Hasta La Gadget!

Continue reading Apple releases iOS 4.3

Apple releases iOS 4.3 originally appeared on Engadget on Wed, 09 Mar 2011 13:07:00 EDT. Please see our terms for use of feeds.

Permalink   |  Apple  | Email this | Comments

If 32GB just isn’t enough to carry your entire collection of Police bootlegs and Greatest Hits compilations, prepare to look enviously upon this bastion of storage salvation. It’s a 64GB iPhone 4, a supposed prototype leaked by a Foxconn employee. M.I.C. Gadget scored a quick video of the phone, even prying off the back cover revealing internals that certainly look official. The back is curiously designated “XXGB,” which thankfully does not indicate a turn to Roman Numerals, while the model number and FCC ID are similarly X’d out. Other details, like a missing silver ring around the camera lens, indeed make this look like a prototype, but of course the question now is whether this plucky little handset will ever follow its dreams of becoming a production model before getting relegated to obsolescence.

Update: We went over to the Sin Tak shopping mall in Hong Kong to get up close and personal with this mysterious device. Shockingly, the folks wanted a hefty HK$13,500 (US$1,734) for their phone, but this alone doesn’t really prove whether it’s a legit Foxconn prototype. We also got some folks to run the numbers through Apple’s system, and while the serial number is invalid, the model number indicates that this device was manufactured in early 2010, which probably means at one point Apple intended to release a 64GB version of the iPhone 4, but then changed its mind. Ah well, let’s see what this summer will bring us.

Continue reading 64GB iPhone 4 prototype spotted in China? (Updated)

64GB iPhone 4 prototype spotted in China? (Updated) originally appeared on Engadget on Wed, 09 Mar 2011 09:33:00 EDT. Please see our terms for use of feeds.

Permalink   |  M.I.C. Gadget  | Email this | Comments

The iPhone's multitasking drawer contains a playback widget for controlling audio apps. Photo: Jon Snyder/Wired.com

The next operating system upgrade for the iPhone, iPod Touch and iPad is landing sometime this week, and customers can look forward to major improvements in Safari performance and AirPlay media streaming.

Wired.com reported on most of the changes in iOS 4.3 when the beta released January. The newest additions that Apple announced last week included a significant speed boost to the Safari web browser, using the new Nitro JavaScript engine, which should make JavaScript run twice as fast.

Additionally, the AirPlay streaming feature has been expanded so that you can stream video played inside Safari or a third-party app to an Apple TV. Previously, AirPlay video streaming was limited to the Videos app on the iPad.

Wired.com got an early look at the iOS 4.3 gold master (usually the official version that ships), and it’s pretty slick.

Benchmark tests Wired.com ran using SunSpider showed a significant speed boost in Safari’s JavaScript performance between iOS 4.2 and iOS 4.3.

An iPhone 4 running iOS 4.3 completed the test in 4,340 milliseconds, and an iPhone 4 running iOS 4.2 completed the same test in 10,626.4 milliseconds (fewer milliseconds is better). That makes iOS 4.3 about 59 percent faster than iOS 4.2 with JavaScript rendering. (See below the jump for the full benchmark results.)

So when iOS 4.3 ships this week, iPad 2 owners won’t be the only people getting a big speed boost. A significant increase in JavaScript performance means web browsing on JavaScript-heavy sites such as Google’s mobile apps (Voice, Docs and Reader), Facebook and others should be zippier.

As for AirPlay, I used an iPhone 4 to play a YouTube video and was able to hit the AirPlay button to send it to straight to the Apple TV 2. Being able to AirPlay YouTube is pretty nifty, because using the Apple TV’s remote control to browse YouTube is a huge pain in the rear; it’s much easier to search for YouTube clips using the iPhone. iOS 4.3’s expanded AirPlay support will be a nice addition for anyone with the new Apple TV.

The iOS 4.3 is due out sometime this week, possibly as soon as Wednesday, according to well-sourced Apple blogger Jim Dalrymple.

SunSpider results on iOS 4.2
============================================
RESULTS (means and 95% confidence intervals)
——————————————–
Total: 10626.4ms +/- 6.8%
——————————————–

3d: 1437.6ms +/- 6.8%
cube: 443.0ms +/- 14.8%
morph: 534.6ms +/- 5.3%
raytrace: 460.0ms +/- 2.0%

access: 1635.8ms +/- 33.6%
binary-trees: 146.2ms +/- 5.8%
fannkuch: 671.2ms +/- 0.6%
nbody: 429.6ms +/- 32.4%
nsieve: 388.8ms +/- 106.4%

bitops: 877.0ms +/- 0.5%
3bit-bits-in-byte: 180.8ms +/- 0.8%
bits-in-byte: 185.2ms +/- 2.7%
bitwise-and: 163.2ms +/- 0.6%
nsieve-bits: 347.8ms +/- 0.5%

controlflow: 143.0ms +/- 1.1%
recursive: 143.0ms +/- 1.1%

crypto: 643.6ms +/- 2.9%
aes: 295.0ms +/- 2.8%
md5: 163.6ms +/- 1.2%
sha1: 185.0ms +/- 11.1%

date: 744.8ms +/- 3.3%
format-tofte: 377.2ms +/- 3.5%
format-xparb: 367.6ms +/- 3.1%

math: 1134.6ms +/- 0.9%
cordic: 425.6ms +/- 1.7%
partial-sums: 372.0ms +/- 0.6%
spectral-norm: 337.0ms +/- 2.6%

regexp: 1444.8ms +/- 1.1%
dna: 1444.8ms +/- 1.1%

string: 2565.2ms +/- 21.6%
base64: 355.8ms +/- 4.3%
fasta: 422.0ms +/- 15.0%
tagcloud: 458.6ms +/- 18.8%
unpack-code: 744.8ms +/- 5.6%
validate-input: 584.0ms +/- 70.2%

====

SunSpider results on iOS 4.3

============================================
RESULTS (means and 95% confidence intervals)
——————————————–
Total: 4340.1ms +/- 0.7%
——————————————–

3d: 557.6ms +/- 1.6%
cube: 213.5ms +/- 3.4%
morph: 136.0ms +/- 0.9%
raytrace: 208.1ms +/- 0.7%

access: 497.2ms +/- 2.4%
binary-trees: 122.2ms +/- 3.3%
fannkuch: 138.7ms +/- 3.0%
nbody: 164.2ms +/- 4.7%
nsieve: 72.1ms +/- 2.1%

bitops: 219.2ms +/- 2.2%
3bit-bits-in-byte: 28.9ms +/- 10.4%
bits-in-byte: 53.6ms +/- 2.3%
bitwise-and: 64.5ms +/- 3.3%
nsieve-bits: 72.2ms +/- 2.3%

controlflow: 25.7ms +/- 4.4%
recursive: 25.7ms +/- 4.4%

crypto: 246.7ms +/- 0.5%
aes: 147.1ms +/- 0.4%
md5: 58.5ms +/- 1.7%
sha1: 41.1ms +/- 2.2%

date: 588.8ms +/- 4.2%
format-tofte: 277.3ms +/- 0.9%
format-xparb: 311.5ms +/- 7.2%

math: 462.2ms +/- 1.1%
cordic: 167.2ms +/- 0.8%
partial-sums: 206.6ms +/- 2.8%
spectral-norm: 88.4ms +/- 1.2%

regexp: 115.4ms +/- 3.5%
dna: 115.4ms +/- 3.5%

string: 1627.3ms +/- 0.9%
base64: 226.3ms +/- 3.2%
fasta: 227.8ms +/- 0.8%
tagcloud: 310.8ms +/- 0.8%
unpack-code: 554.2ms +/- 0.3%
validate-input: 308.2ms +/- 1.9%

Spring is near! The sun is shining, the weather is warming and holy what the mothercakes, gas is four dollars a gallon right now?! And what mister newscaster? It’s going to get more expensive? Geez! This is getting out of hand. We all would appreciate saving some money on gas. Here’s how. More »

The Verizon iPhone's network settings shows an option to enable a personal hotspot. Jon Snyder/Wired.com

For downloading data, the Verizon iPhone is slower than the AT&T iPhone, and it’s also slower than most other Verizon smartphones, according to a study published Monday.

Mobile analysis firm Metrico performed a comprehensive study on several AT&T and Verizon handsets, including the iPhone on both networks. The data suggests the differences between AT&T and Verizon are greater than we initially thought.

The debut of the iPhone on Verizon sheds light on more subtle differences between the Verizon and AT&T networks. Now that both networks have the same data-intensive iPhone capable of performing identical tests, it’s been easier for researchers to collect more comparable data. Independent tests by multiple publications, including Wired.com, have found that AT&T’s network is faster, while Verizon’s is more reliable with phone calls.

Metrico’s tests paint a conflicting picture. Some of their findings:

• The Verizon iPhone performed “below average” in data download speeds when compared to other Verizon phones including the HTC Incredible.
• For voice calls, the Verizon iPhone was one of the highest ranked in noise-canceling performance and was an average performer in voice quality.
• The AT&T iPhone was the top performer in data transfer rates, but it ranks lower in call performance and Bluetooth speech quality compared to some BlackBerry, HTC and Samsung phones.
• The AT&T iPhone completed 10 percent more download sessions than the Verizon iPhone when the handsets were moving — in a car, for instance.
• But the Verizon iPhone had a 10 percent better success rate when uploading data in a stationary environment, like sitting at your desk.

The rest of Metrico’s results tell the story we’ve already heard: The AT&T iPhone’s download speeds are about double the mean download speed of Verizon’s.

However, even though AT&T’s iPhone downloaded files faster, when it comes to loading web pages, the average load time is about the same on both devices, Metrico added.  Metrico explained that many different factors affect web performance, including network latency and the operating system’s rendering times.  (Indeed, a previous bandwidth test performed by Speedtest.net found that both the Verizon and AT&T iPhones turned in the same average latency times of about 350 milliseconds.)

Most peculiar to Wired.com is that the Verizon iPhone is “below average” in download speeds compared to other Verizon handsets. Was the handset purposely slowed down to avoid overloading Verizon’s network? Is there something about Apple’s hardware that makes it slower than other handset makers?

It’s tough to say: Metrico doesn’t have theories to explain the test results.

“I can’t say it’s related to technology at this point,” said Rich McNally, vice president of information product at Metrico. “We’re measuring what happened, not necessarily why.”

To perform the study, Metrico put each phone through live tests on the AT&T or Verizon networks, as well as Metrico’s own simulated network. On the simulated network, the phones completed tests and transferred data to Metrico’s servers for more consistent, controlled results.

The phones were also tested in both stationary and mobile environments. For mobile testing, the phones were tested inside automobiles while driving.

In all, the company performed more than 10,000 web page downloads, 2,000 data download/upload tests, and 4,000 voice calls.