MacRumors

And today, Gizmodo drops the biggest bombshell: They got their hands on the device and have examined it and torn it apart for all to see.

What's new
- Front-facing video chat camera
- Improved regular back-camera (the lens is quite noticeably larger than the iPhone 3GS)
- Camera flash
- Micro-SIM instead of standard SIM (like the iPad)
- Improved display. It's unclear if it's the 960x460 display thrown around before -- it certainly looks like it, with the "Connect to iTunes" screen displaying much higher resolution than on a 3GS.
- What looks to be a secondary mic for noise cancellation, at the top, next to the headphone jack
- Split buttons for volume
- Power, mute, and volume buttons are all metallic

Other changes observed in the device compared to the current iPhone include a glass or shiny plastic back, 16% larger battery, and a slightly smaller but apparently higher resolution screen.

With the launch of new MacBook Pros last week, some observers immediately questioned Apple's decision to continue using Intel Core 2 Duo processors in its 13-inch models while bumping up to Core i5 and i7 chips in the larger form factors. At the time, Apple CEO Steve Jobs noted that the company had opted for "killer graphics" and the extended battery life versus what would have been a very small bump in CPU speed. Ars Technica now provides a thorough look at the factors that played into Apple's decision, showing how space constraints and other issues made the Core 2 Duo processors the only real option.

Much of the issue stems from a licensing dispute that has prevented NVIDIA from offering its integrated graphics packages alongside Intel's new family of Arrandale processors that includes the Core i5 and i7 chips used in the larger MacBook Pros, as well as the Core i3 chips that are a step below those adopted by Apple. The dispute left Apple with only two options if it wanted to use Arrandale processors in the 13-inch model, using Intel's "HD" integrated graphics or offering a separate discrete graphics chip from the likes of NVIDIA.

With Intel HD Graphics proving inferior for more demanding tasks to even the NVIDIA GeForce 9400M used in the previous-generation MacBook Pro, the former was not an option. And the latter meant that Apple would need to find room in an already-cramped 13-inch MacBook Pro body for a discrete graphics chip, a feat that was simply not possible, especially given the effect it would have had on battery life both from increased power consumption and from having to reduce the size of the battery to fit the expanded logic board.

Graphics performance would have taken a significant step backwards if Apple opted to rely solely on the Intel HD IGP that comes with Core i3 or i5, so Apple would have also had to include a discrete GPU like that used on the 15" and 17" models. Besides the added cost, there's simply no room on the 13" MacBook Pro logic board to include an additional discrete GPU.

Together, those factors mean that Core i5 and i7, and even the Core i3, were not feasible options for Apple in the 13-inch MacBook Pro, leaving Core 2 Duo as the fallback processor option where Apple could use NVIDIA's graphics technology. While Apple was able to use a slightly bumped CPU speed for the new models, it recognized that it needed to do more on the graphics front to improve performance, and so it engaged NVIDIA to create the GeForce 320M, a beefed-up, custom version of the company's GeForce 310M. The 320M offers tremendously superior performance over the Intel HD Graphics that would have been implemented with Arrandale processors in the 13-inch model had Apple gone that route, as well as offering more efficient power usage.

Clearly, Apple wasn't actively trying to cripple the 13" MacBook Pro by leaving it "stuck" with yesterday's Core 2 Duo technology. Instead, an Apple spokesperson told Ars that the company wanted to maintain the 13" MacBook Pro's unique combination of performance, portability, and extended battery life, as well as its $1,200 starting price. The combination of a faster Core 2 Duo processor and the NVIDIA 320M allowed Apple to do that.

The Intel-NVIDIA dispute that in large part forced Apple's hand with regard to the 13-inch MacBook Pro component choices has been cited as one of the major factors for Apple's rumored consideration of AMD processors over Intel's offerings for future Macs.

The May 7th date appears to be for new iPad 3G orders as those who have pre-ordered the 3G model were promised delivery by late April. Several readers who pre-ordered the 3G iPads note that their order statuses still indicate a "late April" delivery.

Update: Apple has also sent out emails to those who pre-ordered 3G models, reassuring them that their orders will ship in late April.

Thank you for your recent order of the magical and revolutionary iPad 3G.

We would like to confirm that your order will be shipped in late April as communicated at the time you placed your order. You will receive a confirmation notice when your order has shipped.

You can get up-to-date information about your order, including shipping status and tracking number, at http://www.apple.com/orderstatus

Seems more people are becoming convinced that the iPhone 4G images leaked by Engadget over the weekend are indeed legitimate. In fact, the report that the pictured product is actually a Japanese counterfeit has been retracted and was itself a false claim.

Daring Fireball's John Gruber now believes the images are real and that the next generation iPhone looks like the images depicted. Gruber also adds a detail that he has heard that the new iPhones will have a "fancy glass" back.

Multiple sources familiar with the next iPhone have confirmed to me that the back is made out of some sort of fancy glass -- and looks pretty much exactly what's pictured at Engadget. That's not the only reason I believe Engadget's unit is legit, but it's one.

He points to a 2006 Apple patent application which describes the use of Zirconia as a durable and radio-transparent material that might be used.

A portable computing device capable of wireless communications, the portable computing device comprising: an enclosure that surrounds and protects the internal operational components of the portable computing device, the enclosure including a structural wall formed from a ceramic material that permits wireless communications through.

Now, if you do believe that these images do represent the next generation iPhone, one interesting detail noted by an Engadget commenter is that it appears to allow for a user removable battery.

If you accept this as true, then it casts doubt on Applesfera's claims and photos that this same iPhone was identified as a Japanese counterfeit. (Update: Applesfera's claims have been retracted, so the pictured phone is not a known counterfeit device.)

More photos of the internals of the device have been posted to our forums. The original images came from Chinese forum Weiphone.

Apparently the phone was found on the floor of a San Jose bar inside of an iPhone 3G case. Right now we don't have a ton of info on the device in question, but we can tell you that it apparently has a front facing camera (!), 80GB of storage (weird, right?), and isn't booting at this point (though it was previously, and running an OS that was decidedly new).

Another image of the back of the device. It is clearly labeled as an iPhone though many of the details are simply X-ed out.

Princeton University has detailed wireless issues with Apple's iPad that have resulted in the university blocking several of the devices from its wireless network. While frequently invisible to end users and apparently unrelated to most early complaints from iPad owners about Wi-Fi connectivity issues, the iPad issue documented by Princeton can result in interference of network usage by other devices.

Princeton reports that they have seen a number of iPads with errors in their Dynamic Host Configuration Protocol (DHCP) protocol.

The malfunction we see is that the iPad uses DHCP to obtain a lease, renews the lease zero or more times (as expected), but then continues using the IP address without renewing the lease further. The iPad allows the DHCP lease to expire, but it continues using the IP address after allowing the lease to expire. The incident continues for some time (typically hours); usually it ends when the iPad asks for a new DHCP lease, or the iPad disconnects from the network.

Essentially, the iPad is continuing to use IP addresses that have already been reclaimed by the university's routers. Similar problems could explain certain issues that users have been seeing in their own personal use of the device, although user complaints regarding Wi-Fi performance have primarily focused on poor signal strength and inconsistent connections.

As of April 19th, 25 of the 41 iPads on the campus network had shown this issue while 9 have been blocked due to repeated problems. Princeton is reportedly working with Apple to address this issue, but has developed a tentative workaround to assist with the problems. Some of the nine blocked iPads have since been unblocked after the owners performed the suggested workaround. Princeton's current policy with respect to iPads is to allow them onto their network, and in the case of DHCP lease issues, notify the user and point them to the workaround. Only if problems continue after that point would a specific device be blocked.

Boy Genius Report claims that AT&T has instituted a block on all employee vacations for the month of June, signaling a possible iPhone release sometime during the month. In the past, such policies have been reliable indicators of iPhone launches.

Same story as last year and the year before, but we've confirmed with multiple AT&T sources that the carrier has now put a block on employees taking vacations in June. The only time AT&T does a straight block like this is for iPhone launches.

Apple has typically launched its new iPhone models in June of each year, although the iPhone 3G did not launch until July 11th, 2008, just inside the window predicted by AT&T's vacation block that year.

For the past two iPhone releases, Apple has introduced the new iPhone hardware at its annual Worldwide Developers Conference (WWDC) in San Francisco and launched sales of the new models 32 days (3G in 2008) and 11 days (3GS in 2009) later. Despite the fact that Apple has yet to announce dates for this year's WWDC, a number of hints and rumors have pointed to a late June timeframe for the event. Such scheduling would make for a short interval between the next-generation iPhone's presumed introduction at WWDC and its launch by the end of the month.

AppleInsider reports that Apple and AMD are in "advanced discussions" to bring the chipmaker's processors to upcoming Macs, potentially breaking the exclusive relationship Apple has held with Intel since moving away from PowerPC processors in 2006.

Representatives for the Sunnyvale, Calif.-based AMD have recently been seen on Apple's Commuter Coach buses, and executives for the chipmaker have been spotted on their way out of meetings with members of Apple's top brass, according to people familiar with the matter.

The meetings have reportedly included briefings by AMD that have since enabled Apple to begin working with AMD processors in its labs as part of an initiative to position the chips inside some of the company's forthcoming products.

According to the report, Apple is looking at using AMD's chips in its Mac lines, seeking to increase its flexibility for sourcing chips going forward. In particular, recent shortages of new notebook chips from Intel that reportedly led to a delay in the release of new MacBook Pros and licensing issues that have hindered graphics chipmaker NVIDIA's ability to work with Intel's processors have reportedly caused Apple to adopt a growing sense of dissatisfaction with Intel.

AMD's chip offerings are currently considered subpar when compared to those from Intel, but the report suggests that lower pricing and the ability to bundle AMD processors with ATI graphics chips (ATI is owned by AMD) as possible reasons for bringing a switch under consideration. In addition, Apple may have inside knowledge of forthcoming offerings from AMD that could make the chipmaker more competitive with Intel, or AMD might be more willing to offer custom solutions for Apple than Intel currently is. Finally, the report concedes that the talks could simply be a leveraging tool to assist Apple in its dealings with Intel.

Information Removed at Request of Apple Legal

Apple's 11-character serial number system has utilized an alphanumeric format of PPYWWSSSCCC, comprising two digits "PP" denoting a plant code where the product was produced, a single digit "Y" denoting the year of manufacturing, two digits "WW" signifying the week of the year in which the machine was produced, a three-digit unique identifier "SSS" to distinguish between machines that would otherwise have the same serial numbers, and a final three digits "CCC" specifying the model.

The new 12-digit serial number format makes a few minor tweaks to the format, which we have been able to decipher. The increase in length for the new format, which takes on the sequence PPPYWSSSCCCC, can be attributed to changes to the length of the "P", "W", and "C" portions of the code.

The addition of a third "P" digit to assist in identifying the manufacturing location is a fairly straightforward change, although it remains to be seen whether existing plant codes will simply have a character such as a "0" appended to them or if the system is being entirely reworked. The difference in the "W" component is more significant, however, including a shift to alphanumeric codes that will make it a bit more difficult to decipher a machine's manufacturing at a simple glance. The "W" change is also related to a difference in how the "Y" component is used.

Under the previous format, the "Y" component of the serial number was simply the last digit of the year of manufacture, meaning that a machine produced this year would carry a "0" in that position. The number in that position would obviously recycle every ten years, but it should be relatively clear from Apple's product release history whether a given machine was produced in 2000 or 2010.

While leaving the length the same at one character, Apple has modified the "Y" component to include a letter code instead of a number, and the new system will reflect in that code not only the year of manufacture, but also whether it was manufactured in the first or second half of the year. Apple has elected to use 20 different letters in this position, omitting vowels A, E, I, O, and U, as well as B. As with the previous system, letters in this position will recycle every ten years. For 2010, machines with a "C" in this position will have been manufactured in weeks 1-26, while machines with a "D" will have been manufactured in weeks 27-52 or 53. Next year will see the use of codes "F" and "G", and so on.

With the reduction of the "W" component of the serial number to a single alphanumeric digit, Apple has had to roll out a new system for identifying the week in which a given machine was produced. Previously, the two-digit code simply reflected the week of the year, beginning with "01" and moving on through "52" or "53".

The new format utilizes one of 27 alphanumeric characters to denote the week of manufacture, beginning with 1-9 and moving on to letters, omitting 0, vowels A, E, I, O, and U, as well as B, S, and Z. Because the 27 possible characters can not account for all of the weeks in the year, the "W" component must be paired with the "Y" component to determine whether the machine was manufactured in the first or second half of the year, with the "W" codes recycling every six months.

The three-character alphanumeric "S" code for unique product identification remains the same under the new system, while the alphanumeric "C" code for identifying model numbers has been expanded from three characters to four.

The ability to decipher Apple's serial number code to ascertain when a given machine was produced has been used by many customers wishing to learn more about the age of their machines, particularly when it comes to production issues. Understanding whether production date makes a difference in whether machines do or do not suffer from a given issue can help pinpoint when fixes might have been deployed and either alert or reassure customers attempting to determine whether their machines might be affected.