Give us your best shot as to why this acquisition makes the most sense to Apple at this time.
I've already done that, but, sure, I'll do it again, and expand upon it.
I have to go back a bit in history to that gpu design firm I've mentioned that Apple bought several years ago.
There was much speculation in the industry, as well as here, that Apple would finally take the lagging gpu Mac development into their own hands, but it never happened.
What did happen is that Apple developed the core technologies instead. My thought on that was that Apple wanted the low level knowledge that these people had about gpu hardware and software. Remember that what Apple had done was never done before, and was, and is, significant. So much so, that MS is copying the entire concept.
I feel that some of that is in this deal as well. Apple wants their expertise remember, not their products. Will Apple really take the risk of the responsibility of an entire chip development platform? I think not. That would involve them developing their own compilers, etc, as well as hardware.
This is VERY expensive these days. Then Apple would have to out-fab this.
Look at the problems AMD is having. What would happen if Apple has problems with a processor line that they will need for their products? That's a tremendous risk. I don't see Apple taking it. With all the experience these people have, doing an ARM would be something they have never done. It doesn't matter that years ago, he was one of the main developers of the ARM, he hasn't done work with it for quite a while, and a lot of water has flowed under that bridge.
Since we have no idea what Apple has been doing with them the past three years, despite all the speculation running around, if anything, things we know nothing about could have been brewing.
Does anyone doubt that Apple has been running the iPhone/itouch development on both the ARM AND the x86 hardware? I don't. I think thats very likely. More than likely.
How soon could Apple have a new Apple-made ARM up and running? A year? Two years?
Do you think that somehow, this small firm has been secretly working on ARMs for a while, to please Apple? Maybe, but it's doubtful. This is a small company, and their PPC product has kept them busy.
I do agree that possibly they could help Apple with some specialized chips. But would that be a good enough reason to buy the entire firm? After all, Apple could contract out with any design firm to help design, and build custom chips that would belong to Apple.
So, one other reason could be to build some custom functions into Intel's chips. We don't know what experience people there have had with x86. It could be just as much as with PPC and ARM. Possibly more than with ARM.
Intel might be open to Apple doing this, if it would get them to use Intel, and move off ARM. After all, the speculation is that Apple may sell tens of millions of phones a year eventually, plus iTouches, and who knows what else. Intel would sure like to get that business!
And while some criticize the Atom as being too power hungry compared to the ARM, well, possibly now that's a problem, but not in a year. Intel has the best fabs, they are ahead of everyone in this area. AMD and IBM have announced their metal gate processes for later this year, or next year, but Intel has been using it for months already. They will be on 32 nm a year ahead of everyone else, and the smaller contract fabs, like Chartered, will take even longer to move to that.
I think that PA will possibly help in that area, as well as others, and Apple can leave it up to Intel to do the heavy lifting.
How soon could Apple have a new Apple-made ARM up and running? A year? Two years?
Do you think that somehow, this small firm has been secretly working on ARMs for a while, to please Apple? Maybe, but it's doubtful. This is a small company, and their PPC product has kept them busy.
I do agree that possibly they could help Apple with some specialized chips. But would that be a good enough reason to buy the entire firm? After all, Apple could contract out with any design firm to help design, and build custom chips that would belong to Apple.
Years ago PA Semi attempted to court an iPod contract showing Apple some low power embedded ARM designs. Apple liked what they saw and challenged PA Semi to design a highly energy efficient but much more powerful ARM design. What PA Semi designed impressed Apple so much they wanted to buy the IP and keep it for themselves.
This is a much more linear narrative than Apple buying PA Semi to design x86 processors
A year or two to make a processor assumes Apple and PA Semi have not already been working on it. When it was revelaed Apple bought Fingerworks, multitouch was already made. Which means Apple and Fingerworks had already been long been working together. When it was revealed that Apple bought Silicon Color, Apple Color was released a couple of months later which means Silicon Color and Apple had long been working together.
As for the expense. PA Semi is a much smaller company and used Texas Instruments as its fab and didn't seem to have insurmountable expense or problems. I'm sure their are many ways to structure a deal with a fab that can lower the cost. Such as starting with a generic ARM design and only adding some proprietary components to it. I doubt PA Semi was able to provide a contract for nearly 100 million units to a fab, the way Apple would for the iPod and iPhone.
Mel I think you just want to believe against all evidence to the contrary that Apple will use x86 in the iPod line. I guess you will hold on to this hope until Jobs appears live on stage and says "we are not using x86 in our iPod line".
Years ago PA Semi attempted to court an iPod contract showing Apple some low power embedded ARM designs. Apple liked what they saw and challenged PA Semi to design a highly energy efficient but much more powerful ARM design. What PA Semi designed impressed Apple so much they wanted to buy the IP and keep it for themselves.
This is a much more linear narrative than Apple buying PA Semi to design x86 processors
A year or two to make a processor assumes Apple and PA Semi have not already been working on it. When it was revelaed Apple bought Fingerworks, multitouch was already made. Which means Apple and Fingerworks had already been long been working together. When it was revealed that Apple bought Silicon Color, Apple Color was released a couple of months later which means Silicon Color and Apple had long been working together.
As for the expense. PA Semi is a much smaller company and used Texas Instruments as its fab and didn't seem to have insurmountable expense or problems. I'm sure their are many ways to structure a deal with a fab that can lower the cost. Such as starting with a generic ARM design and only adding some proprietary components to it. I doubt PA Semi was able to provide a contract for nearly 100 million units to a fab, the way Apple would for the iPod and iPhone.
Mel I think you just want to believe against all evidence to the contrary that Apple will use x86 in the iPod line. I guess you will hold on to this hope until Jobs appears live on stage and says "we are not using x86 in our iPod line".
PA never made an ARM chip. What they did was to present Apple with an outline for a new ARM chip. It wasn't a complete design. That's very expensive to do. Apple looked at that outline, said very nice, patted them on the back, and promptly went somewhere else for their chips. So much for PA' ARM designs.
The entire thrust of the company, announced even before the company was formed, was to produce better PPC chips, which they did.
There is NO evidence for what you and a few others are saying.
There is NO evidence that PA has done any work for Apple pertaining to current, or future products.
Fingerworks was formed to produce the very products they had in the works when Apple bought them. Totally different situation.
This one is much closer to the one I brought up.
Apple certainly won't give an order for anywhere near 100 million units. Possibly 25 million. You don't order numbers for several years of production at once. You take options on future production, and if you don't need the numbers, you pay a penalty.
You remind me of before Apple went to the x86. It was heresy to think of that. even afterwards, people were up in arms, and cursing it.
I don't see Apple ever again doing something with this much uncertainty. And I can't even see the point to it.
If Jobs says that they will be designing their own ARM chips, I'll agree that I was wrong. But, so far, I see no reason for him to do that.
Comments
Give us your best shot as to why this acquisition makes the most sense to Apple at this time.
I've already done that, but, sure, I'll do it again, and expand upon it.
I have to go back a bit in history to that gpu design firm I've mentioned that Apple bought several years ago.
There was much speculation in the industry, as well as here, that Apple would finally take the lagging gpu Mac development into their own hands, but it never happened.
What did happen is that Apple developed the core technologies instead. My thought on that was that Apple wanted the low level knowledge that these people had about gpu hardware and software. Remember that what Apple had done was never done before, and was, and is, significant. So much so, that MS is copying the entire concept.
I feel that some of that is in this deal as well. Apple wants their expertise remember, not their products. Will Apple really take the risk of the responsibility of an entire chip development platform? I think not. That would involve them developing their own compilers, etc, as well as hardware.
This is VERY expensive these days. Then Apple would have to out-fab this.
Look at the problems AMD is having. What would happen if Apple has problems with a processor line that they will need for their products? That's a tremendous risk. I don't see Apple taking it. With all the experience these people have, doing an ARM would be something they have never done. It doesn't matter that years ago, he was one of the main developers of the ARM, he hasn't done work with it for quite a while, and a lot of water has flowed under that bridge.
Since we have no idea what Apple has been doing with them the past three years, despite all the speculation running around, if anything, things we know nothing about could have been brewing.
Does anyone doubt that Apple has been running the iPhone/itouch development on both the ARM AND the x86 hardware? I don't. I think thats very likely. More than likely.
How soon could Apple have a new Apple-made ARM up and running? A year? Two years?
Do you think that somehow, this small firm has been secretly working on ARMs for a while, to please Apple? Maybe, but it's doubtful. This is a small company, and their PPC product has kept them busy.
I do agree that possibly they could help Apple with some specialized chips. But would that be a good enough reason to buy the entire firm? After all, Apple could contract out with any design firm to help design, and build custom chips that would belong to Apple.
So, one other reason could be to build some custom functions into Intel's chips. We don't know what experience people there have had with x86. It could be just as much as with PPC and ARM. Possibly more than with ARM.
Intel might be open to Apple doing this, if it would get them to use Intel, and move off ARM. After all, the speculation is that Apple may sell tens of millions of phones a year eventually, plus iTouches, and who knows what else. Intel would sure like to get that business!
And while some criticize the Atom as being too power hungry compared to the ARM, well, possibly now that's a problem, but not in a year. Intel has the best fabs, they are ahead of everyone in this area. AMD and IBM have announced their metal gate processes for later this year, or next year, but Intel has been using it for months already. They will be on 32 nm a year ahead of everyone else, and the smaller contract fabs, like Chartered, will take even longer to move to that.
I think that PA will possibly help in that area, as well as others, and Apple can leave it up to Intel to do the heavy lifting.
At least, that's how I see it.
I remember reading a lot of similar responses to the idea that Apple was going to switch to Intel.
I don't see Apple going back to PPC either but I can't find any other reason that explains this purchase.
Since we know that PA will discontinue the PPC's they are building, Apple won't be using them.
I'll make this clear: They aren't using the Intel Atom platform.
Being in contact with folks who know the fabs know they aren't using the Atom platform.
No one is using the Atom platform yet, what's your point?
How soon could Apple have a new Apple-made ARM up and running? A year? Two years?
Do you think that somehow, this small firm has been secretly working on ARMs for a while, to please Apple? Maybe, but it's doubtful. This is a small company, and their PPC product has kept them busy.
I do agree that possibly they could help Apple with some specialized chips. But would that be a good enough reason to buy the entire firm? After all, Apple could contract out with any design firm to help design, and build custom chips that would belong to Apple.
Years ago PA Semi attempted to court an iPod contract showing Apple some low power embedded ARM designs. Apple liked what they saw and challenged PA Semi to design a highly energy efficient but much more powerful ARM design. What PA Semi designed impressed Apple so much they wanted to buy the IP and keep it for themselves.
This is a much more linear narrative than Apple buying PA Semi to design x86 processors
A year or two to make a processor assumes Apple and PA Semi have not already been working on it. When it was revelaed Apple bought Fingerworks, multitouch was already made. Which means Apple and Fingerworks had already been long been working together. When it was revealed that Apple bought Silicon Color, Apple Color was released a couple of months later which means Silicon Color and Apple had long been working together.
As for the expense. PA Semi is a much smaller company and used Texas Instruments as its fab and didn't seem to have insurmountable expense or problems. I'm sure their are many ways to structure a deal with a fab that can lower the cost. Such as starting with a generic ARM design and only adding some proprietary components to it. I doubt PA Semi was able to provide a contract for nearly 100 million units to a fab, the way Apple would for the iPod and iPhone.
Mel I think you just want to believe against all evidence to the contrary that Apple will use x86 in the iPod line. I guess you will hold on to this hope until Jobs appears live on stage and says "we are not using x86 in our iPod line".
Years ago PA Semi attempted to court an iPod contract showing Apple some low power embedded ARM designs. Apple liked what they saw and challenged PA Semi to design a highly energy efficient but much more powerful ARM design. What PA Semi designed impressed Apple so much they wanted to buy the IP and keep it for themselves.
This is a much more linear narrative than Apple buying PA Semi to design x86 processors
A year or two to make a processor assumes Apple and PA Semi have not already been working on it. When it was revelaed Apple bought Fingerworks, multitouch was already made. Which means Apple and Fingerworks had already been long been working together. When it was revealed that Apple bought Silicon Color, Apple Color was released a couple of months later which means Silicon Color and Apple had long been working together.
As for the expense. PA Semi is a much smaller company and used Texas Instruments as its fab and didn't seem to have insurmountable expense or problems. I'm sure their are many ways to structure a deal with a fab that can lower the cost. Such as starting with a generic ARM design and only adding some proprietary components to it. I doubt PA Semi was able to provide a contract for nearly 100 million units to a fab, the way Apple would for the iPod and iPhone.
Mel I think you just want to believe against all evidence to the contrary that Apple will use x86 in the iPod line. I guess you will hold on to this hope until Jobs appears live on stage and says "we are not using x86 in our iPod line".
PA never made an ARM chip. What they did was to present Apple with an outline for a new ARM chip. It wasn't a complete design. That's very expensive to do. Apple looked at that outline, said very nice, patted them on the back, and promptly went somewhere else for their chips. So much for PA' ARM designs.
The entire thrust of the company, announced even before the company was formed, was to produce better PPC chips, which they did.
There is NO evidence for what you and a few others are saying.
There is NO evidence that PA has done any work for Apple pertaining to current, or future products.
Fingerworks was formed to produce the very products they had in the works when Apple bought them. Totally different situation.
This one is much closer to the one I brought up.
Apple certainly won't give an order for anywhere near 100 million units. Possibly 25 million. You don't order numbers for several years of production at once. You take options on future production, and if you don't need the numbers, you pay a penalty.
You remind me of before Apple went to the x86. It was heresy to think of that. even afterwards, people were up in arms, and cursing it.
I don't see Apple ever again doing something with this much uncertainty. And I can't even see the point to it.
If Jobs says that they will be designing their own ARM chips, I'll agree that I was wrong. But, so far, I see no reason for him to do that.