back to article ARM wrestling: Apple iPad chip to overpower rivals?

The big question is perhaps not whether Apple will announce the eagerly anticipated iPad next week, but what chip will power the media tablet. Since the device is expected to run a version of the iPhone operating system, and given that battery life is likely to be a key factor in the iPad's success, it's hard to imagine Apple …


This topic is closed for new posts.

I doubt it

This has never been Apple's forte. They may 'do design' (PPC) but it is usually WITH a big chip house. This would imply that the product will have a larger appeal than just Apple and then they can get economies of scale. With 5% of the market, I just do not see how they could produce the chipset cheaply enough to keep the costs down and profits up that they want.


Proprietary dreams

"If you're serious about putting good graphics into an ARM-based SoC, Imagination is one of the first companies you call for a chat - if, indeed, it's not at the top of your list."

Imagination being the outfit that Intel stupidly shopped at, presumably because of all their "expertise" in "mobile 3D", which left everyone running Ubuntu on their Dell (and other) gear completely stuffed because the drivers were all proprietary and Intel's chipset was using Imagination's super special secret sauce. Still, Apple fanboys (like the usual parade of breathless idiot analysts) won't care about that: they'll throw their tablet away in a year for the next model.


Poor little linux boy...

...didn't get his driver. Sniff.


Wrong target

My hardware has always had full driver support under Linux, unlike all those people who bought netbooks which won't be upgradeable beyond a few point releases of the kernel. Of course, to "empowered consumers" like yourself, this is a blessing because it gives you the excuse to go out and buy "the new shiny" and feel like you're on the red carpet until you get bored and want something even shinier.

But thanks for the playground-level idiot remarks, tiresome even at the primary school level, I'm sure they pass as the height of wit in your Royston Vasey-level locality.


This post has been deleted by its author


Why not PPC?

There really is no reason Apple would choose ARM over PPC if designing their own chip in-house...

OSX can already run and is well tested on both architectures, and Apple would not need to pay any licensing fees if using a PPC derived core.


RE: Why not PPC?

Joe wrote: There really is no reason Apple would choose ARM over PPC if designing their own chip in-house...

One reason could be power use. Apple stated that their main reason for going with x86 instead of PPC was that the x86 used considerably less power for the same performance. Designing a low-power PPC core would be a much larger undertaking than designing a SoC around an existing low-power core (such as ARM Cortex A9).

Another is software: iPhone applications run on ARM, so choosing another CPU family for the new product will require applications to be recompiled. That might be easy enough for those that Apple control, but all third-party applications would also require recompilation.

Conversely, I see no reason to choose PPC. Apple walked away from that one and has little or no incentive to go back. True, they bought a company that had plans of making a PPC core, but those plans never got very far, AFAIR.

Silver badge

Re: Why not PPC?

Power consumption. That's why not.

Jobs Horns

Except... (ppc vs arm)

That ARMs have already been proven in very low power consumption configurations. I am actually not aware of any specific PowerPC mark (sadly) that can compete with the most power-frugal ARM. Maybe they exist, but are keeping a low profile, for what reason I do not know.

Arms are (or were traditionally) much simpler processors, I do not believe they ever made a sophisticated behemoth the likes of a 970, nor did they have brethren like POWER...

I would much prefer a ppc, quite honestly, but an ARM would be just fine. Or if the MIPS-esque crowd come up with something good, sure, that too. Anything so long as it's not a frikin' atom / x86.

Actually, I'm probably not going to buy one of these lame DRM crippled Apple things (out of principle) so I don't quite actually give a crap, actually.

Fuck you, Apple, good luck with your iPlank... (/shows the finger)

Paris Hilton


OK, one reason the G5 consumed so much power is IBM was lazy and synthesized the thing. IBM had people who did low power (PPC 4xx guys) but who got sold off to AMCC. The people who put the G5 together went down the P4 path.1

IBM royally @#$@#$ up the PPC. PA did full custom for many parts of their PPC chip and had efficient clock gating, lowering the power of the chip significantly but by the time it hit the market it was a vanity company for Dan. That product (PA6T-1682M) shipped though, Torben.

That said, why reinvent the wheel? Doing another PPC at this stage is a science project. Why not just do a multi-core ARM chip with full-custom where it helps your MIPS/watt and integrate the graphics stuff they bought?

Paris 'cuz after all these years I'd still like to royally

. 3

Current ipods

Err... the current generation of ipods and the iphone use Imagination IP for their tile-based 3D rendering. There's not really a whole lot of competition for hardware 3D on handheld devices.

And wrt to Linux drivers, well the Pandora supposedly will run Debian won't it? Not to mention there are several Linux smart phones with PowerVR SoCs.


binary blobs

Pandora, BeagleBoard, n900 and so on are using a TI OMAP chip with the powerVR graphics core. They need a binary-only driver if you want the accelerated graphics to work, but so far TI seem reasonably serious about keeping it available and up to date, which is nice but sadly rather unusual.

More common is a mess like the intel driver previously mentioned, or broadcom's adsl driver, used in modems like the dg834gt, that only works if you're using an ancient (2.6.8?) kernel version. The intel case is particularly sad because the rest of their recent graphics support in linux is open.


Exactly what we can expect of Apple

@Schultz: Apple's computers may only account for 5% of systems sales, but their cashflow is actually more on par with Big MS themselves. As for per-unit profit - can you say "ecosystem"?

Jobs has seen to it that all his company's products have a variety of selling points. (I won't mention them here, fanbois are attracted to flames.) If they truly have managed to steal a march on everyone else in terms of processing power, battery life and chip size, then you can be damn sure that they'll make the utmost of the opportunity to dominate the new* tablet device market. It's a feasible set of moves.

*It did exist before rumours of Apple's entry, but (lets be honest) outside of a few niche applications no-one cared,


Agreed BUT

"@Schultz: Apple's computers may only account for 5% of systems sales, but their cashflow is actually more on par with Big MS themselves. As for per-unit profit - can you say "ecosystem"?"

You are correct BUT

Apple almost went belly up selling hardware that was mostly proprietary. (PPC) I am not saying it was good or bad hardware. Nor am I saying that Apple is a big or small company in terms of unit sales. It is just that I would doubt that SJ has forgotten that Apple almost died. This is why I am betting on a bit of hardware that will have enough sales to drop the per item cost. (Maybe not at first but soon enough.)


One thing is sure

Whatever Apple comes up with will *not* brag with "20 hours runtime in standby" like other tablets (which is a joke when you need to shut down the thing over night because otherwise the battery will be half empty in the morning). Booting sucks even with notebooks and a tablet should need to reboot only when you update the OS. Click on, use, click off. Instant-on is one thing that really needs to be there with these things and I really doubt Apple would bother with a tablet if they hadn't solved this. And they're really good at power management anyway.

I may sound like a fanboi but I really expect something really thin, really great and with a battery life that blows the other tablets out of the water. Otherwise it may be just a big iPhone with some twist to it, I don't care.


Please get a grip: ARM SoC's are not full custom hardware....

They are unique, yes, but you get a standard processor, a standard bus system (AMBA, the ARM on chip answer to PCIe), a recognised programming model with very good compilers, hardware debuggers, etc. You also get access to a huge library of AMBA modules rather like PCI peripherals from 3rd parties. It is just like assembling a custom PC but much harder.

You get TSMC, Charter, STMicro or one of the other foundries to make them for you cheaply. Lots of CMOS bulk manufacturers out there, only Intel and IBM have their own CMOS foundries now.

What you get is the huge power advantages of having it on one chip (power savings for no pad ring, no unnecessary extra hardware and no pointless charging of PCB traces). You do need to engineer it for low power but they have a team hugely experienced in successful designs - PA Semi.

You also get ultimate lockin - no 3rd party OSs and noone else can use your software.

Lots of programmers in the embedded space who know what an ARM looks like.

Google make their own ethernet switches and PCs for the same reason - done carefully and smart it can be of huge value..

Apples PPC processor was a fail because they relied on Motorola and IBM to design it. Motorola gave up the PPC and IBM doesn't do mobile - only big iron. Apple didn't want to do a desktop and laptop processor and chipset. It takes Intel teams of 100s to design them. ARM SoCs are much easier.


one point

Agreed but teams of 100s? No way. Teams of thousands.

Thumb Up

"You also get ultimate lockin - no 3rd party OSs"

"You also get ultimate lockin - no 3rd party OSs and noone else can use your software."

I like most of what you wrote but that particular statement puzzles me.

I have ARM-based PDAs that started life with WinCE/PocketPC that can easily run a 3rd party OS (e.g. Linux) and 3rd party applications.

I have ARM-based routers that started life with vendor-supplied OS that can easily run a third party OS (e.g. Linux-based DD-WRT, or OpenWRT, or in the right circumstances, VxWorks) and 3rd party applications.

I have ARM-based industrial comms gear that starts life with a vendor-supplied OS, that can alternatively run a selection of third party OSes, although 3rd party applications might not be so readily available yet.

I have an ARM-based mobile phone and I have no idea how to run, and no interest in running, a 3rd party OS on it. But that's the exception to the rule.

Where's the "ultimate lock-in", please?

This topic is closed for new posts.


Biting the hand that feeds IT © 1998–2017