The problem is simple, Apple can afford to pay for the plant, but once the plant has run its course and technology moves to the next NM phase, Apple is stuck with everything while TSMC can use it on older products. It is better for Apple not to get into this game and leave it to others more skilled at chip manufacturing.
Not that Apple can't, it wouldn't be able to. China controls A LOT of raw materials and have been stockpiling them for years. Apple might build it's own lab in US, but wouldn't be able to import the rare metals it needs. China would hike the prices so high it would not be profitable at all. That's why US needs to really rethink it's relationship with China.
I wonder if apple would ever consider going back to IBM. I know apple wasn't happy with IBM at the end of the PowerPC era, but those problems were not all IBM's fault. IBM could not make the needed investments to keep PowerPC competitive in large part because apple was such a tiny customer. But i wonder if IBM could do a much better job for apple now that apple is buying 100 times as many CPUs today as they were back then, is doing the chip design work itself, and could easily pay for an entire fab.
Why go with IBM? IT would be easier for apple to work with intel and get their processors up to apples specs for cell phones. I wouldnt be surprised if apple has demo units at their offices of iphones with intels new x86 cell phone processors .
Rejection might be a reason to buy TSMC outright. Wouldn't hurt apple to have its own fab. Also how about R&D in glass. If they make tv's and iphones they might try that too. Good for TSMC if it was better business for them not to or they know apple is seriously looking to move away from Samesung so maybe they are playing hardball. Whatever the scenario its good for apple to know money wont always be enough.
Why go with IBM? IT would be easier for apple to work with intel and get their processors up to apples specs for cell phones. I wouldnt be surprised if apple has demo units at their offices of iphones with intels new x86 cell phone processors .
Intel is the best manufacturer of microprocessors in the world -- no doubt about it. But intel is used to very fat profit margins AND selling the same product to multiple customers. I think those things make intel an unlikely partner for apple outside of macs.
IBM, on the other hand, is the brains behind most non-intel chip fabbers. Samsung, global foundries, etc all depend on IBM to do a lot of the heavy lifting in terms of R&D for new fab tech. So indirectly, apple already is using IBM. If the two companies could get over hard feelings from the PowerPC days, they could form a very powerful alliance.
Note, btw, that Tim cook worked at IBM for 12 years.
Samsung have just invested >4 billion to expand/renovate their Austin, Texas plant.
A new plant would be small change for apple, but probably not a wise investment. As margins are low.
Quote:
Originally Posted by Blastdoor
Intel makes 60% margins and even lowly AMD makes margins in the 40s. Both of those are a lot better than the minuscule interest apple earns on its cash hoarde.
The real reason apple would not want to build a fab is that it would just take too much time to hire the management and technical staff who could pull it off, then build it, then test it, and finally actually make something with it. We are talking about a 5 to 10 year effort. Apple may have more money than everyone else, but time is limited for all.
There's a lot more to the story and a bunch more that goes into the margin numbers. Cost of actual raw materials is the biggest, not to mention logistics. If Apple has their own plant they would have to compete against the rest of the 'big dogs' and probably end up paying a bunch more for materials due to economies of scale. As Apple continues to grow the return on such an investment is probably becoming attractive, but it's not there yet. My biggest guess is it's still much more beneficial to the bottom line to execute things as is.
We all know that Apple drive a hard bargain so dedicated lines would be consistent but simply wouldn't be as profitable as shorter runs for other companies.
Looking at the photo of the chips I can't help but wonder if Steve Jobs lost his obsession late in his life over the design of parts not seen. I see badly printed and meaningless numbers on several pieces and one piece that looks to be stuck in a bit crooked.
Is that what happens with patents--you get forced to use components provided by other companies and thus stuck with the lack of attention to visual detail? Or is it that Jobs and Apple eventually decided it was better to save money on components and concentrate on the visual appeal of the overall construction?
Comments
I don't mean going back to PowerPC. I mean having IBM fab A chips.
Huh???
The problem is simple, Apple can afford to pay for the plant, but once the plant has run its course and technology moves to the next NM phase, Apple is stuck with everything while TSMC can use it on older products. It is better for Apple not to get into this game and leave it to others more skilled at chip manufacturing.
Not that Apple can't, it wouldn't be able to. China controls A LOT of raw materials and have been stockpiling them for years. Apple might build it's own lab in US, but wouldn't be able to import the rare metals it needs. China would hike the prices so high it would not be profitable at all. That's why US needs to really rethink it's relationship with China.
Quote:
Originally Posted by Blastdoor
I wonder if apple would ever consider going back to IBM. I know apple wasn't happy with IBM at the end of the PowerPC era, but those problems were not all IBM's fault. IBM could not make the needed investments to keep PowerPC competitive in large part because apple was such a tiny customer. But i wonder if IBM could do a much better job for apple now that apple is buying 100 times as many CPUs today as they were back then, is doing the chip design work itself, and could easily pay for an entire fab.
Why go with IBM? IT would be easier for apple to work with intel and get their processors up to apples specs for cell phones. I wouldnt be surprised if apple has demo units at their offices of iphones with intels new x86 cell phone processors .
Rejection might be a reason to buy TSMC outright. Wouldn't hurt apple to have its own fab. Also how about R&D in glass. If they make tv's and iphones they might try that too. Good for TSMC if it was better business for them not to or they know apple is seriously looking to move away from Samesung so maybe they are playing hardball. Whatever the scenario its good for apple to know money wont always be enough.
Intel is the best manufacturer of microprocessors in the world -- no doubt about it. But intel is used to very fat profit margins AND selling the same product to multiple customers. I think those things make intel an unlikely partner for apple outside of macs.
IBM, on the other hand, is the brains behind most non-intel chip fabbers. Samsung, global foundries, etc all depend on IBM to do a lot of the heavy lifting in terms of R&D for new fab tech. So indirectly, apple already is using IBM. If the two companies could get over hard feelings from the PowerPC days, they could form a very powerful alliance.
Note, btw, that Tim cook worked at IBM for 12 years.
Quote:
Originally Posted by cycomiko
Samsung have just invested >4 billion to expand/renovate their Austin, Texas plant.
A new plant would be small change for apple, but probably not a wise investment. As margins are low.
Quote:
Originally Posted by Blastdoor
Intel makes 60% margins and even lowly AMD makes margins in the 40s. Both of those are a lot better than the minuscule interest apple earns on its cash hoarde.
The real reason apple would not want to build a fab is that it would just take too much time to hire the management and technical staff who could pull it off, then build it, then test it, and finally actually make something with it. We are talking about a 5 to 10 year effort. Apple may have more money than everyone else, but time is limited for all.
There's a lot more to the story and a bunch more that goes into the margin numbers. Cost of actual raw materials is the biggest, not to mention logistics. If Apple has their own plant they would have to compete against the rest of the 'big dogs' and probably end up paying a bunch more for materials due to economies of scale. As Apple continues to grow the return on such an investment is probably becoming attractive, but it's not there yet. My biggest guess is it's still much more beneficial to the bottom line to execute things as is.
No ones stuck with anyone. It is pie in the sky whether Apple has skilled up to emulate that Samsung design licensed from ARM.
We all know that Apple drive a hard bargain so dedicated lines would be consistent but simply wouldn't be as profitable as shorter runs for other companies.
Looking at the photo of the chips I can't help but wonder if Steve Jobs lost his obsession late in his life over the design of parts not seen. I see badly printed and meaningless numbers on several pieces and one piece that looks to be stuck in a bit crooked.
Is that what happens with patents--you get forced to use components provided by other companies and thus stuck with the lack of attention to visual detail? Or is it that Jobs and Apple eventually decided it was better to save money on components and concentrate on the visual appeal of the overall construction?
Quote:
Originally Posted by Shidell
And so Apple is stuck with Samsung.
rofl.
Oh snap! you are right! I forgot that Samsung and TSMC are the only ones they could possibly go to.
I am happy that you are happy though, its like seeing a baby laugh the the bubbles when he farts in the bathtub.
Quote:
Originally Posted by Galbi
You sound butt hurt. lol
Non sequitur. Try harder. lol
Quote:
Originally Posted by Blastdoor
I don't mean going back to PowerPC. I mean having IBM fab A chips.
Oh! Idk if IBM would fab non PowerPC chips, and if they did if they're lithography is what Apple currently uses.
Originally Posted by dorkus maximus
I see badly printed and meaningless numbers on several pieces and one piece that looks to be stuck in a bit crooked.
I'm gonna go ahead and guess that those numbers probably have meaning.