Android browser 52% faster than iOS Safari in browsing speed test

Browser speed has always been a talking point in the never-ending mobile OS war(a war we hope doesn’t end anytime soon). According to Blaze, a website optimization company, Android’s native browser blows iOS’s Safari out of the digital water. After 45,000 tests, Blaze was confident in saying that the Android browser was faster to load a page 84% of the time. The tests focused on real world sites and shied away from using the classic benchmark tests, like the SunSpider Javascript test and Acid. Blaze wanted to see how the browsers stood up to real time browsing.

The test was done using an iPhone 4 and a Google Nexus S. All websites loaded were Fortune 1000 company websites, and the sites were loaded more than once, and on different days. The testing was done over WiFi, so carrier discrepancies can be ignored. There were multiple devices running the tests, and the median time of each site across the devices was used as the result.

Android beat Safari 84% of the time, and had a median load time that more than a second faster (2.144 vs. 3.254), making it 52% quicker. However, Android’s dominance wasn’t as pronounced on mobile sites, where the browser was only 3% faster. Despite the fact that both Android 2.3 and iOS 4.3 have much improved Javascript engines, this didn’t affect the overall load time relative to previous versions of the OS. So, while the Java benchmark tests would show marked increases in performance, it didn’t translate to real life advantages.

Blaze does acknowledge that a perfect page load test is practically infeasible. The amount of unique variables that factor into browsing performance is staggering, and trying to standardize each one would be a waste of time and resources. So, while this isn’t definitive by any means, it definitely favors Android, and there’s something to be said for that.

Update: Blaze has admitted that the test wasn't as accurate as they first thought. In fact, the entire test is based on custom browser apps that completely disregard performance updates to the Safari browser. Apple came out and said as much in response to the study. Blaze admitted that they were, indeed, using the custom browser instead of the actual mobile Safari. However, they've voiced disapproval of Apple's "embedded" browser, and updated their claim to say that Android's "embedded" browser is better than Apple's. That claim doesn't have nearly the same clout as before, and Blaze hopes that apple will update their embedded browser to enable the tests to be completed fairly.

Image Source: Technorotic

Report a problem with article
Previous Story

Internal Microsoft video shows general direction for next-gen gaming

Next Story

Windows Phone 7 Facebook application woes [Update]

56 Comments

Commenting is disabled on this article.

Kind of a pointless test. There are other better browsers than Safari to use on the same operating system.

jd100 said,
Kind of a pointless test. There are other better browsers than Safari to use on the same operating system.

Apple sells an "intuitive" device. In case you don't what Apple means by "intuitive", it means "Easy to use and understand because your too stupid to figure it out". Since that is the case, what makes you think your average person has any clue on what a Firefox, Opera is?

You maybe tech savvy but the general user is not. They will be using whatever stock browser came with there device.

Android > iOS .

Gotta love the fanboys, it's always flawed one way or another. I'm happy with my Nexus One, go and be happy with your iPhone and leave it at that.

here is update of this story: source - ars

"It turns out that Blaze didn't actually use Safari in its tests on the iPhone; instead, Blaze used its own iOS app, which Apple points out does not make use of any JavaScript performance increases that came with iOS 4.3. Blaze told CNet in a statement that this is indeed true, so keep this in mind when reading the results."

Does it mean that Android browser is 200% faster than Safari as it loaded flash video/ads as well during the tests while Safari had blank spaces (less data to load) ?

alexalex said,
Does it mean that Android browser is 200% faster than Safari as it loaded flash video/ads as well during the tests while Safari had blank spaces (less data to load) ?

Are you seriously asking that question? You have to compare apples to apples. So obviously Flash is disabled otherwise Safari is a god awful mobile browser.

Me running SE x10i default 2.1 and I find no difference in speed between Opera Mini and Android default.. But some how, personally i like using Opera mini

Can't say I'm surprised by the results. My girlfriend has an iPad and I was shocked at just how crap the browser is. The moment you start scrolling you just get the checkerboard; whereas on my Windows Phone 7 on the same sites it's very difficult to get anything other than smooth scrolling content.

According to Blaze's own documentation the “measurement itself was done using the custom apps which use the platform's embedded browser. This means WebView (based on Chrome) for Android, and UIWebView (based on Safari) for iPhone.”

UIWebView isn't using the Nitro engine in 4.3. Page loading will definitely be slower than they actually are in real world browsing with this speed test.

giga said,
According to Blaze's own documentation the “measurement itself was done using the custom apps which use the platform's embedded browser. This means WebView (based on Chrome) for Android, and UIWebView (based on Safari) for iPhone.”

UIWebView isn't using the Nitro engine in 4.3. Page loading will definitely be slower than they actually are in real world browsing with this speed test.

Of course, it's very important to point out that they had no idea about that limitation when they ran their tests. That only came out maybe 2 days ago.

giga said,
According to Blaze's own documentation the “measurement itself was done using the custom apps which use the platform's embedded browser. This means WebView (based on Chrome) for Android, and UIWebView (based on Safari) for iPhone.”

UIWebView isn't using the Nitro engine in 4.3. Page loading will definitely be slower than they actually are in real world browsing with this speed test.

Ahh, I didn't even read that part. So in fact that would mean Android's current browser is as fast as Safari without the new optimizations.

But then, that brings up an interesting point since Android added browser optimizations in Honeycomb, so really, all around, this is a last gen test. Here is an actual current gen test: http://news.cnet.com/8301-17938_105-20043455-1.html

Last gen and current gen, they are both neck and neck, and both VERY fast.

What a terrible article...Some of it's even inaccurate.

"Some Android disadvantages are; poor email support and email clients - Cannot move a message to another folder, for example, poor PDF reader as the text comes out garbled."

I have never had a problem moving emails in the mail application (synced to an Exchange 2007 account), or with the Gmail app. The PDF reader works for me. :\

Also, give us more specifics on the test - and make it visual! Graphs > Text. This article just flat out sucks. If a "tech site" is going to publish something, make it worth while.

tsupersonic said,
What a terrible article...Some of it's even inaccurate.

"Some Android disadvantages are; poor email support and email clients - Cannot move a message to another folder, for example, poor PDF reader as the text comes out garbled."
I have never had a problem moving emails in the mail application (synced to an Exchange 2007 account), or with the Gmail app. The PDF reader works for me. :\

Also, give us more specifics on the test - and make it visual! Graphs > Text. This article just flat out sucks. If a "tech site" is going to publish something, make it worth while.


me think u in wrong article, this is browser war article. u digg?

ApuBo said,

me think u in wrong article, this is browser war article. u digg?

Try clicking through to the source article ...

episode said,

Try clicking through to the source article ...


, well u still think we still in the this article,
Mobile browser war....

I imagine this is simply down to the more powerful processor in the Nexus S. The A4 is a year old, while the HUmmingbird is relatively cutting edge. I wouldn't consider this a comparison of the OS's, but rather the devices - they both use WebKit to render HTML, and that's most of what a browser will do unless loading a real "we app," which tend to be quite heavy on JavaScript.

A fairer test would be comparing an Android device running 2.3, from July/August of last year, to the iPhone 4 running iOS 4.3.

Simon said,
I imagine this is simply down to the more powerful processor in the Nexus S. The A4 is a year old, while the HUmmingbird is relatively cutting edge. I wouldn't consider this a comparison of the OS's, but rather the devices - they both use WebKit to render HTML, and that's most of what a browser will do unless loading a real "we app," which tend to be quite heavy on JavaScript.

A fairer test would be comparing an Android device running 2.3, from July/August of last year, to the iPhone 4 running iOS 4.3.

Galaxy S released on June 2010 is using Hummingbird as well.

Simon said,
I imagine this is simply down to the more powerful processor in the Nexus S. The A4 is a year old, while the HUmmingbird is relatively cutting edge. I wouldn't consider this a comparison of the OS's, but rather the devices - they both use WebKit to render HTML, and that's most of what a browser will do unless loading a real "we app," which tend to be quite heavy on JavaScript.

A fairer test would be comparing an Android device running 2.3, from July/August of last year, to the iPhone 4 running iOS 4.3.

I mostly agree with you and see your point. The hardware specs are beefier on the newer Android phones and that makes it difficult to discern the differences between javascript engines. However, when it comes to someone making a buying decision today they have the iPhone 4 which is still priced the same as it was last July/August competing against newer Android phones which are priced about the same if not cheaper in some cases.

Shadrack said,
I mostly agree with you and see your point. The hardware specs are beefier on the newer Android phones and that makes it difficult to discern the differences between javascript engines. However, when it comes to someone making a buying decision today they have the iPhone 4 which is still priced the same as it was last July/August competing against newer Android phones which are priced about the same if not cheaper in some cases.
I can definitely agree with this - I wouldn't even recommend buying an iPhone right now, they're reaching their end of life.

My real point is that there are too many variations to say "Android has a faster browser than iOS." The real point of this study is that "Some Android devices render web pages faster than the iPhone 4." This "study" would be just as accurate if it were comparing an iPad 2 to a Nexus S. It isn't a comparison of the OSes, but a comparison of 2 devices. It doesn't matter if you run a test 45,000 times; if you use the same two devices, your results will not be an accurate reflection of an entire OS.

To be clear, I'm not denying that Android's browser is fast, nor am I saying that iOS is infallible - it does, and it isn't. But if you're going to do a study, and claim that it's a sweeping analysis of two operating systems, compare the OS on as many devices as possible. Otherwise you're wasting money, time, and making yourself look foolish.

tanjiajun_34 said,
Galaxy S released on June 2010 is using Hummingbird as well.

Not only that but the Hummingbird and the A4 are the same CPU, so in fact it's the perfect test.

tanjiajun_34 said,
Galaxy S released on June 2010 is using Hummingbird as well.

Thanks for clarifying that. These Apple boys will come up with Amy excuse in the book. One of the reasons I love Android is because it isn't specific to any chipset and there's always a processor that out performs the Apple ( recently Samsung) chipsets. Choice FTW.

The hardware between the two devices is definitely comparable...Don't make excuses.

@Jmaxku, if only Samsung worked on the software department. Great hardware, crap software.

Jmaxku said,

Thanks for clarifying that. These Apple boys will come up with Amy excuse in the book. One of the reasons I love Android is because it isn't specific to any chipset and there's always a processor that out performs the Apple ( recently Samsung) chipsets. Choice FTW.

Recently Apple and Samsung chipsets have been nearly identical, closer than any two other SoCs out there. The CPUs are the same, the SoC is modified a little by Apple I think though.

Steeley said,
Woah! With that saved second I could come up with a cure for cancer or something!
Woah! You're so lucky to only have to load one webpage in your whole lifetime!

Whoah, damn, I am SO glad that I am not that busy, that waiting 1 second longer is actually bugging me! Talk about real world advantages.

This probably includes the flash animation which iOS would just ignore and Android would downlaod and render. Which is awesome for Android

Would love to see Chrome on Android, though; Dophin is fast, mini faster, Opera mini crap, and Miren awesome but phone gets way too hot.

dancedar said,

Would love to see Chrome on Android, though; Dophin is fast, mini faster, Opera mini crap, and Miren awesome but phone gets way too hot.

Yeah same here, the browser in honeycomb is much closer to the desktop chrome and natively syncs across all platforms. I'd imagine it's only time before this is integrated into the smartphone branch.

dancedar said,
This probably includes the flash animation which iOS would just ignore and Android would downlaod and render. Which is awesome for Android

Would love to see Chrome on Android, though; Dophin is fast, mini faster, Opera mini crap, and Miren awesome but phone gets way too hot.

What about IE9 on Android ? :-)

it definitely favors Android

It favours Android 2.3, and as we know from another article posted on the forums today, most people are running 2.2.

what said,

It favours Android 2.3, and as we know from another article posted on the forums today, most people are running 2.2.

Did you read:

Despite the fact that both Android 2.3 and iOS 4.3 have much improved Javascript engines, this didn't affect the overall load time relative to previous versions of the OS. So, while the Java benchmark tests would show marked increases in performance, it didn't translate to real life advantages.

All iPhones running 4.2 will be able to get 4.3. Only a handful of Froyo phones will be getting Gingerbread. Would be nice to see a comparison between iOS 4.3 and Android 2.2.

what said,
All iPhones running 4.2 will be able to get 4.3. Only a handful of Froyo phones will be getting Gingerbread. Would be nice to see a comparison between iOS 4.3 and Android 2.2.

There's a comparison between 2.2 and 2.3 with a 65millisecond difference in favour of 2.3. Reading the actual article does wonders

what said,
All iPhones running 4.2 will be able to get 4.3. Only a handful of Froyo phones will be getting Gingerbread. Would be nice to see a comparison between iOS 4.3 and Android 2.2.
Wrong, iPhone 3G won't

what said,
All iPhones running 4.2 will be able to get 4.3. Only a handful of Froyo phones will be getting Gingerbread. Would be nice to see a comparison between iOS 4.3 and Android 2.2.

Depends if you choose to root or not. Plus, new phones coming out will get 2.3.

Hollow.Droid said,

Did you read:

Despite the fact that both Android 2.3 and iOS 4.3 have much improved Javascript engines, this didn't affect the overall load time relative to previous versions of the OS. So, while the Java benchmark tests would show marked increases in performance, it didn't translate to real life advantages.

Except for that its blatantly not true. 4.2.1 vs 4.3 and surfing the same pages at the same time, the 4.3 is MUCH faster.

what said,

It favours Android 2.3, and as we know from another article posted on the forums today, most people are running 2.2.


what said,
All iPhones running 4.2 will be able to get 4.3. Only a handful of Froyo phones will be getting Gingerbread. Would be nice to see a comparison between iOS 4.3 and Android 2.2.

Actually although we haven't had definitive numbers from many manufacturers, it seems that the majority of devices running 2.2 now are getting 2.3. Certainly the most popular devices - The Nexus 1, HTC's Desire range and the Galaxy S line are all getting it.

what said,

It favours Android 2.3, and as we know from another article posted on the forums today, most people are running 2.2.

Nope. 2.2 was placed against 4.2 and still beat that pants off it. Android has been the front runner in web browsing since Froyo. Sorry.

what said,
It favours Android 2.3, and as we know from another article posted on the forums today, most people are running 2.2.

I don't think they even had any browser performance enhancements between Froyo and Gingerbread. Gingerbread's differences are fairly minor and low level (mostly support for new hardware and new development tools).

episode said,

Except for that its blatantly not true. 4.2.1 vs 4.3 and surfing the same pages at the same time, the 4.3 is MUCH faster.

Source/statistics for this? Would be interesting to see.

Hollow.Droid said,

Source/statistics for this? Would be interesting to see.

My phone vs my business partners phone, side by side.

what said,
All iPhones running 4.2 will be able to get 4.3. Only a handful of Froyo phones will be getting Gingerbread. Would be nice to see a comparison between iOS 4.3 and Android 2.2.

fail! iPhone 3G isn't getting 4.3 - well not yet anyway

what said,

It favours Android 2.3, and as we know from another article posted on the forums today, most people are running 2.2.

You all miss the point of this test. It's not the devices or browsers that are over 50% faster on Android. Think about it.......!!! If they'd tested Jailbroken iOS Devices against Android, it would have been more even. The simple truth is that; mobile web is not the REAL WEB and content is much larger and complex. But that's not all..... iOS encrypts and tunnels all data through their Garden Walled Network on one single port. It's called Hypertunneling and means it's all data is channeled through Apple's own Network like AOL or Compuserve used to be and Xbox 360 is today. It's the price to pay for Security of your locked in a FISHBOWL Existence!

Don't like it? .....then your only choice is Jailbreaking. Because what Apple is not saying is that they aren't allowing any tests to be run on the native browser...... BECAUSE they are afraid the results will be just the same! ...no because they KNOW the results would be just the same!!! ^_*