation file-size checks which are performed on key workload output files every iteration; the overall score is to be invalidated if failures are found.

 

Without a doubt, we will continue including this type of benchmark with the new tablets we will receive as samples and build our own database right here, inside the article.

 

 

 

We would like to thank again to BAPCo for supplying us with an iPAD Air 2 sample in order to be able to feature this article.