iPhone 3G S: First peek under the hood

Updated on June 19 at 12:15 p.m. PDT: adding Apple processor discussion in addition to earlier iFixit photo update.

What makes the iPhone 3G S tick, and why is it the "fastest iPhone ever"? Rapid Repair and iFixit tear down the iPhone in Paris and London, respectively, providing some clues.

First, a quick glance at Apple's claims for the iPhone 3G S. The ad copy on Apple's Web site claims that it launches applications faster and that Web pages "render in a fraction of the time." Apple doesn't stop there: "Improved performance and updated 3D graphics deliver an incredible gaming experience, too," according to Apple.

Apple iPhone 3G S disassembled

(Credit: Rapid Repair)

Rapid Repair said Thursday night that the 3G S uses a 600MHz Samsung processor based on the ARM Cortex A8 design. The iPhone also integrates a PowerVR SGX graphics chip from Imagination Technologies, according to Rapid Repair.

This, not surprisingly, beats the iPhone 3G (sans "S"), which uses an older ARM design--that runs at about 400MHz--and a prior-generation graphics chip.

The iFixit team also has done a teardown of the iPhone 3G S, with further analysis to follow.

Though the markings on the main Apple processor do not readily identify the chip, both Rapid Repair and iFixit say it is the Samsung S5PC100. Both sites confirmed this in response to e-mail queries.

The "Samsung ARM Cortex A8-based S5PC100 Mobile Application Processor...adopts a 32-bit ARM Cortex A8 RISC microprocessor and a 64/32-bit internal bus architecture, and operates up to 833MHz," according to documentation on Samsung's site.

The chip is also capable of high-definition (HD) 720p video, Samsung states.

iPhone 3G S main circuit board with Apple-branded processor

(Credit: iFixit)

CNET Reviews has validated Apple's speed claims, saying "for most native applications that don't depend on a cellular or Wi-Fi connection, the iPhone 3G S was consistently faster."

The review continues: "The iPhone 3G S opened the iPod player almost 5 seconds faster, and it was much quicker at loading some notoriously slow apps. For example, Bejewled 2, which can take up to 12 seconds to load on the iPhone 3G, started in just 5 seconds on the 3G S."

And: "The iPhone 3G S also started up much quicker than the iPhone 3G--we were up and running in 26 seconds instead of 50 seconds."

The Palm Pre also uses a Cortex A8 application processor from ARM and the PowerVR SGX graphics from Imagination.

[canberra_photographer]

Palm Pre looks a bit old there.

[mrbofus]

How does the Palm Pre look a bit old when it uses the same processor and graphics as the iPhone 3G S?

[David Dudley]

I own both the Pre and iPhone and wish that the iPhone could do multitasking. That is the biggest killer app on the Pre though the major, major downside is the battery life of the Pre. The iPhone 3G is generally more responsive across the board than the Pre except when it comes to anything network intensive. It's clear from owning both devices that AT&T's network is just as bad as TMobile's - no coverage, pathetic data transfer rate, whereas Sprint's network is much bigger and their data IO wipes the mat with AT&T's sometimes Edge sometimes 3G network.

[ikramerica--2008]

Data speed depends on your area. Spring has the clear edge in some regions, but ATT is wicked fast in others.

Mutlitasking is the reason the Pre has bad battery life. Apple made a choice about this, and publicly stated that choice. People scoffed, but now we see the proof.

There will be multi-tasking eventually, but Apple will have to work out a battery management scheme and a multi-tasking protocol for third-party apps so that battery life doesn't drop so low it becomes useless.

[mbenedict]

No, it does NOT look like multitasking has anything to do with the Pre's battery problems.

Seems like there are two different problems affecting SOME (but not all) Pre users: 1) email syncing gets caught in an infinite loop and keeps spinning until the battery is drained; 2) in a poor reception area, it seems the Pre tries to keep reconnecting and again drains the battery.

The good news for affected Pre owners is that both should be correctable with a firmware update. Other Pre owners can multitask all day and not encounter any of these issues, so the problem is not related to multitasking.

[monkeyfun14]

@ikramerica

Maybe Apple should let the users choose want they want and not what they think is best for the user...

Perhaps make it a option?

Every smartphone but the iPhone can multitask.

[seven7dust]

well the Iphone does multi-task
it just doesn't do it with third party apps
IMO the only real viable multi-tasking is provided by the palm pre thx to the cards UI
in other operating systems you have to close the app everytime you need to open a new one So it really isn't multi-tasking if you ask me ,sure they run in the background
but it's not a desktop type of experience !
plus when you consider the no. of applications available for the iphone
many of which are graphic intensive games such as myst etc.
then Multi-tasking can become a double edged sword


eitherway if you want multi-tasking dont buy the iPhone
it's just one of those negatives again , just like the Virtual keyboard etc.
no phone is perfect every phone has negetives iPhone or Pre
anyway lack of multi-tasking of third party apps
might affect a few people but most people can do without it ,
cause 90% of the time it isnt used !

[ralfthedog]

The iPhone multithreads quite well. They just don't multitask on the application level because it eats memory quite fast (Not a problem with junk applications like iFart, however any two complex games or heavy business applications would nuke a first generation iPhone or a 3G).

Having two different modes, one multitask one singletask would be a bad thing. This would increase complexity, and hurt performance. Applications would need to be rewritten to know when they are on top and when they don't have focus. The better solution would be to have much faster flash memory so a large 100MB application can quickly save state, exit then restore if it is suspended.

[vmlenigma]

ATT SUCKS

[lusciatti]

Really? No problems for me, good coverage and good speed too.

[kewell82]

I agree with............................vmlenigma.

[dlowe402]

They ALL suck in their own way!

[pithenumber]

@dlowe402
except AT&T seems to suck more than the other carriers

[ausernamenoonehaschosen]

This article should mention that what Engadget posted: that the hardware could be able to play full HD 720p video.

http://www.engadget.com/2009/06/19/iphone-3g-s-gets-the-quick-and-dirty-tear-apart-treatment-alrea/

[droe2990]

ikramerica, I like what you said about it being a option. They could set it up under settings on a per app basis. I my self could care less about multitasking, in fact some of these push notifications are getting on my nerve's. I feel like I have a nagging wife on my hip. As for AT&T I have 16 salespersons and 12 truck drivers and support people that I switch to the Iphone at the end of last year, and have never had a problem with AT&T (We are in PA, MD, DC, DE, NY, WV and VA, Plus I do a lot of traveling and never had 1 problem with dropped calls or data. In fact last fall I hiked the AT Trail starting in lower VA to the end in GA and would get service about 60% of the time, my friend had verizon and would only get service about 40%. I was with Verizon for 10 years and now AT&T because of the Iphone and they are about the same (they both have bad customer service)

[badmojo42]

Just curious, do you have documented proof of this?

[ckh1272]

Why? Do you have documented proof? After all, it is at least possible that AT&T coverage is actually decent in those areas. Heck, I hear talk about Verizon and Sprint having better coverage, yet while Verizon is good where I live (Arizona), it sucked overall in North Carolina (where I used to live) and most of the southeast Sprint, on the other hand, had good coverage in NC but it sucks most of Arizona (other then Phoenix). That's two examples. Should i provide documentation?

[badmojo42]

good god, chill out. I wasnt asking you ckh1272. I only asked because droe1990 put actual "statistics" (40% vs 60%) in the post and wanted to know.

[ckh1272]

@badmojo42-Someone sure peed in your corn flakes, didn't they?? Judging by some of your previous posts regarding anything Apple (or crApple as you like to refer to it), I think my point remains valid. The guy gave estimated numbers, so I don't get why you would ask him to provide documentation. The post never mention that the numbers were based on fact, thus there is no need for him/her to provide you with documentation.

[jjwansch]

Does this phone support Blue tooth stereo?