Obsessed with technology?
Subscribe to the latest tech news as well as exciting promotions from us and our partners!
By subscribing, you indicate that you have read & understood the SPH's Privacy Policy and PDPA Statement.
Feature Articles

Investigating notebook battery life with PCMark 10's new battery benchmark

By Kenny Yeo - 22 Jun 2019

Introduction

Note: This article was first published on 29 May 2019.

The problem with battery tests

The great thing about notebooks is they let you work anywhere but not if it runs out of juice.

Testing any device’s battery is not as easy as it sounds, especially if it’s a notebook. To begin, a notebook is an extremely multifaceted device that can do many things. It can be used for work, it can play games, and it can show videos, and each task affects its battery life in different ways.  

And then there’s the matter of manufacturer’s claims. Each manufacturer measures and reports their notebook’s battery life differently, making it hard to draw any kind of meaningful conclusion.

Should we and can we believe manufacturers' claims? (Image source: LG)

Here’s ASUS testing methodology when quoting the battery life of VivoBook S14:

Battery tests conducted by ASUS in July 2018 using local video playback. Test settings: Wi-Fi enabled, display brightness set to 150nits, and volume at 50%.

And here’s HP for the Spectre X360. Very specific and detailed but different from others:

Battery life tested by HP using continuous FHD video playback, 1080p (1920x1080) resolution, 150 nits brightness, system audio level at 17%, player audio level at 100%, played full-screen from local storage, headphone attached, wireless on but not connected.

And then there’s Apple’s:

The wireless web test measures battery life by wirelessly browsing 25 popular websites with display brightness set to 12 clicks from the bottom or 75 per cent. The iTunes movie playback test measures battery life by playing back 1080p HD content with display brightness set to 12 clicks from the bottom or 75 per cent. The standby test measures battery life by allowing a system, connected to a wireless network and signed in to an iCloud account, to enter standby mode with Safari and Mail applications launched and all system settings left at default.

And some, like LG, are quite vague about how they measure battery life. This is the footnote for their LG gram notebook:

Battery life can vary, depending on the model, settings, and user environments based on the MobileMark® 2014 standard.

To be clear, there’s nothing wrong with the way their notebooks are tested — the test parameters used are entirely plausible. It’s just that they are inconsistent so there's no way of drawing any kind of worthwhile conclusion between each manufacturer’s claims.

 

Introducing PCMark 10's new battery benchmark

PCMark 8 was a useful tool for testing battery life but its parameters were too general.

So to address this issue, we need a benchmark that is consistent and produces repeatable results.

Our old battery testing benchmark, the Home workload in PCMark 8, was good in that respect. But its parameters are too broad and probably not entirely representative of most use cases.

It runs a workload that cycles through the following:

  • Web Browsing
  • Writing
  • Casual Gaming
  • Photo Editing
  • Video Chat

While web browsing and productivity are fairly common use cases, the remaining three are less so. Furthermore, they put a great strain on the battery, which gives us results that are a little skewed. While it’s fair to say that the final results represent the “worst-case scenario,” the same argument could be made that they are not entirely representative of normal use cases. That's because, as we will find out later, the worst-case scenario battery life and the battery life that you will normally get if you use a notebook sensibly is very different.

So what then are normal use cases? We have identified this to be web browsing, productivity, and video watching.

PCMark 10's new battery profiles aim to more accurately mimic real-world usage.

This brings us to the latest version of PCMark 10 which adds a much-requested battery test feature that was designed to more accurately mimic real-world usage by offering a couple of test scenarios.

These are the test scenarios and what they represent according to UL Benchmarks (the people behind this benchmark, also previously known as Futuremark) :

  • Modern Office: Battery Life for everyday work
  • Applications: Battery Life for Microsoft Office
  • Video: Battery Life playing back HD Video
  • Gaming: Battery Life while gaming
  • Idle: Runtime at idle

The ones we are most interested in its Modern Office, Video, and Gaming. There’s some overlap between Modern Office and Applications since the former includes web browsing and word processing, which is also tested under the Applications workload. However, Modern Office includes video conferencing which makes for a more varied run, whereas Applications solely runs productivity workloads which is too specific in our opinion.

The Video workload will give users a good idea of how long the machines will run if they only want to binge on movies and shows. Finally, the Gaming workload represents a worst-case scenario of how long the notebook would last if it was taxed to the maximum.

Loading...