Sidebar Sidebar
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

Performance of Web Apps Saved to Home Screen Hampered in iOS 4.3

I'm confused. So if I create a shortcut to a website and place it on my desktop it will run slower than if I open safari and go to the same website?

In thought the shortcut just launched safari and then loaded the webpage.
 
kepardue said:
Apple has always maintained that it supports two platforms: native apps and web apps. If intentional it gives Apple an unfair advantage to maintain it's 30% cut. Why go out of their way to cripple the developing web if they don't feel threatened by it?

All that said, I'm really thinking this is a bug. Doesn't seem like something they'd necessarily do.
Click to expand...

Yeah, I'm going with 'bug' too. People are quick to point out the 30% thing, but you know who's really good at web apps? Google. Guess who'd take full marketing advantage of an iPhone that doesn't do web-apps well.

If Apple really is doing this on purpose they'll trade 'getting more app store sales' for 'more people buying Android phones.' Which one do you think Apple cares more about? I'm betting on "selling more iPhones."

If anyone here believes that then they really should be thinking "bug" here along with us.

amberashby said:
I'm confused. So if I create a shortcut to a website and place it on my desktop it will run slower than if I open safari and go to the same website?

In thought the shortcut just launched safari and then loaded the webpage.
Click to expand...

The issue seems to be (I think) that Apple introduced improvements into Safari in 4.3 but those improvements don't load when you launch from the home screen.

That's the factual part that everyone agrees on. The controversy is over whether or not Apple did that on purpose or if it was a mistake. No one really knows that part.
 
Anyone know yet if this is true of UIWebViews as well? That would probably be an even bigger problem for most users.
 
IMHO, not taking advantage of Nitro isn't nearly as big of a deal as not being able to take advantage of offline cache. Here's a question for anyone having previously written these things, did offline cache work for homescreen launched apps before iOS 4.3?
 
WestonHarvey1 said:
So they haven't been crippled, they are just performing the same as they did in 4.2?
Click to expand...

As I understood it (but haven't actually tried), partially yes - the performance is much as it was in 4.2 rather than enhanced 4.3 speed and rendering quality. *BUT* 4.3 also stopped these web apps from running offline by not caching their files, which I'm assuming had previously worked? Like I say I've never tried it, but given the complaints am assuming that this is a change in functionality?

Steve.
 
Wirelessly posted (Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_2_1 like Mac OS X; en-us) AppleWebKit/533.17.9 (KHTML, like Gecko) Mobile/8C148a)

sj up to his old tricks! what a sneaky bastard... hehe.
 
jonessodarally said:
Does anyone have enough knowledge in the area to check whether or not this affects web apps that are encased in a Web View in a native app?
Click to expand...

Donz0r said:
It's pretty bold to say that this was intentional or to use the word ' hampered' - the web apps just haven't received the safari update- could be a simple oversight
Click to expand...

Tastic Bycrom said:
Really? Does anyone actually think this was on purpose to "favor App Store apps over web app experiences?" Like there some evil IOS developers laughing manically that they denied you performance... get real.

If it was even a known issue, it was likely not addressed due to time constraints or priorities
Click to expand...

It seems to me that the iOS development team @ Apple are either incompetent or jerks. Incompetent because if this was unintentional, then they obviously didn't spend enough quality time to test everything. Why not just use the same Javascript engine in the whole iOS?

On the other hand, if it's intentional, seems very profit-mongering. I understand Apple wanting to make money, but I feel there's more to life than just making profits. Plus, if they made the performance as good as possible, wouldn't make some people want to pay for more stuff?
 
ChrisA said:
If the problem persists for more than a few weeks in is intensional. This is one of those things where the fix is simple.
Click to expand...

How are you to judge whether the fix is simple or not? Do you have access to the codebase and therefore know exactly what the issue is and how long it'd take? No, you don't. There could easily be some underlying issue that needs to be fixed before Web Apps are updated with the latest WebKit. You simply don't know what the situation is.
 
Theory as to WHY this is happening

My guess is that apps that are installed as web-apps on the springboard using this :
Code: 
<meta name="apple-mobile-web-app-capable" content="yes" />

are being opened up using UIWebViews instead of loading the entire Safari app. Developers already know that UIWebViews do NOT benefit from the upgrades of Safari in iOS4.3.

When Apple decides to upgrade the UIWebView component (which is not a trivial task), all the webapps will gain in speed.

Apps on the desktop that are merely shortcuts should still benefit from full-speed Safari.
 
bartszyszka said:
This is ridiculous. As a user I see the home screen icons as shortcuts to sites I go to frequently, not as separate webapps. Why don't they just make them work as well as they do in Safari? Or just open directly into Safari?
Click to expand...

Did they change the behavior for how they open in 4.3? In 4.2 they just open in Safari right alongside every other page I have open, So I don't see how it would behave any different that opening a page from a bookmark. Seems like a bug to me.
 
Total BS...

I just did a test of my own multiple times via the browser and the web app and didn't get anything like they're claiming. My iPhone 4 funning 4.3 never gave back a result higher than 4020ms. They have to be doing something fishy to get results like that. Maybe running it once on wifi and once over the at&t network.
 
Daveoc64 said:
Some people have reported that is a problem in UIWebViews as well.
Click to expand...

Yes, UIWebViews demonstrate the same behavior. I wrote a quick native app to run sunspider and I am getting the results below.

Mobile Safari: http://dl.dropbox.com/u/196588/Photo%20Mar%2015%2C%2012%2047%2000%20PM.jpg

UIWebView: http://dl.dropbox.com/u/196588/Photo%20Mar%2015%2C%2012%2046%2051%20PM.jpg

My guess is that the chrome-less home screen app is really just a special native app with a fullscreen UIWebView, not Mobile Safari. This would explain the behavior matching what we see in 3rd-party apps using UIWebViews. If you have ever tried to use advanced html5/css3 features from Mobile Safari in UIWebViews before you probably already know that they do not perfectly match.

My best estimate:

- Mobile Safari uses its own rendering and javascript engine. This engine was bumped in 4.3.
- 3rd-party apps and the chrome-less browser you get when you save to homescreen use UIWebViews. This was not bumped in 4.3 and is the same engine as 4.2.
- This doesn't explain loss of html5 cache manifests in the chrome-less browser. Not sure about that.
- This doesn't mean we won't see improvements to UIWebView rendering and javascript performance in later OS versions.

Note: This is all just conjecture.
 
griz said:
Did they change the behavior for how they open in 4.3? In 4.2 they just open in Safari right alongside every other page I have open, So I don't see how it would behave any different that opening a page from a bookmark. Seems like a bug to me.
Click to expand...

Webapps and Bookmarks/shortcuts are not the same!
 
You must log in or register to reply here.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.
Top Bottom
Back