Hacker News Reader

Comment by Legion | Hacker News Reader

I don't think that's the case, but there definitely isn't a market for tablets that fail to compete on features or price.

The brilliance of what Apple has done with their mobile products is hard to overstate. They put all this effort into designing mobile chips and a mobile OS for the iPhone and iPod Touch. They took the fruits of those efforts and stuck them into a tablet format, and bam, you've got an iPad. They took them and stuck them into a set-top box, and bam, you've got the Apple TV 2G.

They get to spread the development costs of their silicon and OS over these four product lines. Talk about efficient reuse of your development investment. I own three separate devices powered by an A4 CPU and iOS operating system.

View on HN

Yet somehow the same strategy is not working that well for Android tablets and STBs.

You mean the one platform besides iOS that actually does have a foothold in the market?

Android is now 20% of the tablet market, while the BB Playbook is nowhere and the HP TouchPad can't even stay alive for two months.

I would like to see where that number is coming from. I've seen a lot of number's based on units shipped, but none on "actual" units sold. Curious what the 20% is actually based on.

Edit: Someone posted the link below.

Sorry, should have included it myself.

Main reason being they can't even compete on price without those fat carrier subsidies that it took to get Android for phones off the ground. They're pretty much at price parity with the market leader with a fraction of the useful apps, and surprised why they're not selling that well.

Anyone who thinks Android would be the leading smartphone OS if their devices were sold at price parity with iPhones is deluding themselves.

Similarly, if Apple wanted to overtake Windows PC markethsare with Macs (they don't) they would be foolish not to compete on price. When you are asking the user to try something new (or new to them) and non-mainstream, "pay more for this risky new thing" isn't a very attractive proposition.

It's even more efficient when you just relabel a standard chip you bought from Samsung and just tell everyone that you've put lots of effort into designing it. Works much better than I'd have imagined anyway.

Apple does more than "relabel" their chips. They bought Intrinsity specifically to create customized Arm chips for their products.

Intrinsity does not "create customized ARM chips", they optimize existing designs. There's been nothing impressive so far in Apple's chips, they're pretty standard design though thought to have been optimized by Intrinsity's folks.

Innovative custom designs were expected from the PA Semi acquisition. Apparently did not pan out, since it's been three years and there isn't much to show for it.

So optimizing an existing design just for Apple isn't customizing it? I disagree.

> So optimizing an existing design just for Apple isn't customizing it?

It's customizing the design, not creating a custom design.

Okay, I see what your saying. Thanks!

At least for the A4/Hummingbird, Intrinsity optimised the existing Cortex-A8 design for Samsung and before Apple bought them, so you disagree based on faulty facts.

SAMSUNG and Intrinsity Jointly Develop the World's Fastest ARM® Cortex™-A8 Processor Based Mobile Core

http://www.samsung.com/global/business/semiconductor/newsVie...

"Seoul, Korea, Austin, Texas - July 27, 2009 : Samsung and Intrinsity today jointly announced the industry's fastest mobile processor core implementation of the dual-issue ARM® Cortex™-A8 processor architecture in 45 nanometer (nm) Low Power (LP), low leakage process technology. This Cortex-A8 implementation, code-named Hummingbird, delivers 2000DMIPS at 1GHz. The Hummingbird comes with 32KB each of data and instruction caches, an L2 cache, the size of which can be customized, and an ARM® NEON™ multi-media extension. Performance and power consumption of the Hummingbird have been validated in silicon. Samsung is currently developing standard mobile SoC products using this new core."

At least for the A4/Hummingbird, Intrinsity optimised the existing Cortex-A8 design for Samsung and before Apple bought them, so you disagree based on faulty facts.

You realize this doesn't make sense...right?

Your going to have tell me what you think is wrong, because I can't read minds. I can't see any problem, but I'll rephrase to see if it's just the sentence structure that's confusing you:

Samsung worked with a company called Intrinsity. They took the Cortex-A8 design, and Intrinsity made it a bit more power efficient. Samsung sells these chips, and uses them in their devices. They also sell them to Apple. Apple gets Samsung to put Apple logos on them, and refers to them as the A4. Apple bought the company called Intrinsity after all the work on the A4 was finished, nothing they've done in the pay of Apple has any impact on the A4 or Hummingbird core.

I responded to someone claiming that Intrinsity customised the chips for Apple. This isn't true. I linked to a press release that shows this.

So what am I missing?

"There's been nothing impressive so far in Apple's chips... it's been three years and there isn't much to show for it."

Meh, ARM is a space where a number of companies are already in cutthroat competition and no one is going to walk in and totally change the game.

Apple's MO here is pretty clear:

(1) investigate their options thoroughly,

(2) back the best price/performance,

(3) possibly make some optimizations,

(4) then lock up the supply if possible.

...

And the results actually are pretty impressive, iPad 2 leaves every other tablet in the dust:

http://images.anandtech.com/graphs/graph4605/40363.png

While getting the best battery life in the category:

http://images.anandtech.com/graphs/graph4605/40257.png

The iPad 2 has a smaller, lower resolution display than the other tablets in those charts. Once you compensate for that it still has an advantage, but barely. It hardly leaves the other tablets "in the dust".

This is a common objection but if you actually do the calculation you find it's bunk.

44 fps / 1.30 = 33.8 fps, still ahead by anywhere from 45% to 200%.

You appear to be claiming that this is a demonstration of how much better Apple's custom ARM chips are, when it's just a better GPU in the core, just like Samsung's Hummingbirds in the Galaxy and Nexus S used the next GPU up compared with the iPhone 4. They're all PowerVRs. It's about as technically impressive as us both buying a laptop from Dell and me buying a laptop with a better GPU than yours.

I suppose you could give Apple 10% credit seeing how they bought that much of the company that makes them (again, after starting to use their products) but then Intel should get even more credit, because they've owned more, for longer.

There's two sides to this argument and you appear to have missed the important half. Anyone can slap a better GPU on and get better framerates. Not just anyone can slap on "the next GPU up" and still beat you in battery life.

Everything I've ever read about Apple's supposed optimizations is that they have to do with power efficiency.

(x) make it a branded, hard-to-copy ingredient of your product

Reply on news.ycombinator.com