Internet Explorer still provides the best laptop and tablet battery life

While Internet Explorer is often treated as the red-headed step child of web browsers, there is one area in which Microsoft's browser has traditionally reigned supreme, battery life.

Microsoft's own research has demonstrated IE's power efficiency, and recent test results from the website 7tutorials have once again proved that IE enables user to eke out the maximum amount of battery life from their Windows laptops and tablets.

This particular test relied on the Peacekeeper battery test and was performed across a variety of devices including a Toshiba Encore 8" tablet, a Surface Pro 2, and an older HP Pavilion laptop. Each round of testing compared the performance of the latest versions of Internet Explorer (touch and desktop), Firefox, Chrome, and Opera web browsers.

On the Toshiba Encore, the touch version of IE trounced the competition with a run time of 8 hours and 52 minutes, compared to 7:45 for the desktop version of IE. The other three browsers produced similar results with approximately 6 hours of battery life.

The battery test on the Surface Pro 2, however, produced some rather odd results. The touch version of IE exhibited a battery run time that was an hour less than the desktop version of IE, an inversion of the results from the previous test. It even performed worse than the Firefox and Opera web browsers. These results are especially confusing given that both the Encore and Surface tablets run full (non-RT) versions of Windows 8. It appears that there are still some Surface Pro 2 firmware issues that remain to be addressed.

Source: 7tutorials images via 7tutorials

 

Report a problem with article
Previous Story

Bill Clinton will be speaking at Microsoft's SharePoint Conference 2014

Next Story

Xbox One 2014 plans reportedly leak, white and cheaper models inbound

37 Comments

Commenting is disabled on this article.

Chrome actually is a CPU hog, it also sets the Platform Timer Resolution to the minimum (1ms) which impacts battery life severely. IE11 leaves the Platform Timer Resolution at 15.6ms unless when you play video, when it sets the Platform Timer Resolution to 10ms which is still nowhere as bad as Chrome.

Now I'm a bit mad at Neowin. I know it's unprofessional journalism but to select data like that is quite disturbing. I know you are typically Microsoft biased, but to select a part of the test results to make a point is very disturbing. In fact, there were actually three computers tested, not only the two presented, and on two of them, the touch version of IE performed worst, when the Neowin article suggests that bad touch performance is a fluke on only one of the tests due to bad firmware. The results for the other browsers are inconclusive at best, Firefox seems to beat Chrome in two out of three tests, and IE desktop seems to beat every other browser, but while these separate statements seem to be confirmed by data, the actual results aren't consistent: the most energy efficient rendering engine is, in two of the tests, both the best and worst performer; the second best performer in one test scores last in other two; Firefox is either the second best or the second worst,...
In conclusion, given the uncertainty, the best one can conclude is that battery life varies by about 20% in all browsers, with no clear winner, at best a suggestion of better IE desktop performance. The test has clearly conditioning issues.
I'd suggest the author to (1) present all data and (2) not suggest that the bad IE touch results are a fluke, and limit itself to presenting IE desktop as systematically the best of the pack, arguably the only good conclusion one can make of that test, though not worth much given the issues stated with the methodology.

Okay, I will address the issues I believe that you have brought up in no particular order.

First, I made no attempt to hide the fact that this test was carried on three different machines. I make a clear reference to this aspect of the test. I decided not to include the third graph because I believe the results were not very interesting. The author of the original article also appears to feel the same way because he stated that "on a laptop with a very poor battery, using a browser vs. another doesn't help much with saving power." Of course, I included a link to the original article in case people did care to see the results.

Second, your claim that on two of the machines "the touch version of IE performed worst" is not true, although close to true. Only on the last machine, the HP laptop, did the IE version of touch provide the least amount of battery life. It performed better than Chrome on the Surface Pro 2.

Third, the information from the test as a whole confirms that IE, one version or the other, provides the best laptop and tablet battery life. IE desktop does not "seem" to beat every other browser. In these test, it does. And on one of the tablets, IE touch provides a very significant increase in battery life. I do not believe it is disingenuous to refer to both IE desktop and IE touch, as I do in the headline, as Internet Explorer.

I also stand by my conclusion that the inversion of results from the Toshiba Encore to the Surface Pro 2 is "odd." Both fairly modern machines are described as running Windows 8.1 but turn in different scores. Clearly the hardware in both machines is the differentiating factor. Unless there is something inherently different about the Core i5 Architecture, I don't see why a firmware update could not provide better battery life when using IE touch.

Thank you for the thoughtful comments.

Frustrating as many sites have issues with IE 11 edge DOM JavaScript and IE lacks plugins and is years behind W3C support compared to Chrome.

However Chrome is having issues now and flash hardware acceleration is disabled on my system where CPU usage skyrockets to 30% when I listen to music on youtube.

Firefox does not offer low rights mode, sandboxing, or per cpu/tab threading right out of 2008

Browsers are regressing again sigh.

sinetheo said,
Frustrating as many sites have issues with IE 11 edge DOM JavaScript and IE lacks plugins and is years behind W3C support compared to Chrome.

However Chrome is having issues now and flash hardware acceleration is disabled on my system where CPU usage skyrockets to 30% when I listen to music on youtube.

Firefox does not offer low rights mode, sandboxing, or per cpu/tab threading right out of 2008

Browsers are regressing again sigh.

I also would like to see IE move faster with standards; however, there are reasons. Microsoft has a bad history with implementing not finalized standards and getting burned by them. IE5/6 for example.

The extended development time also allows Microsoft to ensure quality of code where they want it.

Which is at the heart of the argument; would you rather have Chrome being buggy and supporting more of the W3C or would you rather have it run well like IE seems to be doing?

The only reason I upgraded to 11 is because IE 10 crashes by the hour. It feels like IE 7 all over again.

Chrome I have to use beta as no menus work on the task bar if you do this on a freshly imaged computer! Seriously a new chrome install is broken by default if it has never been installed. I reimaged and same problem appears.

But since IE 11 is so incompatible I can't use it. I am going back to Firefox even though everyone who uses it eventually gets malware or chrome beta.

I haven't had these many issues since I had Netscape and IE 6 to choose from

sinetheo said,
The only reason I upgraded to 11 is because IE 10 crashes by the hour. It feels like IE 7 all over again.

Chrome I have to use beta as no menus work on the task bar if you do this on a freshly imaged computer! Seriously a new chrome install is broken by default if it has never been installed. I reimaged and same problem appears.

But since IE 11 is so incompatible I can't use it. I am going back to Firefox even though everyone who uses it eventually gets malware or chrome beta.

I haven't had these many issues since I had Netscape and IE 6 to choose from

If you are having crashes, you may want to consider there is a bigger problem than IE10/11.

The first place to look is hardware, and if you are having crashes in other software that may be pushing the system. For example, IE shoves the GPU more than other applications.

See if anything is overheating, see if your system can pass a few memory or burn in tests, etc.

Also try a 'test' clean install of your OS with only IE10/11 and see if the same crashing occurs. (Heck even grab a evaluation version of Windows 8.1 and see if it does better.)

If a clean install with only IE10 added and it still crashes, you definitely have a hardware problem.

Also check the sites you are having problems with IE11 on, as it does a better job than IE10 does. See if you have adblocker or some other network level malware filter/protection running that could be altering the site code to make it display improperly.

A first party browser whose components are part of and preloaded by the operating system uses less power than third party browsers whose components are loaded in additional to the first party's...

And in other news, Bears like honey...

simplezz said,
A first party browser whose components are part of and preloaded by the operating system uses less power than third party browsers whose components are loaded in additional to the first party's...

And in other news, Bears like honey...

Except IE on the desktop runs in the same capacity. Its code isn't preloaded (which would only affect startup times) and runs just like any other Browser has the ability to run.

IE doesn't have some secret kernel level code running all the time - trust me.

Order_66 said,
"Microsoft's own research"

Well no surprise there.

/s

This isn't rocket science, run your own tests, all you need is a notebook.

The results above are fairly close to our own testing on various tablets/notebooks.

Only reason I use IE at all is because of Windows 8.1 with a touch screen on a laptop. It just performs much better with those gestures.

Otherwise, it doesn't get used at all, as I dislike ads as well as the loss of functionality between what I can get on Chrome or Firefox.

Raa said,
You don't use an adblocker for IE (there are a number of options)?

I do, but I also lose functionality between other addons as mentioned. I still use IE, it's just that I only use the app version of it on my Yoga since it works great with touch screen laptops. For quick browsing, works in a pinch. When I'm sitting down to do real work or long-term browsing though, it's just not up to snuff for me.

TPLs are not perfect. It can't do much about ads which originate from the same domain. Yes, I've tried the EasyList TPL with limited success.

Adblocking options are limited. The best IMO is Adblock IE, but it's $20 for a license. The IE Adblock Plus is buggy - some sites crash the browser tab.

In short, the OP has a point - the options are limited and aren't great especially when compared to the solutions offered on competing browsers.

Denis W. said,
TPLs are not perfect. It can't do much about ads which originate from the same domain. Yes, I've tried the EasyList TPL with limited success.

Adblocking options are limited. The best IMO is Adblock IE, but it's $20 for a license. The IE Adblock Plus is buggy - some sites crash the browser tab.

In short, the OP has a point - the options are limited and aren't great especially when compared to the solutions offered on competing browsers.

It is still better to have a tracking protecting list running in the browser code, than a 3rd party chunk of code.

If the user needs more than a TBL, look at the Firewall tools in Windows (there are also 3rd party that attach to the Windows Firewall if the advanced settings are too much for a casual user.)

Doesn't surprise me. Safari yields better battery life on MacBooks than 3rd party browsers in my experience as well. 1st party optimized is hard to beat by 3rd parties...

Not surprised by the results either. The two Blink browsers - modern Opera and Chrome - tend to spike the CPU quite frequently.

The tradeoff for IE is that pages don't render and draw as fast as Opera/Chrome, and scrolling performance is kinda crap on pages like long Neowin forum threads. On the upside, I've noticed it suspends any updates for tabs in the background, on top of not revving up the CPU fan as frequently.

Denis W. said,
...The tradeoff for IE is that pages don't render and draw as fast as Opera/Chrome...

In Browser benchmarks, specifically rendering and drawing is where IE 9/10/11 consistently beats Opera/Chrome/Firefox. Sometimes by a factor of 50-100 times faster.

Head over to: http://ie.microsoft.com/testdrive/Views/SiteMap/

Yes these are made by Microsoft but are 100% standards compliant with a focus on very rich web site content examples. (The slow times you will find in non-IE browsers is why RIA is still struggling to emerge on the web, as it still runs too slow.)

Run some tests in IE and other browsers, especially drawing or graphical based tests. Notice how well IE11 does, also notice (if you can) on older hardware, the demos still run respectably.


Historical Note:
IE11 now supports WebGL, which ironically gained popularity because of Chrome trying to compete with the graphical IE Fish demo running on IE9.

At the time, Google couldn't get close to the FPS/Speed that IE9 was doing in CSS/HTML, so they created their own IEFish demo, except they wrote it in WebGL and had it render to an HTML5 canvas. (Google was cheating by using the OpenGL rendering instead of their own; however, it did help WebGL to become popular.)


**Side Note to get better browsing Speed out of IE...
A lot of sites have background ads and junk, so turn on at least a basic protection list. You can always disable them site by site if content, like a Google Ad doesn't display that you want.

Open IE on the Desktop
Click - Tools...
Click - Manage Add-ons
Click - Tracking Protection
Down Below Click - Get a tracking protection List online...

This can make a huge difference in browsing speed, and since you can't install AdBlock on IE11 Touch/Modern, you get the basic benefit.

(Be sure to turn them off on sites you support, like Neowin.)

My point about drawing speed still remains on IE. Here's a demo I whipped up months ago. On Blink browsers you don't see that kind of repainting. I should note that on IE10 it was fine, but since IE11 scrolling performance has tanked.

http://www.youtube.com/watch?v=ha1v16zqX_c

Those drawing demos are nice, but to be fair all browsers do tests that cater to their own and show how better they are. My daily browsing habits take me to Neowin, amongst other sites, not fancy drawing demos.

Denis W. said,
My point about drawing speed still remains on IE. Here's a demo I whipped up months ago. On Blink browsers you don't see that kind of repainting. I should note that on IE10 it was fine, but since IE11 scrolling performance has tanked.

http://www.youtube.com/watch?v=ha1v16zqX_c

Those drawing demos are nice, but to be fair all browsers do tests that cater to their own and show how better they are. My daily browsing habits take me to Neowin, amongst other sites, not fancy drawing demos.

Working backwards...

I agree, it is our daily browsing habits that define what work best for us. However, the bigger point is that a lot of what CSS3 and HTML5 promise could have been in use several years ago if Chrome/Safari/Opera/Firefox would have moved forward.

IE can run fairly rich internet applications that the other browser simply cannot; thus, the progression of technology and more specifically developers are being held back.

IE since version 9 does use a completely different model than the other browsers. Not only in practice, but in how was designed conceptually.

It treats even a few words on an HTML page like a page of code and then executes that code. IE8 and the other browsers treat a few words on an HTML page like a document that it then displays.

Now, at the basic HTML text level, the performance gain is nonexistent; however, on a very complex 'dynamic' CSS page with a lot of bitmap and vector graphics, IE takes all the content and converts a bunch of it to fairly native code and allows it to execute, with the 'compiled content' shoving CPU and GPU instructions directly, even allowing vector level rendering updates directly to the composer.

This is how IE gets more out the CPU and the GPU than just the Direct2D rendering that Firefox also used or the OpenGL minimal acceleration Chrome uses.

The irony is that Google is using Chrome like an Application platform, yet the base browser technology is still effectively treating content like a document(PD) and rendering it.

Google of ALL companies should be following the IE model of how to treat content; however, they don't want to write it themselves from the ground up or abandon their investment in WebKit which will probably never approach web content like IE does. -This includes their WebKit break with Blink, as it is not redesigning the model and is still just a document/content display engine.

We have been somewhat rich on CPU and GPU processing power for the past few years, with little more than a few engineering titles and games pushing the needs of our current hardware. This will end, and Chrome will be one of the most prominent and hardest hit products unless Google or people working on WebKit wake up.


As for the video link, this can be the result of a couple of things; however, I honestly can't replicate it. Even on a throttled connection on an Acer One Z270 Atom/945, I can't produce the rendering/scrolling lag under Windows 8.1.

(If others can confirm this, or if anyone at Neowin has insight to the site code that might explain why it is having trouble, I am all ears. If there are specific steps to get this to happen, I am curious. -Neowin should be as well if it is still a problem.)

It could be as simple as the site formatting the content for a browser it doesn't properly recognize, or even 'ads' in the site failing to properly identify the browser an creating delays to ensure the ad becomes visible. (Or 50 other things.)


I have to agree, if I was seeing this type of problem, I would agree that using IE11 was painful and complain really loudly to Microsoft. If someone can help me reproduce, I will.

Denis W. said,
My point about drawing speed still remains on IE. Here's a demo I whipped up months ago. On Blink browsers you don't see that kind of repainting. I should note that on IE10 it was fine, but since IE11 scrolling performance has tanked.

http://www.youtube.com/watch?v=ha1v16zqX_c

Those drawing demos are nice, but to be fair all browsers do tests that cater to their own and show how better they are. My daily browsing habits take me to Neowin, amongst other sites, not fancy drawing demos.

The scrolling is a mouse software issue. I had the same problem on Windows 7 with IE 10 if I install the mouse software for my Logitech Performance MX. I have tried both with and without Smooth Scrolling Enabled in IE and I ended up uninstalling the software and now everything works fine. The issue is because Logitech has an option for smooth scrolling and also installs an add-on called something like scrolling browser helper (which even when both are disabled still cause problems).

To be honest I have also noticed it. Even from the fan speed spinning I can tell that Chrome does really bad compared to IE. Better battery life and cooler PC are probably one of the main reasons I use IE as my main browser.