Posts Tagged ‘Browser’

iOS 6 SunSpider benchmark

September 19, 2012

Apple released new hardware such as the new iPhone 5 and the new iPod Touch, but for current customers there’s also some goodies: the new version of iOS is now available for the older generations of the iPhone, iPad and iPod Touch.

One thing that always impresses me is the download of the update itself: despite millions of people all around the world receiving their iOS update at the same time, my fiber optics connection bandwidth was maxed out downloading the update in a couple of minutes from the Apple servers.

After installing the iOS 6 update on my iPad 2, I ran the SunSpider javascript benchmark in order to look for performance improvements over the previous version of iOS.

As you can see above, Safari completed the benchmark in just 1417.2 ms which, compared to the iOS 5.1 result at around 1800 ms is a huge improvement over the previous iOS version: now that’s just amazing!

iOS 6 might not bring many new features compared to the previous versions, but Apple seems to be unstoppable perfecting the current ones: the browser javascript performance has improved over and over in each version and that is what keep Apple users happy with their devices.

When most competing products usually have a decrease in performance after major software updates (e.g.: Sony Ericsson Android ICS update), Apple just keeps improving the older products performance in each software update until every last bit of performance is extracted from the hardware!

Chrome vs stock Android Browser

September 7, 2012

Since I’ve recently installed Android ICS on the Sony Ericsson Live with Walkman, I’ve been able to install the Google Chrome mobile web browser that’s available in the Google Play store.

Chrome brought the fast V8 javascript engine to the Android platform, so it’s supposed to be much faster handling benchmarks like the SunSpider test.

As I wrote in the previous blog post, the Android ICS stock browser did it in 3099.8 ms, which was an improvement over the Gingerbread stock browser.

I’ve now run the same benchmark in Chrome.

As you can see above the test was completed in 2886.9 ms. Although there’s a definite improvement over the stock browser, the 200 ms advantage is not as big as I thought it would be.

I believe Chrome will eventually replace the current browser and become the default one in future versions of Android.

Sony Ericsson Live with Walkman performance with Android ICS

August 30, 2012

Sony Ericsson promised that all the 2011 Xperia line along with the Live with Walkman model would get updated to Android 4.0 Ice Cream Sandwich.

They took their time and the carriers also took an extra time before releasing the firmware to the carrier-locked devices. Recently I was finally able to go through the ICS update process with the Live with Walkman I’ve posted a short hands-on a while ago.

The update process is pretty straight forward as long as you know how to use a computer, since Sony Ericsson (now only Sony without the Ericsson part) demands the user to connect the phone to the PC Companion Windows software or to the Bridge for Mac app in order to get Android 4.0. Apparently the update it too big to be handled as a simple OTA (over the air) update!

Sony also warns on their website that Android 4.0 is heavier than the older 2.x versions and that might have a negative impact on how the smartphone performs.

I’ve installed the official Android 4.0.4 update and ran the Quadrant benchmark to check if the update had any impact on the performance figures.

The score is, as the manufacturer warned, lower than before: 1381 is indeed not as good as the score Android Gingerbread achieved, around eighteen hundred points.

I’ve run the test twice to see if there was any improvement in the second run but, as you can see below, it just got a little bit worst, achieving 1362 points: disappointing!

I also ran the SunSpider web-browser javascript benchmark and fortunately, unlike the Quadrand score, the results improved as you can see below.

Now the test is completed in 3099.8 ms which is better than the old result (3342.7 ms)!

Overall, and despite the bugs and the performance hit, I do recommend updating because newer apps require Android 4.0 ICS to run, like Google’s own Chrome web browser!

Firefox 14 lacks Android 3.x HoneyComb support

June 28, 2012

Mozilla just launched Firefox 14 for Android and I was eager to discover if the latest version was any good.

Unfortunately, unlike what’s written in the system requirements webpage, Firefox 14 is in fact not available for the Samsung Galaxy Tab 10.1 running Android 3.2 as you can see in the picture below.

This might be related to what some people call “the Android fragmentation”, but it’s really unfortunate that a less than one year old Android tablet is not able to run the latest software like Google’s own Chrome browser…

Firefox 13 for MeeGo Harmattan

June 25, 2012

Firefox keeps getting updated at a regular pace for the Nokia N9, version 11 and 12 came in a quick succession and now Firefox 13 has arrived at the Nokia Store.

After a failed initial release and a few days missing from the Nokia Store, I was successfully able to install it and run the SunSpider javascript benchmark.

The test ran in approximately the same time as the previous version, which is also, more or less, the same time the PR1.2 stock browser took to complete it: definitely not a brilliant result although acceptable if we took the not so high-end SoC into consideration.

Samsung Galaxy Tab 10.1 SunSpider Browser Benchmark

June 15, 2012

Recently I’ve been using a Samsung Galaxy Tab 10.1, last year’s Samsung flagship tablet that runs Android HoneyComb (3.2).

This device was, probably, the main iPad 2 competitor, so I was curious about how it performed. I ran the SunSpider javascript benchmark on several different browsers I found on the Play Store.

One of the browsers I was most curious to try was the Google Chrome browser that recently got an Android version but unfortunately it only runs on the latest 4.0 ICS (Ice Cream Sandwich) version of Android so it was a no-go.

Nevertheless I tried several other browser as you can see below:

The stock Android HoneyComb web browser scored 2304.4 ms, so this can be seen as a base line for the other third-party browsers.

Dolphin browser (version 8.2.1) completed the benchmark test in 2439.7 ms, which is a little worst than the stock browser regarding javascript performance.

Mozilla’s Firefox, version 10.0.5, performed surprisingly good, running the test in 1883.4 ms, result that almost competes with the iPad 2 Safari result (that was around 1800 ms)!

The Firefox Beta, currently available at the Play store in version 14.0, performed worst than the stable v10 version, completing the benchmark in 1983.4 ms.

Opera Mobile, currently at version 12.0, ran it in 2094.7 ms which, compared to Firefox, is not brilliant although better than the Android 3.2 stock browser.

Finally, Skyfire 4.1.0 ran it in 2276.2 ms which is about the same as the stock browser, so, not a result to be particularly proud of.

Although packing a Tegra 2 SoC and more RAM, the Galaxy Tab 10.1 does not beat the theoretically less powerful iPad 2, that’s still the king of my SunSpider tests on tablets, with a result as low as 1744.0 ms achieved with iOS 5.0.1.

iOS 5.1 SunSpider benchmark

March 8, 2012

It’s been 24 hours since Apple released the iOS 5.1 update to several iDevices and I’ve ran on my iPad 2 the SunSpider javascript benchmark to check if there were any performance improvements.

As you can see in the screenshot above, there is absolutely no improvement… at all! The benchmark is completed in the same amount of time iOS 5.0 took.

The main noticeable improvement, at least for me in this first day of usage, is in the stability department: in iOS 5 and 5.0.1 it was common to watch Safari close by itself, especially when dealing with a large amount of tabs but with iOS 5.1 things appear to be much more robust.

Android lag

December 30, 2011

Recently I’ve been reading a lot of buzz on the web around the Android lag issue: some Google engineer started the controversy with a post on Google+ and then other devs wrote their own opinion on that.

First of all, I believe that if this subject was discussed widely on the web, it probably means that the problem exists! I want to emphasize this because first I was led to believe by some online forums that it was an issue exclusive to the Galaxy S I had. Apparently this might affect the whole platform and, as a consequence, more or less, every single Android device.

Second, the conclusion I get from all the posts I’ve read is that Android, unlike iOS or Windows Phone, does a lot of different stuff all at the same time instead of focusing on the screen response to the user when the screen is touched. It seems pretty obvious to me that the screen responsiveness is a major thing on a touchscreen device user experience. Google apparently doesn’t share entirely this view with their competitors and believe that, at some point in the future, the hardware will be so powerful that this problem will sort itself out.

In the meantime, I felt curious about everything I’ve read and made a short video where I try to see what’s the approach Google and Apple took on this by playing some CSS animation on both OS stock browsers.

By the way, on a side note, after watching this video I noticed something I didn’t before: recording the video at the same time, on the same place, connected to the exact same wireless network, the iPhone 4S shows a much higher level of signal reception compared to a very poor signal showed by the Sony Ericsson Live with Walkman.

Is the iPhone WiFi signal level too optimistic or is the Sony Ericsson antenna really bad?

Sony Ericsson Live with Walkman hands-on impressions

November 28, 2011

Recently I had the opportunity to try one of the newest and cheapest Android phones from Sony Ericsson, the Live with Walkman.

When I tried it for the first time I had no expectations whatsoever because it’s being sold, as a prepaid cell phone in some European carriers, at a really low price that beats most, if not all, of it’s competitors (mainly Chinese companies like ZTE and Huawei).

What took me by surprise is that the phone turns out to be awesome!

I was completely convinced by the good build quality, by the rather nice camera (with 5 MP and 720p video recording) and ultimately by the performance. This low-cost Android handset has exactly the same hardware (SoC, memory, etc.) the more expensive Xperia brothers (such as the Arc and the Neo) have inside but because this hardware (SnapDragon 1GHz CPU, Adreno 205 GPU, 512 MB RAM) only needs to run a smaller (3,2″) and lower resolution display (320×480) the phone actually is as fast (or even faster) when compared to the older brothers.

If you run the Quadrant benchmark, as you can see below, it is able to score between low seventeen hundreds up to 1869, which is a great result for a phone that’s competing near the €100 price point!

I’ve also ran the SunSpider javascript benchmark and the result left me very impressed… again: 3342.7 ms!

Of course perfection is a thing one can always aim at but never achieve and this phone has one thing that keeps it from being perfect: the screen! I’m not complaining about the low resolution (which brings some performance benefits) but the contrast and the color reproduction is not really good (there is no mobile Bravia Engine) and the phone does not adjust the screen brightness automatically, so here you have the phone’s Achilles’ heel!

My conclusion after playing with it a couple of days is that this phone is actually a great bargain, does very well what most Android phones do (regardless of price), and even if the screen isn’t the greatest one around it’s still worth every penny!

Thoughts on Android 2.3.5 for the Galaxy S

November 24, 2011

As I mentioned before, I’ve been using for some time now the Android 2.3.5 update for the Samsung Galaxy S and it deserves some comments.

Good news first:

– The previously reported Android OS battery drain bug is apparently gone! This just means the phone does not drain the battery all by itself in a dozen hours or so, which was a huge problem. The Android OS process is now at a more reasonable 6% share of the overall battery consumption.

– The browser scroll now works properly as in any other smartphone in the market

Unfortunately, there’s a few problems with this firmware:

– As I said in the previous blog post, the TouchWiz launcher crashes randomly.

– The phone process is also crashing a lot, especially soon after the end of a phone call, as you can see in the picture below.

– The browser may now present a correct scroll behavior but the stability is worst than before, which is not good because the browser was never the most stable app in Android.

It wouldn’t be so bad if it did compensate the lack of stability with improved performance, but running the SunSpider javascript benchmark returned a result of 6505.1 ms which is a regression compared against the 6305.4 ms scored by Android 2.3.3.

Apparently stock browser javascript performance optimization is not a priority for Samsung, neither is improving the stability of the official firmwares they release for the Galaxy S.