or Connect
AppleInsider › Forums › Mobile › iPad › Apple to expand CPU design group beyond iPad A4
New Posts  All Forums:Forum Nav:

Apple to expand CPU design group beyond iPad A4 - Page 3

post #81 of 170
Quote:
Originally Posted by wizard69 View Post

I just see Apple patenting a lot of technology that frankly isn't of much use unless you are doing a lot of heavy design in and around the CPU core. Unfortunately I can't remember where I saw the patents. As to patentlyapple i've ben to the site a couple of times and frankly the search mechanism sucks.

I agree with the patently apple search -- I'll email Jack and see what he can do to add an advanced search -- simple && || ! and "" exact match would add a lot of utility.
"...The calm is on the water and part of us would linger by the shore, For ships are safe in harbor, but that's not what ships are for."
- Michael Lille -
Reply
"...The calm is on the water and part of us would linger by the shore, For ships are safe in harbor, but that's not what ships are for."
- Michael Lille -
Reply
post #82 of 170
Quote:
Originally Posted by wizard69 View Post

As noted I'm practically in the same boat thinking instead that the new hardware will go in a tablet. As a side note a processor optimized for the iPhone is also likely. The thing is iPhone could use a lower power, as in energy , processor than the iPad. IPhone performance isn't to bad for what it does but the IPad on the other hand pretty much sucks.


They have the potential to implement hardware that is markably faster than competing systems for the same amount of energy consumed. Especially if code is kept as native while the competition is running some sort of VM. In the end you are right the winner will be the company delivering the best performance at the lowest power point. There is little else to distinguish hardware.

------------------------------------------------------------------------

Why hasn't anybody mentioned the possibility that Apple will purchase a company that can implement much more efficient memory module solutions such as Hypercloud Technology by Netlist. These two companies have worked together in the past and it would completely destroy Google in the cloud while also giving Apple a much needed speed advantage and patent protection. Netlist is a 70 million dollar company. Isn't this Apple's MO?
post #83 of 170
Quote:
Originally Posted by pmz View Post

First of all, who cares,

"I don't care about it, so there's no reason for anyone to care."

Quote:
second of all, have you ever heard of boot camp? Sure you have, you just pretend it won't exist in the future.

You'll want to re-read my posts, or at least read all of them. A new architecture would make Boot Camp impossible. Virtualization is never out of the question.

Quote:
buy a $99 windows box

Can you BUY a computer for $99? Because if you can, I would and just make it a Hackintosh.

Originally Posted by asdasd

This is Appleinsider. It's all there for you but we can't do it for you.
Reply

Originally Posted by asdasd

This is Appleinsider. It's all there for you but we can't do it for you.
Reply
post #84 of 170
Where PoP = Package on Package

In theory the only way to get better RAM performance would be to have that RAM on the SoC itself. Even then you aren't likely to get better all around performance because of the different processing used to build hi density RAM.
Quote:
Originally Posted by roocka View Post

------------------------------------------------------------------------

Why hasn't anybody mentioned the possibility that Apple will purchase a company that can implement much more efficient memory module solutions such as Hypercloud Technology by Netlist. These two companies have worked together in the past and it would completely destroy Google in the cloud while also giving Apple a much needed speed advantage and patent protection. Netlist is a 70 million dollar company. Isn't this Apple's MO?

No. I'm not sure what you see at Netlist that has you so excited. All they have is RAM modules for servers. There really isn't anything "cloudy" about them.

Besides the last thing we want to have happen is for Apple to get to attached to the cloud and end up making their hardware useless for anybody but the brainless. People need to realize that cloud computing is a way for companies to keep their fingers in your wallet. The advantages to the user are few and for the most part rather specific.
post #85 of 170
Quote:
Originally Posted by wizard69 View Post

No I said the current iPad is slow and that future machines will make it look significantly worst. This becomes readily apparent when iPad attempts to do anything non trivial or for that matter do Javascript heavy web sites.

You really feel the over all performance of a machine should be measured by random javascript on random websites?


Quote:
Well maybe because for many users the processor performance is not the primary indicator of good performance. Look at it this way the iPad processor can be best seen as a 486 class performance.

You hit the nail on the head. Processor performance is not the primary indicator for good performance.

What exactly are you using as a benchmark to measure the iPad against? Slow in comparison to what?

Quote:
That is Apple has to significantly increase performance just to realize its goals. IPad 1 can barely deliver on current needs so it should be obvious that a much faster SoC is coming.

Apple has sold roughly 14 million iPads in 2010. You feel it can barely deliver on its current needs. What other evidence is there of this?
post #86 of 170
Quote:
Originally Posted by Tallest Skil View Post

Windows may not be made for installation on Macs, but it's possible.

And?

The reasons for running Windows on a Mac are rapidly diminishing...

Not that I expect Apple to abandon Intel for desktops or laptops any time soon - but it doesn't matter near as much as it did just a few years ago, and a few years from now it will matter even less.

This, more than anything probably scares the crap out of Intel and Microsoft both. The x86 architecture has a ton of baggage, and it's a miracle they have gotten it to perform as well as they have, but that baggage exists for Windows backwards compatibility. Mac OSX and now iOSX really don't need that baggage.

Longer term, it will be interesting to see how this all plays out...
post #87 of 170
Quote:
Originally Posted by Eriamjh View Post

One doesn't need to be an insider to know these things. They are obvious requirements to Apple's design cycle. I don't need to know any details about them. The chip will be faster, more powerful, etc.

Duh!

Yup - no kidding
post #88 of 170
Quote:
Originally Posted by Hiro View Post

samab does not recognize the concept of modification to a core. He strongly believes and stated in another thread that his absolute views of using unmodified IP or totally designing a core from scratch are the only two ways to use ARM IP, and he strenuously does not believe that Apple is an ARM architecture licensee so he doesn't think Apple can touch a single transistor inside the Cortex core.

No kidding. People like samab are still in denial that Apple knows what the heck they are doing.

Trust me, of all people Steve Jobs perfectly understands what happened with the original Mac and he has no intention of following the same path with the iOS.

Those that think that Apple doesn't have a long term plan for the iOS are sadly misguided.

2014 will be nothing like 1984 or 1994. The A4 was just a first step. That 4 year pipeline samab is so worried about? He needs to go back and re-read Steve Jobs comments about when they started working on the iPad that lead to the first iPhone and then the iPad... It's pretty obvious Apple has been working on these things for more than four years
post #89 of 170
Quote:
Originally Posted by wizard69 View Post

The whole reason Macs have been so successful as of late is that they have no restrictions on what can run on the machines. Why would Apple screw with that?

If even half of the Mac's out there have had Windows run on them (Virtual PC, Bootcamp or otherwise) I would be very astonished. While Intel was probably a significant influencer a few years back, I doubt it's much of an influencer for the majority of Mac purchasers.

Remember, the majority aren't geeks like us that are reading this board. There are far more of "them" than "us"

Quote:
Frankly if you have been keeping track of Apples CPU patents over the last couple of years you have to wonder what is taking them so long. It looks like their intentions are to supplement the ARM instruction set with instructions that accelerate the execution of Objective C.

It's probably because just because Apple can do a thing, it doesn't mean they will until it makes absolute sense. That precedent has been set pretty clearly.

For all intents and purposes, the iPhone pretty much came out of no where. Yeah, there were some vague rumors and rampant speculation, but nothing of significance until it was practically announced/released.

I have no doubt that whatever Apple is working on, it will be equally or more significant than the original iPhone. And it will arrive just as quick.

Quote:
In the mobile arena this is nothing to get negative about, in fact it should be just the opposite. A highly optimized high performance processor is exactly what is needed to really make mobile devices fly.

Yup, and I sincerely doubt we are the first to think such a thing is a good idea
post #90 of 170
Quote:
Originally Posted by wizard69 View Post

I think you missed something here, no one has indicated that they are getting rid of i86 hardware in Macs.

Not with the current or maybe even the next generation or two of chips.

But to think that Apple will stick with x86 chips forever is pretty silly. All the moves they are making with LLVM and other technologies, they are intentionally freeing themselves from CPU and architecture dependancy.

Who knows, this might be the opportunity for someone else to mature an architecture. Heck, given Steve Jobs and Larry Ellison's relationship, who knows. Perhaps the Mac's of the future could be running SPARC. I can think of worse things! It's certainly a cleaner architecture.

The point is, Apple isn't afraid to make bold moves if it has to. And that is what makes this very interesting to me. I can't wait to see what next rabbit comes out of the hat!
post #91 of 170
Quote:
Originally Posted by Tallest Skil View Post

Not yet, anyway.

I agree. Apple will stay wedded to x86 only as long as it continues to make sense.

The days of Apple being beholden to others are coming to an end. I wouldn't be surprised if some of that cash horde they are sitting on wasn't destined for lower level supply chain such as screens or even their own CPU foundry line.
post #92 of 170
Quote:
Originally Posted by wizard69 View Post

Seriously one big selling point for Macs is that the user can run just about anything in a VM.

For whom? A few geek hobbyists? I guarantee the majority of current Mac purchasers couldn't care less.

Quote:
Apple simply isn't big enough to do its own processor to match the i86 families. Not even close.

Seriously? Second largest company on the stock market sitting on a cash horde and amassing all kinds of low level design experience?

Underestimate Apple at your own risk. Just ask RIM and Microsoft...

Quote:
I see zero chance of this happening.

It doesn't take much imagination to have limited vision...

Quote:
It is one thing to run a program that uses ARM IP it is a totally different thing to build a processor that is functionally more impressive than the i86 hardware on the market.

There are other mature architectures out there that can be used. And we don't know how far Apple is with ARM. They were an original contributor in the '90s.

Quote:
I don't know about that, if I was AMD I'd be all over Apple and very willing to build whatever SoC they wanted. Imagine if one of AMD's Bobcat based Fusion products was tweaked for Apple, with Apple IP on board.

And who knows, they could be working with or perhaps even acquiring AMD for all we know.

Quote:
Well yeah in mobile devices, I suspect this is their goal. The problem is when you go beyond that, the issues are massively non trivial. More importantly the spark that got Macs to selling was i86 more than anything else.

Yup, but as I said, the need for that spark is long gone and I doubt the fact that Mac's come with Intel chips matters to the majority of current Mac buyers.
post #93 of 170
Quote:
Originally Posted by Smallwheels View Post

I wish Apple would actually manufacture things. They could start off by letting Americans assemble iPods and iPhones. Then they could have Americans manufacture the cases to their computers and idevices in the USA. In time Apple could manufacture their own circuit boards and chips in the USA. With the premium prices of Apple products the company could afford to do these things within the USA. Just the good public relations alone would get them more sales from Americans who want to support US manufacturers.

I was with you until you started spouting fantasy gobbledygook such as the above. It's a pipe dream -0 manufacturing is not coming back to the US any time soon. Not with our current economic situation. The "premium" you speak of is trivial compared to what it would take to accomplish you desires.

Quote:
If Foxcon employs one million people in China then Apple could eventually move that production to the USA and create at least that many jobs.

Not gonna happen. There are far to many socio-economic reasons why it isn't going to happen any time soon. This isn't an issue of Apple's doing, BTW. If you wan't to blame someone, you can blame our politicians and business leaders for selling our soul out over the past 20 years for short term economic gains.
post #94 of 170
Quote:
Originally Posted by Tallest Skil View Post

Here we go. What better way to completely shut the user out of the computer than making the whole thing proprietary? Mac OS will only work on Apple's architecture (read: Apple computers), and Windows, et. al. won't ever be installable because they'll have no need to build Apple versions.

It'd take forever, but it may happen.

Please, by the time Intel Apple products are phased out I'm sure there will be excellent Windows virtualisation/ emulators for ARM.

Edit: But I see this has been covered extensively already by other posters.
post #95 of 170
Quote:
Originally Posted by DocNo42 View Post

If even half of the Mac's out there have had Windows run on them (Virtual PC, Bootcamp or otherwise) I would be very astonished. While Intel was probably a significant influencer a few years back, I doubt it's much of an influencer for the majority of Mac purchasers.

I doubt that. Seriously it is a security feature in the sense that it gives people a sense of a fall back. On top of that there are many that simply need that compatibility for part of there computing needs. Further it isn't just Windows that people want to be able to run.
Quote:
Remember, the majority aren't geeks like us that are reading this board. There are far more of "them" than "us"

Well the geek mentality I understand. However I think you mis the practical mentality where people chose a Mac for its good points and then run the odd app they need to in a VM. Using a VM really isn't geeky anymore.
Quote:
It's probably because just because Apple can do a thing, it doesn't mean they will until it makes absolute sense. That precedent has been set pretty clearly.

Of course moving forward is always an issue of timing. However Apples continued research in this area tells me they have goal that sees some of this research going into products. Seriously why bother hiring engineers that do this level of work if you don't have intentions of putting at least some of that effort to work in product.

From what I can see A4 is NOT an example of implementing such hardware or engineering effort.
Quote:
For all intents and purposes, the iPhone pretty much came out of no where. Yeah, there were some vague rumors and rampant speculation, but nothing of significance until it was practically announced/released.

I find it hard to believe that you would put into print the above.
Quote:
I have no doubt that whatever Apple is working on, it will be equally or more significant than the original iPhone. And it will arrive just as quick.



Yup, and I sincerely doubt we are the first to think such a thing is a good idea
post #96 of 170
This is a truly excellent post. Remember, we do not know how this convergence of phone, tablet, laptop, desktop + other is going to play out.

Definitely there will be Intel Macs through 2015. But the dominance and importance of ARM Apple products leading up to 2015 can never be underestimated.

By mid-2012 we could easily imagine a MacBook Air 10" running ARM OSX. The thing is, by mid-2012 our understanding of "computing" will be changed and challenged further.

AMD is not a serious long-term option. Sure, Fusion has some potential but mainly in the transition to the first Mac ARM products.

Apple definitely is spreading its bets in any case, and this is a good thing. Their Intel relationship is definitely lukewarm at the moment, though of course as I mention Apple will need Intel for a few more years at least.

Quote:
Originally Posted by SockRolid View Post

Exactly what I'm thinking. $278 million for PA Semi plus $121 for Intrinsity is too much to spend on a slightly warmed-over single-CPU ARM design. The purchases prevent competitors from acquiring that IP and engineering talent, but that's only a small bonus.

I think Apple's long-term goal is to create their own proprietary SoC for not just their iDevices but for Macs as well. This would help lower their hardware costs since they won't be paying off-the-shelf prices for one of the most expensive components in their products. And lower hardware costs will help Apple maintain their margins.

But there are two more gigantic benefits. First, Apple could conceivably transition Mac OS back from Intel to their custom multi-core ARM. They have already transitioned Mac OS through several CPU changes: 68k to PowerPC, then PowerPC to Intel. Been there, done that, got the developers to come along too. And that could enormously benefit iDevices in the next decade. Eventually mobile device CPU power will exceed that of today's desktop computers, and Apple could prepare for that future by transitioning Mac OS to run on their ARM-based mobile CPUs.

Second, and this is perhaps the most important benefit for Apple, using a custom ARM chip on all their computing products would free them from dependence on an outside chip designers. For decades, from the 6502 to Intel Core i7, Apple has been at the mercy of the Motorolas, IBMs, and Intels of the world. Each of which have different goals than Apple. Motorola and IBM were more concerned with the embedded versions of their PowerPC chips than efficient and speedy desktop and laptop computer versions. Intel is more concerned with optimizing Windows performance than anything else. (And the CISC design of their CPUs uses vast areas of silicon for the execution of obscure backward-compatible x86 instructions generated only by Microsoft's compilers.)

None of those chip makers really wants to build a bespoke chip just for Apple. Intel, just after Apple completed the PowerPC-to-Intel transition, gave Apple their newest chips first. The original MacBook Air had an avant-garde chip that eventually was used in other laptops. But that was presumably because Apple paid them for that privilege, an unsustainable tactic, and the honeymoon ended.

Apple could eliminate their co-dependence on other chip designers now that they have acquired PA Semi's and Intrinsity's intellectual property. And that will set them up for their next decade or two of innovation. No other tech company in the world will be as well prepared.
post #97 of 170
Quote:
Originally Posted by DocNo42 View Post

I was with you until you started spouting fantasy gobbledygook such as the above. It's a pipe dream -0 manufacturing is not coming back to the US any time soon. Not with our current economic situation. The "premium" you speak of is trivial compared to what it would take to accomplish you desires.

I work in US based manufacturing and let me tell you it is tough. The biggest issue that you simply can't compete unless you are highly automated or are in some way protected. Even these things are of limited value these days, as many of the automation companies locally are now sending entire automation lines to China for various efforts. The low labor rate and lax regulation means that they often can do a completed item for the cost of the raw material for a case in the US. There is a rolling snowball effect due to everything being far cheaper to source in China.
Quote:
Not gonna happen. There are far to many socio-economic reasons why it isn't going to happen any time soon. This isn't an issue of Apple's doing, BTW. If you wan't to blame someone, you can blame our politicians and business leaders for selling our soul out over the past 20 years for short term economic gains.

Yep politicians of all strips and colors. Few if any have said boo about the out sourcing of manufacturing to China. It is rather sad because manufacturing is key to a countries strength.

About the only way things will change is if a major war flares up and the people of the USA give Washington no choice but to discontinue trade with China. Up to this point nothing has galvanized the American public to a suitable extent. Unfortunately war would in a literal sense whip out companies like Apple that have focused to much of their production effort on China. A conflict would mean Apples revenues would disappear over night with no easy fix to get back to a solid cash flow. It is really sad that the boards governing companies like Apple have never heard about keeping all your eggs in one basket.
post #98 of 170
Quote:
Originally Posted by DocNo42 View Post

For whom? A few geek hobbyists? I guarantee the majority of current Mac purchasers couldn't care less.

I doubt that.
Quote:
Seriously? Second largest company on the stock market sitting on a cash horde and amassing all kinds of low level design experience?

Seriously yes, very much so. Look at the trouble AMD has keeping up with Intel and they have had years of i86 experience. The difference between doing an ARM tweak and a full blown custom i86 processor is massive. As to emulation NO I don't think ARM hardware will be up to that anytime soon.
Quote:
Underestimate Apple at your own risk. Just ask RIM and Microsoft...

What about Apples customers? It would be foolish to underestimate them. Frankly any would barf at the thought of another architecture transition for the Mac.
Quote:
It doesn't take much imagination to have limited vision...

No; it takes imagination to see the trouble another processor transition will cause Apple. Beyond that Apple simply doesn't have the volume to support the engineering required to produce i86 class CPU's for the Mac market.

Everybody has been talking about how big Apple is but functionally they are a tiny company when it comes to CPU design.
Quote:
There are other mature architectures out there that can be used. And we don't know how far Apple is with ARM. They were an original contributor in the '90s.

Apple can do whatever they want with ARM for iOS devices, but Macs will need to remain i86 for the foreseeable future. More so they will not buy into yet another half successful architecture. If they did not learn from the PPC day we are all in trouble.
Quote:
And who knows, they could be working with or perhaps even acquiring AMD for all we know.

This is certainly possible and possibly a more rational approach to putting Apple IP into Macs.
Quote:
Yup, but as I said, the need for that spark is long gone and I doubt the fact that Mac's come with Intel chips matters to the majority of current Mac buyers.

It is a big deal and I think it is silly to dismiss out of hand. There are the issues of compatibility and performance and nothing Apple can produce with ARM will be able to match anything Intel has anytime soon.

It wold be great to see Apple produce a 64 bit processor that gives equal performance per watt to the latest Intel or AMD i86 but it is also wishful thinking. Even AMD is way ahead of the curve with its new Bobcat core.
post #99 of 170
Quote:
Originally Posted by nvidia2008 View Post

This is a truly excellent post. Remember, we do not know how this convergence of phone, tablet, laptop, desktop + other is going to play out.

While this is true, few of us can see into the future, we can look at past market behaviors. Iphone gave people a compelling reason to give up their old cell phones. I do not however see a compelling reason for one to give up an OS like Mac OS/X. Can Apple pull one out of the hat here, possibly but many uses for an OS like Mac OS/X will not be easily replaced by something drastically different.
Quote:
Definitely there will be Intel Macs through 2015. But the dominance and importance of ARM Apple products leading up to 2015 can never be underestimated.

In iOS based products this is very much the case. It is much harder to see where a transition to ARM in the desktop/laptop line up would make sense. These products literally have to develop in a different direction.
Quote:
By mid-2012 we could easily imagine a MacBook Air 10" running ARM OSX. The thing is, by mid-2012 our understanding of "computing" will be changed and challenged further.

Honestly I can't imagine a viable Mac of any sort running an ARM processor. Certainly not by 2012. As for "computing" as you call it, there is a huge difference between what that means to a consumer and what that means to somebody that uses a Mac Professionally. IPad has grabbed considerable consumer uptake because it satisfies certain needs really well. Even on the commercial side of things it has its good points but it is not a work station by any means. In fact it can't even pass as a computer for a clerk.
Quote:
AMD is not a serious long-term option. Sure, Fusion has some potential but mainly in the transition to the first Mac ARM products.

Apple would have a far better chance of putting their IP in a successful i86 chip if they work with somebody like AMD. AS for ARM show me a road map anywhere where it is indicated that core performance will come close to an AMD Bobcat based core anytime soon. Mind you this is a real 64 bit core.
Quote:
Apple definitely is spreading its bets in any case, and this is a good thing. Their Intel relationship is definitely lukewarm at the moment, though of course as I mention Apple will need Intel for a few more years at least.

Apples best near term weapon against Intels stupidity and belligerence, is AMD. All Apple really needs to do is to implement a couple of successful AMD based platforms to get Intel to pull its head out of its ass. I'm sure Intel is motivating Apple so that we can agree upon.

I just see the idea of extending ARM IP into anything that is suitable to run against current i86 hardware as being more than Apple can chew. People are expecting Apple to throw together a chip in a few years that has all the benefits of the i86 evolutionary process. This makes no sense even with the advanced help of ARM IP.
post #100 of 170
I'm sitting on route 4, in the middle of Florida, with a faster data connection than what I had in my hotel. (iPhone 3G) Traffic stopped due to a nasty accident, looks really ugly from about a mile away.

I only mention this because of the idea that 4G ought to be in Apples hardware plans. It just highlights that CPU engineering is only part of the equation. A faster SoC is great and I do hope it is the advancement we are hoping for, but a more balanced update would be welcomed.
post #101 of 170
Apple will not buy AMD. It's far simpler to design a spec that leverages AMD's CPU/GPU => APU solution than it is to take on billions in debt.

Intel and AMD will not be surpassed by ARM in CPU designs for Workstations, Desktops and high-end laptops.
post #102 of 170
I've been browsing Patently Apple and found this:

Quote:
An Apple A4 Chip Related Patent Surfaces

Today, in a newly published patent application from Apple Inc., we get to see a glimpse behind one of the many processes behind their new powerhouse A4 processor. Apple's patent reveals systems and methods for providing a system-on-a-substrate. In particular, this patent relates to systems and methods for reducing the total size of a system's circuitry by providing all of the components of the system on the same microchip. A microchip that the patent reveals is behind the iPad, iPhone and likely to be used in other future Apple products such as Apple TV.


Patent Background


Systems, such as systems for an electronic device, are often created from multiple components. For example, the components of a system could include one or more of a processor, memory (e.g., RAM, SDRAM, DDR RAM, or ROM), CODEC circuitry, Input/Output ("I/O") circuitry, communication circuitry, accelerometers, capacitors, inductors, or any other suitable components. Traditionally, each of these components is a distinct "entity" and could be created on a separate microchip or could be included in a separate package.

To create the circuitry for the entire system, the separate components (e.g., separate microchips) are typically coupled together through a printed circuit board ("PCB") or other suitable medium. The PCB could be fabricated with the appropriate wiring or routing to suitably connect all of the separate components.


Patent Summary


Apple's patent generally relates to systems and methods for providing a system-on-a-substrate (like an SOC). For example, rather than including the components of a system as discrete entities (e.g., as discrete microchips or as discrete parts), the components of a system could be formed together in "bare die" form. In other words, the components could be formed together on a single substrate, such as a silicon die or a die of other suitable material. In this manner, the components of an entire system could be densely and efficiently packed together, thus allowing the system to achieve a smaller size than a system using components that are discrete entities.

The components could include, for example, one or more of a processor, memory (e.g., RAM, SDRAM, DDR RAM, and ROM), CODEC circuitry, Input/Output ("I/O") circuitry, communication circuitry, accelerometers, capacitors, or any other suitable components.

In some embodiments, a die including the components of a system could be coupled to a substrate. The substrate, in turn, could be coupled to a flexible printed circuit board ("flex"). The substrate and the flex could include any suitable wiring and routing to electrically couple the die to other parts of the system such as, for example, a flash memory. In some embodiments, the flex can be coupled to a different surface of the substrate than the die. In some embodiments, the flex can be coupled to the same surface of the substrate as the die.

In some embodiments, the flex could include a ledge to which one or more components could be coupled. In some embodiments, a system could be created which does not include a substrate. In this case, all necessary wiring could be provided through the flex. In some embodiments, test points could be provided for a component of a die. For example, the test points could be included in a portion of the flex located substantially below the component to be tested.

In some embodiments, rather than being included together in a single die, the components of a system could be included as discrete entities. The discrete entities could be coupled to a substrate rather than being coupled to a printed circuit board ("PCB"). As a substrate can have more stringent design rules than a PCB, coupling the discrete entities to the substrate could allow for a system that is smaller and more compact in size. For example, the wiring for the system could be created using fewer layers and could be formed more densely in a substrate than in a PCB.

http://www.patentlyapple.com/patentl...aces.html#more


I assume that by intelligently packaging the components of a "computer" system that you can reduce size and distance -- and that smaller size/distance will normally yield better performance at lower power.


What I find interesting, and don't understand is:

"In some embodiments, a system could be created which does not include a substrate."

and

"The discrete entities could be coupled to a substrate rather than being coupled to a printed circuit board ("PCB")."

I assume that by intelligently packaging the components of a "computer" system that you can reduce size and distance -- and that smaller size/distance will normally yield better performance at lower power.
"...The calm is on the water and part of us would linger by the shore, For ships are safe in harbor, but that's not what ships are for."
- Michael Lille -
Reply
"...The calm is on the water and part of us would linger by the shore, For ships are safe in harbor, but that's not what ships are for."
- Michael Lille -
Reply
post #103 of 170
Quote:
Originally Posted by mdriftmeyer View Post

Apple will not buy AMD. It's far simpler to design a spec that leverages AMD's CPU/GPU => APU solution than it is to take on billions in debt.

Intel and AMD will not be surpassed by ARM in CPU designs for Workstations, Desktops and high-end laptops.

Any thoughts about ARM servers -- I understand the big attraction is the green factor -- low power, cooling, space, etc.
"...The calm is on the water and part of us would linger by the shore, For ships are safe in harbor, but that's not what ships are for."
- Michael Lille -
Reply
"...The calm is on the water and part of us would linger by the shore, For ships are safe in harbor, but that's not what ships are for."
- Michael Lille -
Reply
post #104 of 170
Quote:
Originally Posted by Dick Applebaum View Post

I've been browsing Patently Apple and found this:

http://www.patentlyapple.com/patentl...aces.html#more

Has anyone else even come close to matching Apples iPhone 4 PCB compression?

Dick Applebaum on whether the iPad is a personal computer: "BTW, I am posting this from my iPad pc while sitting on the throne... personal enough for you?"
Reply
Dick Applebaum on whether the iPad is a personal computer: "BTW, I am posting this from my iPad pc while sitting on the throne... personal enough for you?"
Reply
post #105 of 170
Quote:
Originally Posted by solipsism View Post

Has anyone else even come close to matching Apple’s iPhone 4 PCB compression?


That's pretty impressive -- nice find.


This is the closest I can find:

It's amazing how far we've come since 1976.

The picture shown below is about 1/4 the actual size.

The nickname for this card set was the "Rotisserie" as you could prepare a small chicken when running.

I suspect that gold plating on the 100 x 2 contacts would buy several iPads.

These are only the graphics boards -- other similar sized boards were used, each separately, for the: computer, RAM, I/O, etc.




Quote:
The Cromemco Dazzler S-100 Video Board

The Cromemco "Dazzler" dual S-100 boards were the boards that launched the Cromemco company. It went on to be the longest survivor of the major S100 companies building a series of outstanding commercial systems. The board set was made up of only 7400 TTL chips and allowed the display on a TV in color of graphic games and text. The resolution was primitive by today's standards. But we must remember that this was the era of "Pong".

http://www.s100computers.com/Popups//Dazzler.htm

Quote:
The Dazzler supported four graphics modes in total, selected by setting or clearing bits in the control register (0F) that controlled two orthogonal selections. The first selected the size of the frame buffer, either 512 bytes or 2 kB. The other selected normal or "X4" mode, the former using 4-bit nybbles packed 2 to a byte in the frame buffer to produce a 8-color image, or the later which was a higher resolution monochrome mode using 1-bits per pixel, 8 to a byte. Selecting the mode indirectly selected the resolution. In normal mode with a 512 byte buffer there would be 512 bytes × 2 pixels per byte = 1,024 pixels, arranged as a 32 by 32 pixel image. A 2 kB buffer produced a 64 by 64 pixel image, while the highest resolution used a 2 kB buffer in X4 mode to produce a 128 by 128 pixel image.[8] In normal mode the color was selected from a fixed 8-color palette with an additional bit selecting intensity, while in X4 mode the foreground color was selected by setting three bits in the control register to turn on red, green or blue (or combinations) while a separate bit controlled the intensity.

http://en.wikipedia.org/wiki/Cromemco_Dazzler

Edit: BTW, the colored, multi-wire flat cables were common in that era and were the inspiration for the Apple color logo -- originally used on the Apple ][.

"...The calm is on the water and part of us would linger by the shore, For ships are safe in harbor, but that's not what ships are for."
- Michael Lille -
Reply
"...The calm is on the water and part of us would linger by the shore, For ships are safe in harbor, but that's not what ships are for."
- Michael Lille -
Reply
post #106 of 170
Quote:
Originally Posted by wizard69 View Post

Well maybe because for many users the processor performance is not the primary indicator of good performance. Look at it this way the iPad processor can be best seen as a 486 class performance.

erm no. The i486 was a single issue in order design. In Intel incarnations it never hit more than 100MHz and SIMD was a spark on the horizon when it was designed. It has L1 on chip but not L2.

The Cortex A8 is a dual issue in order design capable of hitting ~1GHz with L1 and L2 cache on chip and has SIMD capabilities. If you insist to trying to compare total system performance with historical x86 processors you will have to go up into at least Pentium 3 territory.

Comparisons are false though because Apple's SoC is fabricated on a modern process with a modern memory subsystem. A PIII would never have been able to decode H.264 at HD resolutions, Apple's SoC can.
post #107 of 170
Quote:
Originally Posted by TenoBell View Post

You really feel the over all performance of a machine should be measured by random javascript on random websites?

Well actually yeah. I have no control over how well a particular web site is written so when visiting you need to rely upon fast hardware to drive the site.
Quote:

You hit the nail on the head. Processor performance is not the primary indicator for good performance.

Well not for many uses. However the A4 falls flat on its face when with with anything non trivial. for example any video that the hardware decoder can't handle directly. This is what really sucks about iPad right now, anytime something can't be handled by the special hardware units or the GPU comes up performance becomes glaringly bad.

This is for CPU bound apps but there is a similar issue with the lack of RAM that leads to poor or impossible performance.
Quote:

What exactly are you using as a benchmark to measure the iPad against? Slow in comparison to what?

Slow as to what I expect as a user. There is no need to compare it to anything, if the machine is not fluid or hesitates to much then it is slow. It is an issue of feel.
Quote:
Apple has sold roughly 14 million iPads in 2010. You feel it can barely deliver on its current needs. What other evidence is there of this?

There is lots of evidence some of it coming directly from Apple. For an Apple example iPhone 4 comes with 512MB of RAM.
post #108 of 170
Quote:
Originally Posted by chelgrian View Post

erm no. The i486 was a single issue in order design. In Intel incarnations it never hit more than 100MHz and SIMD was a spark on the horizon when it was designed. It has L1 on chip but not L2.

The Cortex A8 is a dual issue in order design capable of hitting ~1GHz with L1 and L2 cache on chip and has SIMD capabilities. If you insist to trying to compare total system performance with historical x86 processors you will have to go up into at least Pentium 3 territory.

Comparisons are false though because Apple's SoC is fabricated on a modern process with a modern memory subsystem. A PIII would never have been able to decode H.264 at HD resolutions, Apple's SoC can.

I will have to see if I can dig up some performance numbers. Somebody has run a limited set of benchmarks on the A4 someplace. In any event the CPU performance isn'[t as good as many seem to believe. Apps like VLC and other video players highlight this.
post #109 of 170
Quote:
Originally Posted by Dick Applebaum View Post

Any thoughts about ARM servers -- I understand the big attraction is the green factor -- low power, cooling, space, etc.

With Bulldozer hitting the streets and Sandy Bridge, as well, I can see the first people attracted to ARM server form factors being the Telcos for very specific needs, routers/switches, but not for Data Clusters. That will be where Bulldozer and Sandy Bridge own the space, not to mention the newer Oracle boxes and it's vertical selling.

ARM is targeting the ARM A-15 for Home and Web 2.0 servers, but nothing that would match the big folks.

One observation is for certain: LLVM trunk has ARM Cortex A-9 build support.
post #110 of 170
Quote:
Originally Posted by wizard69 View Post

I will have to see if I can dig up some performance numbers. Somebody has run a limited set of benchmarks on the A4 someplace. In any event the CPU performance isn'[t as good as many seem to believe. Apps like VLC and other video players highlight this.

Again this comparison is not good. The Apple H.264 decoder is highly optimised for the hardware. The codecs built into VLC mostly do not have hand crafted NEON optimisations and are using C fall-back routines running on the integer core.

Optimising software *first* before throwing hardware at the problem is always a power win. I don't doubt that future iPads will have faster processors in them however it will be driven by things that physically can't be done by the current processor, 1080p H.264 decode for example.
post #111 of 170
OT

I am a creature of habit.

I monitor several sites like AI for:

1) New Articles that might be of interest to me

2) New postings to discussion forums of articles That I have read and want to follow


So, I monitor about 20 sites for articles of interest -- I have a single web page with 20 tabs -- 1 tab for each main site that may contain articles of interest to me.

Periodically, i will reload all the tabs, then sequentially browse through them to see if any new articles have been posted to the main site.

If I find an article of interest, I open that article in a separate window.

Upon reading the article, I may decide to post a comment and/or review others' comments. Usually, I will just follow the "posts" link within the article window -- overwriting the article content with the posts content.


After a few hours of this, I have my master window with 20 tabs, and sometimes 15 or more "article/posts" windows of the stuff I am actively reading or posting.


To keep current, on all this, I must periodically:

1) refresh the main site page (reload all the tabs) and sequentially open each tab to see if there are any new articles of interest

2) sequentially reload each active "article/post" window to see if there are any new posts.


I waste a lot of time and browser resources doing this!


I am aware that RSS feeds can do some of this but you kinda' need to know in advance what you are looking for and you can be easily overloaded with info.


Some of the article sites where I post, will email notifications when anyone else posts to an article to which I have posted. This is better than nothing... But a bit kludgey.



What I think I want is an app/browser/service that allows me to:

1) register the main pages of sites like AI that may contain articles of interest

2) be notified automatically when any new article is posted to these sites.

3) when I find an article of interest, I want the option to register this article as an article of interest

4) be notified automatically when any new post is posted to these articles.


Ideally, this would be a single app/window on the desktop or iPad that aggregates updates to all the things (sites/articles/posts) that I am following.


I know enough about RSS, ScreenScraping and Push Notifications -- that this is not a major task!


Does anything like this exist?

.
"...The calm is on the water and part of us would linger by the shore, For ships are safe in harbor, but that's not what ships are for."
- Michael Lille -
Reply
"...The calm is on the water and part of us would linger by the shore, For ships are safe in harbor, but that's not what ships are for."
- Michael Lille -
Reply
post #112 of 170
Quote:
Originally Posted by Dick Applebaum View Post

Post just above

Simultaneously seconded, thirded, and fourthed. I've never understood RSS and I'd love something just like this.

Originally Posted by asdasd

This is Appleinsider. It's all there for you but we can't do it for you.
Reply

Originally Posted by asdasd

This is Appleinsider. It's all there for you but we can't do it for you.
Reply
post #113 of 170
Quote:
Originally Posted by wizard69 View Post

Well actually yeah. I have no control over how well a particular web site is written so when visiting you need to rely upon fast hardware to drive the site.

Can you link to anyone else of significance making this complaint about the iPad?

Quote:
Well not for many uses. However the A4 falls flat on its face when with with anything non trivial. for example any video that the hardware decoder can't handle directly. This is what really sucks about iPad right now, anytime something can't be handled by the special hardware units or the GPU comes up performance becomes glaringly bad.

The only video codec the iPad can play is H.264. At most 720P, 30FPS, main profile. The A4 handles that with no problem.

It cannot play any other video codec at all. I'm confused as to what are you talking about?

Quote:
Slow as to what I expect as a user. There is no need to compare it to anything, if the machine is not fluid or hesitates to much then it is slow. It is an issue of feel.

Can you link to anyone else of significance making this complaint about the iPad?

Quote:
There is lots of evidence some of it coming directly from Apple. For an Apple example iPhone 4 comes with 512MB of RAM.

So now you are making the argument that Apple should not increase the RAM in newer devices to prove their was enough RAM in the previous device?
post #114 of 170
That's essentially what I use twitter for. Every website and blog I read has a twitter feed. I scan the articles for what interests me and open the link to anything that jumps out.

Quote:
Originally Posted by Dick Applebaum View Post

Does anything like this exist?
post #115 of 170
Quote:
Originally Posted by TenoBell View Post

That's essentially what I use twitter for. Every website and blog I read has a twitter feed. I scan the articles for what interests me and open the link to anything that jumps out.

Is there a twitter feed for AI?

Is there a twitter feed for individual posts in AI?

Do you get notifications when new articles are posted to Ai?

Do you get notifications when a post is made to an AI article you are following?

If so, do you have a link that shows how to set this up?


I have a twitter account -- but only use it to follow a few people.

I get an email when a person I follow posts something... Email notification is not ideal, but better than nothing!


TIA

Dick
"...The calm is on the water and part of us would linger by the shore, For ships are safe in harbor, but that's not what ships are for."
- Michael Lille -
Reply
"...The calm is on the water and part of us would linger by the shore, For ships are safe in harbor, but that's not what ships are for."
- Michael Lille -
Reply
post #116 of 170
Quote:
Originally Posted by Dick Applebaum View Post

Is there a twitter feed for AI?

Is there a twitter feed for individual posts in AI?

Do you get notifications when new articles are posted to Ai?

Do you get notifications when a post is made to an AI article you are following?

If so, do you have a link that shows how to set this up?


I have a twitter account -- but only use it to follow a few people.

I get an email when a person I follow posts something... Email notification is not ideal, but better than nothing!


TIA

Dick

Yes, under the Twitter symbol next to the RSS symbol on the main page.

No, it has no association with the forum.

Yes, new articles are added to Twitter.

No, that is what the vBulletin is for.
Dick Applebaum on whether the iPad is a personal computer: "BTW, I am posting this from my iPad pc while sitting on the throne... personal enough for you?"
Reply
Dick Applebaum on whether the iPad is a personal computer: "BTW, I am posting this from my iPad pc while sitting on the throne... personal enough for you?"
Reply
post #117 of 170
Quote:
Originally Posted by solipsism View Post

Yes, under the Twitter symbol next to the RSS symbol on the main page.

No, it has no association with the forum.

Yes, new articles are added to Twitter.

No, that is what the vBulletin is for.

Thx.

I guess I have some work to do -- experimenting with twitter and vbulletin?

I was considering writing a Mac or web app to handle RSS feeds and screen scraping (if necessary) for the artile posts -- then post notifications to an app on the iPad and Mac. I have bits of existing code that do these things... Not a great challenge.

But, if it's readily available -- no need to reinvent the wheel.
"...The calm is on the water and part of us would linger by the shore, For ships are safe in harbor, but that's not what ships are for."
- Michael Lille -
Reply
"...The calm is on the water and part of us would linger by the shore, For ships are safe in harbor, but that's not what ships are for."
- Michael Lille -
Reply
post #118 of 170
I'm actually shocked at the number of patents apple has under its belt. I did not find what I was looking for tonight but did discover a large number of flash related patents and a few MMU and DMA controller patents.

Fair warning though, the search system is anything but user friendly!!!!

Dave
post #119 of 170
Quote:
Originally Posted by Dick Applebaum View Post

AFAICT, Both Apple and Samsung are licensed to Design and Manufacture (two separate licenses) ARM CPUs.

If Apple has Samsung manufacture the A4 (and follow-on) chips) -- does that mean that Samsung could use the same design in their competitive CPUs?

I don't think so, I believe they have different flavors of license. Samsung with a license to use the CoreAxx line in a SoC and Apple with a full license to the ARM IP, allowing mucking about with the core as well. Neither Apple nor ARM has said so explicitly, but there is a fair bit of circumstantial and "talking around the concept" evidence.
.
Reply
.
Reply
post #120 of 170
Quote:
Originally Posted by wizard69 View Post

I'm actually shocked at the number of patents apple has under its belt. I did not find what I was looking for tonight but did discover a large number of flash related patents and a few MMU and DMA controller patents.

Fair warning though, the search system is anything but user friendly!!!!

Dave

http://www.latestpatents.com
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: iPad
AppleInsider › Forums › Mobile › iPad › Apple to expand CPU design group beyond iPad A4