Slashdot Mirror
Nexus S Beats iPhone 4 In 'Real World' Web Browsing Tests
bongey writes "In a series of measured real-world web load tests, the Android-based Nexus S phone spanked the iPhone 4. The Android phone and iPhone 4 median load times were 2.144s and 3.254s respectively. The sample size was 45,000 page loads, across 1000 web sites. It also follows rumors that Apple is intentionally slowing down web apps to make their native apps more favorable."
Maybe they weren't holding the iPhone correctly.
They were using a custom app. Not the default browser. So what they are saying is that their app runs faster on the Nexus S. Not that the Nexus S is faster then the iPhone.
Page load speed, that's their metric? And 50% faster is spanked? We're talking about computers, not 100m dash times - I expect an order of magnitude difference. How is the actual browsing experience - how easy is it to read and navigate on a 4" device?
I will go so far as to quote from TFA:
"Users don’t always notice the speed gap because websites are sometimes tailored to mobile phones, Blaze said. The difference will become more obvious as users demand richer experiences and move to tablet computers with larger screens.
So the metric their using to judge the devices isn't very noticeable, and probably won't matter on a device this size ever. Great. Guess if you have to break out a ruler to feel good about yourself...
Is it just my observation, or are there way too many stupid people in the world?
Oops meant 1110. That's more serious. Considering it takes me an average of 7865.349 msec to plug in my charger, still a fair trade..
Give a man a fish and you have fed him for today. Teach a man to fish, and he'll say "WHERE'S MY FISH, YOU IDIOT?"
Someone pointed out already that the way they tested is with apps that use the browser engine available to apps. As the second link says in the main story (probably, I'm too lazy to RTFA, I read others already), the iOS browser engine doesn't use the Nitro javascript engine.
I found one link that discusses it, but I'm sure there are better ones:
http://www.informationweek.com/news/personal-tech/smart-phones/showArticle.jhtml?articleID=229301178
That's nice.
Now, how quickly does it play Netflix movies? What's it's Hulu Plus app like, does it work nicely?
You don't say.
Seriously, for shame. I really do want an Android phone. It just isn't as functional yet. Another year or two of maturity and I think I'll finally get to switch.
There's a spot in User Info for World of Warcraft account names? Really?
Since the beginning, the iPhone has had busted CSS support for position: fixed; elements, which is terribly unfortunate as it makes Game! difficult to play. How does the Nexus S fare?
Game! - Where the stick is mightier than the sword!
Isn't the iPhone's A4 CPU supposedly some hundred MHz slower than the the one in the Nexus S, giving it better battery life? I don't think this has anything to do with strangling web apps, just different design goals.
The iPhone 4 is 777 MHz while the Nexus S is 1 GHz. Both are based on the ARM Corext-A8 and both have 512 MB of RAM. Given the difference in CPU speed, the results of the page load tests don't seem far departed from what would be expected. While the Nexus S is still proportionally a little faster, it isn't so wildly so that it can't be attributed to some minor tweaks in the OS or browser software. Using the term "spanked" seems a bit sensationalist in this instance.
Why not also include WP7? Has it been written off before people even try it?
This has a lot to do with hardware acceleration in the GUI, which for the most part isn't there in any Android below 2.3. I bought my Droid 2 last september and noticed exactly what you mention. In 2.3, that's no longer true. It feels MUCH smoother. In fact, my wife went out a month ago and picked up a low end device (with 2.3) that has a much better response rate and feel despite having a processor only half as fast.
Really, this is pretty much a new low in comment-baiting for Slashdot.
This so-called "test" is so utterly and completely unscientific as to be not worth the service space it is stored on.
Period.
It's supposed to be NEWS for Nerds, and this hardly qualifies. And, not content to troll on its own, the summary has to link to ANOTHER Flamebait summary to "support" its "point".
Note to Slashdot: You can do better than this; so DO it already!
Touch choice
Pun intended? :-)
Creationist Textbook Stickers Declared Unconstitutional by CowboyNeal
People don't care about X
Eventually Apples popularity will start to fade and people WILL care.
1 second difference can add up to a lot of time if you read many web pages, or you are searching for something. Just do the math. Say 100 modest amount web pages a day , 365 days a year. So you have (100*365)/3600 = 10.13 extra hours spent a year staring at screen that is doing nothing. In both tests they used the embedded browser for both handsets respectively. From their testing suite I don't see how they could throw off the benchmark that much, 45,000 samples is a pretty significant sample size.
More on there testing methodology is here http://www.blaze.io/mobile/methodology/ .
Finally the second link is complaints from Apple iOS developers. iOS 4.3 browser cannot use the new Nitro javascript engine in full screen mode, html 5 caching is missing, and mode in which the page is drawn on the screen has changed such that it is slower than native apps. Bug or not, it currently slower and no one knows why except Apple.
Amen brother! What good is a smart phone if I can't use it to view porn videos while driving???
I've abandoned my search for truth; now I'm just looking for some useful delusions.
First, Apple isn't "intentionally slowing down web apps to make their native apps more favorable." They have added a new JS interpreter (actually a just-in-time JS compiler) to Safari, but not to the "normal" web views that other apps can embed. This means only Safari is faster now, others are as fast as before.
Second, this test is flawed since it does not use Safari. It uses a custom app which uses neither the new JS engine nor the better caching of Safari or asynchronous multithreading.
I'm not convinced by this. The gallery app and the app drawer are both hard-ware accelerated, and neither is as smooth as the iphone. I think it is more to do with the slowness of java/dalvik, and the garbage collector, which only recently became concurrent.
Although nowhere near 45,000 tests, Anandtech recently ran a preview of the iPad 2 and did some browsing benchmarks to test the CPU where they loaded the pages for the iPad 2, Xoom, and the original iPad. Obviously the two tablets are different animals than the two phones, but given they run essentially the same OS and have beefier CPUs, we should expect similar results.
However, the iPad 2 is clearly faster in 7 of the 8 tests and the same speed as the Xoom in the remaining 1. It's possible that the websites used aren't good predictors for general load time though. Given that the two both have dual-core ARM chips running at similar clock rates, we shouldn't be seeing those results, especially if the ones from this study are a valid indicator of performance. The only other conclusions that can be drawn are that performance regressed in Android 3.0 (Or at least Motorola's implementation of it.), the Tegra 2 is a pile of crap, or that Apple is now somehow capable of making a significantly better SoC than many established players.
I can't speculate regarding the first, but given that the Xoom has a similar SunSpider result and a better BrowserMark result than the iPad 2, it's unlikely that either of the other two conclusions are correct. Would like to have additional data before concluding one way or the other, but it does appear as though some things are not adding up.
From experience, iOS will skip frames or otherwise cut the eye candy animations if it needs to. They have never slowed down the system, as they only use otherwise idle power. It's much harder to see this happen on the newer phones, but I saw it plenty of times back on the iPhone 3G as the OS became more complex. Apple does put a lot of effort into this, complete with dedicated system frameworks to drive the eye candy. Apple's view on the eye candy is that it is there to bring attention to what is occurring when transitioning states, and helps provide that "smoother" feeling when using their devices. This applies to both the OS X and iOS sides.
It's also why input latency and audio latency is something Apple pays a lot of attention to, more so then most of the competition. When you touch an iOS device to scroll, it feels like your moving the content on the screen with your finger. Other systems tend to have a larger time difference between touch input and display action, making the experience feel a bit disconnected. While highly technical people might not mind the lack of smoothness, the general populous is going to go with what feels better. It's not a separation between "Apple Users" and the rest of us, it's a separation between those who know and care about the little minute details behind the scenes and those who don't.
Windows Phone 7 on the other hand did seem to put a priority on the animations, to the point they would slow down and frustrate me when trying to do basic things. IE took over two seconds to recover from an orientation switch, and every time it would always play the animation. Orientation corrections in other areas were much faster.
Blaze backed away from its conclusion in light of the new data. Chief Technology Officer Guy Podjarny told CNET in a statement:
This test leveraged the embedded browser which is the only available option for iPhone applications. Blaze was under the assumption that Apple would apply the same updates to their embedded browser as they would their regular browser. If this is not the case and according to Apple's response, it's certainly possible the embedded browser might produce different results. If Apple decides to apply their optimizations across their embedded browser as well, then we would be more than willing to create a new report with the new performance results.
Read more: http://news.cnet.com/8301-30685_3-20044325-264.html#ixzz1GtaYoees