Sorry for my bad English capabilities but do you agree with me or not? No product released since the switch sofar is a real "child" of the switch. Everything could have been realeased in PowerPC version (maybe a little latter ). Jobs seems to be architecture agnostic. If the PPC world would offer something better for less, they are in again. Jobs didn't see this offer hence the switch.
Sorry for my bad English capabilities but do you agree with me or not? No product released since the switch sofar is a real "child" of the switch. Everything could have been realeased in PowerPC version (maybe a little latter ). Jobs seems to be architecture agnostic. If the PPC world would offer something better for less, they are in again. Jobs didn't see this offer hence the switch.
Apple is a public corporation with a fiduciary responsibility to return a profit to its shareholders. But forced to make a choice, I would have to disagree with you. To be sure, Intel's standard support chips will help Apple cut costs in its production of servers, laptops, and servers. But cutting costs is not the best way to earn a profit. The best way is to create new products which create new markets. Apple is product-driven. The switch to Intel enabled products which were not possible with the PowerPC. For example, it is not possible to design a laptop based on the G5 (PPC 970). The very thought of a G5-based iPhone may cause the sick and elderly to die laughing.
There are many who believe that the switch to Intel is irrevocable. Hence, they believe, Apple will drop support for everything else and concentrate solely on the x86. Well, Apple is using processors other than the x86 in the iPhone and Apple TV. By keeping MacOS X processor-agnostic, Apple will be able to develop other new and innovative products based on the OS and whichever processor is most appropriate.
Neither performance nor watt where the reason for the switch. It was money, that's all. Through the switch Apple doesn't have to design their own chips, motherboards ,etc. And they don't have to pay for the cpu development. Intel does all that for Apple and they probably get a special price too. IBM can build more or less any CPU (see POWER6) you want, but you have to pay for that (like Sony, MS and Nintendo did). Apple didn't want to. As both didn't really needed each other, they seperated. Same with Freescale. Apple would use McDonalds CPUs if they thought they could make more money with them.
There were several factors. It wasn't just any one.
Performance was one.
energy efficiency was two.
New chip feature sets was three.
Manufacturer reliabilty was four.
The advantage of Intel's vast board design and manufacturering abilities was five.
x86 developers was six.
All of the above would lead Apple to build products at prices they felt they couldn't build before, and along the way, attract more developers who otherwise didn't want to bring products to a new chip family, using tools they weren't familiar with, as WELL as a new OS.
There were several factors. It wasn't just any one.
Performance was one.
energy efficiency was two.
New chip feature sets was three.
Manufacturer reliabilty was four.
The advantage of Intel's vast board design and manufacturering abilities was five.
x86 developers was six.
All of the above would lead Apple to build products at prices they felt they couldn't build before, and along the way, attract more developers who otherwise didn't want to bring products to a new chip family, using tools they weren't familiar with, as WELL as a new OS.
I would have to add to that by saying six reasons was just the tip of the iceberg. They had more reasons than those, but they all played a part in their decision to switch, and it really wasn't that big of a decision once you add them all up. Apple was right to leave. THey would have had a hard time growing with a manufacturer like IBM who couldn't even supply sufficient amounts of processors for existing buyers let alone new ones.
With full 64 bit support in Leopard, universal binaries may contain code for up to 4 processor architectures: 32 bit Intel and PPC and 64 bit Intel and PPC. The info on Apples Leopard pages suggests that they are going to have ONE version of Leopard, not 4. For comparison, Windows has different versions for 64-bit. That is, Apples Frameworks and libraries are going to be universal (4 architectures) as well.
As soon as developers move to XCode, compiling universal applications is simple. It took Adobe CS3 a bit longer because they used different tools and (probably) some of their code base was not portable enough. The open source projects usually pay big attention to portability (even if performance is the price to pay), and this is not likely to change.
I am sure Apple is going to keep their code portable in the future. Now, when they improved portability with Intel transition and Leopard, why break it?
With full 64 bit support in Leopard, universal binaries may contain code for up to 4 processor architectures: 32 bit Intel and PPC and 64 bit Intel and PPC. The info on Apples Leopard pages suggests that they are going to have ONE version of Leopard, not 4. For comparison, Windows has different versions for 64-bit. That is, Apples Frameworks and libraries are going to be universal (4 architectures) as well.
As soon as developers move to XCode, compiling universal applications is simple. It took Adobe CS3 a bit longer because they used different tools and (probably) some of their code base was not portable enough. The open source projects usually pay big attention to portability (even if performance is the price to pay), and this is not likely to change.
I am sure Apple is going to keep their code portable in the future. Now, when they improved portability with Intel transition and Leopard, why break it?
Apple is interested in moving on from old computers and they have made that obvious. They are not making versions of Aperture compatible with every old Mac, same goes for Motion, and other applications. Even leopard doesn't have GUI features on G4's that are available on G5's. Apple isn't going to make Universal binary versions of every bit of software forever. They may do it in house, but they wont keep releasing them. Apple is interested in selling new computers.. not seeing how long you can get by on a 486 processor.
With full 64 bit support in Leopard, universal binaries may contain code for up to 4 processor architectures: 32 bit Intel and PPC and 64 bit Intel and PPC. The info on Apples Leopard pages suggests that they are going to have ONE version of Leopard, not 4. For comparison, Windows has different versions for 64-bit. That is, Apples Frameworks and libraries are going to be universal (4 architectures) as well.
As soon as developers move to XCode, compiling universal applications is simple. It took Adobe CS3 a bit longer because they used different tools and (probably) some of their code base was not portable enough. The open source projects usually pay big attention to portability (even if performance is the price to pay), and this is not likely to change.
I am sure Apple is going to keep their code portable in the future. Now, when they improved portability with Intel transition and Leopard, why break it?
Because it's a waste of time and money, and they will have no use for it.
As soon as developers move to XCode, compiling universal applications is simple. It took Adobe CS3 a bit longer because they used different tools and (probably) some of their code base was not portable enough.
Your reason sounds right to me -- Adobe no doubt had been using older tools for a long time. Let's hope Adobe converted everything to xCode this time, making future modifications easier.
There were several factors. It wasn't just any one.
Performance was one.
energy efficiency was two.
Well IBM can build a suitable CPU for every customer but Apple? No. The real difference Apple didn't want to pay for it, that's why we never saw a mobile G5.
Quote:
Originally Posted by melgross
New chip feature sets was three.
Name one.
Quote:
Originally Posted by melgross
Manufacturer reliabilty was four.
Ok maybe.
Quote:
Originally Posted by melgross
The advantage of Intel's vast board design and manufacturering abilities was five.
Which only save them the money to design it on their own.
Quote:
Originally Posted by melgross
x86 developers was six.
If you build a universal, do you care about x86 or PPC ?
Quote:
Originally Posted by melgross
All of the above would lead Apple to build products at prices they felt they couldn't build before, and along the way, attract more developers who otherwise didn't want to bring products to a new chip family, using tools they weren't familiar with, as WELL as a new OS.
Name those magic products from "attracted" developers.
The Power 6 may be fanastic but for the price of say 4 of those Apple could buy many many more Intel CPUs and still make a good rendering farm. If there is no G3 support in 10.5 I would not expect any PPC support in 10.6 nor would I care even if I still like my G4 tower.
The Power 6 may be fanastic but for the price of say 4 of those Apple could buy many many more Intel CPUs and still make a good rendering farm. If there is no G3 support in 10.5 I would not expect any PPC support in 10.6 nor would I care even if I still like my G4 tower.
good point, but apple would never have used the POWER6 in one of their products, but we wont get into that, its been/being discussed elsewhere in the forum.
as regards to ppc support in the future, apple has to draw the line somewhere. while they pride themselves on being able to run there most up-to-date software on machines as old as 6 or 7 years, i think it is, they cant do so forever. im thinking:
10.5 = g3 support lost, everything else supported
10.6 = last release for g4 and g5, but would need pretty high specs on the g4 to run the OS, mainly being moderately good precessor, ram etc. might need other certain hardware specs such as a hd-dvd or blu-ray drive for installation as well, if apples made its mind up by then...
Well IBM can build a suitable CPU for every customer but Apple?
I don't think any serious observer claimed that IBM could not build a suitable processor for Apple. They do claim, however, that IBM had turned its attention elsewhere--until it was too late. If you remember, after Apple announced its switch to Intel, IBM tried to persuade it to change its decision.
Quote:
Originally Posted by User Tron
No. The real difference Apple didn't want to pay for it, that's why we never saw a mobile G5.
....
Apple was a bad customer? How high did you have to reach up your butt to pull that one out? The PPC is not one processor, but a family of processors used by IBM and in numerous embedded applications. Every G5 (PPC 970) processor produced so much heat that it required its own custom-mated cooling unit. I am aware of no information that IBM ever tried to develop a processor using he PPC 970 ISA which was suitable for use in a laptop.
Well IBM can build a suitable CPU for every customer but Apple? No. The real difference Apple didn't want to pay for it, that's why we never saw a mobile G5.
IBM can build a suitable CPU for use in boxes that do not demand 'SPEED BOOSTS' every 6 to 9 months. Microsoft contracted with IBM for the 360 and guess what... the CPU that they used when building the very first box is the same speed as the CPU they are going to put into the box in 2008.
You think IBM can build a CPU for *everyone* but you're wrong, they can't... Even in the "power' line a CPU that would NEVER be usable in a box costing less than $10k / $20k (to start) doesn't get 'speed bumps' at anywhere near the frequency of 'normal desktop processors'.
IBM has a good CPU but they simply can't advance their CPUs at the pace demanded by desktop/laptop users.
Just my .02
Dave
P.S. Oh and I really don't mean **they can't** but simply they choose not to..
Well IBM can build a suitable CPU for every customer but Apple? No. The real difference Apple didn't want to pay for it, that's why we never saw a mobile G5.
And you are pointing out a big reason for going with Intel. Intel spreads out the cost of developing desktop and laptop CPUs over all PC makers.
Apple was at a big disadvantage with IBM, and Motorola, because Apple was the only customer for these PC CPUs. With Apple footing the bill for CPU development, and support chips too, Apple could not compete as well.
The only economic way Apple could have continued with IBM is if IBM had been willing to absorb most of the PPC development cost, as a marketing strategy to stay in the personal computer market with the PPC. IBM evidently did not want to follow this course.
Now, to address the question of this discussion, IBM might be willing to foot the bill to develop a competitive PPC cell phone CPU that Apple could use. Such a move would not surprise me at all. By the way, do we know what processor is in the iPhone?
I am sure Apple is going to keep their code portable in the future. Now, when they improved portability with Intel transition and Leopard, why break it?
Quote:
Originally Posted by melgross
Because it's a waste of time and money, and they will have no use for it.
Quote:
Originally Posted by onlooker
Apple is interested in moving on from old computers and they have made that obvious. They are not making versions of Aperture compatible with every old Mac, same goes for Motion, and other applications. Even leopard doesn't have GUI features on G4's that are available on G5's. Apple isn't going to make Universal binary versions of every bit of software forever. They may do it in house, but they wont keep releasing them. Apple is interested in selling new computers.. not seeing how long you can get by on a 486 processor.
You guys miss the point entirely!
Look back at Mac OS history: first on Motorola 68k, then Intel, then PowerPC, than Intel again. The biggest problem for the last 2 transitions was Mac OS 9 legacy code, which was not portable. It was resolved with Carbon and dropping Classic for Intel. Now that the issues are fixed, why should Apple break them??? Of course, they are not going to release universal applications with binaries for every processor out there, only for subset of supported ones. But they *WILL* keep their code base portable.
Go have a look at the open source stuff out there! Now, with Apple switching to Intel, what is the reason to keep compatibility with other processors and OSes? But they DO support other processors and OSes! And Apple is using open source heavily. Darwin - the core of the OS - is open sourced!
Now, go to some scientific projects and you will find that some guys are using/porting FORTRAN libraries from the 60-ies! (FORTRAN is a programming language, for those of you less computer educated ).
And now you are telling me that Apple is going to abandon code portability because Intel Rules?
Edit: Oh, and now, when Apple is spreading MacOS to a wide range of devices, they will drop portability, "Because it's a waste of time and money, and they will have no use for it."?
Well IBM can build a suitable CPU for every customer but Apple? No. The real difference Apple didn't want to pay for it, that's why we never saw a mobile G5.
That's absurd! No customer should ever have to pay for more than the product. It's not as though Apple was the only customer for the chips, even though they were the largest. This was up to IBM. If they wanted to sell their chips to a wider audience (and remember they used those chips in their own servers) then they had the responsibility to do their own investments. IBM was pushing the idea of a ciommunity of PPC users. It never got off the ground. If Apple had to bribe IBM, by paying them an extra hundred million or two each year, to help pay for R&D, Apple's costs would have risen too high, and then you would have complained about Apple's even higher pricing.
Quote:
Name one.
If you're so far out of it that you haven't been paying attention to what Intel is doing, and to what IBM hasn't been doing, you won't believe me anyway. Read some of the articles here about Intel.
Quote:
Ok maybe.
Ok, definitely.
Quote:
Which only save them the money to design it on their own.
Hello! What are we talking about?
Quote:
If you build a universal, do you care about x86 or PPC ?
Read the posts here please. I'm not going to go over that again.
Quote:
Name those magic products from "attracted" developers.
I'm not going to go looking for a lot of software. But this is an example of what I mean. We're seeing some heavy hitters in the Windows business space moving to the Mac because of the switch to Intel. I have others bookmarked, But I don't have time now to find them.
Look back at Mac OS history: first on Motorola 68k, then Intel, then PowerPC, than Intel again. The biggest problem for the last 2 transitions was Mac OS 9 legacy code, which was not portable. It was resolved with Carbon and dropping Classic for Intel. Now that the issues are fixed, why should Apple break them??? Of course, they are not going to release universal applications with binaries for every processor out there, only for subset of supported ones. But they *WILL* keep their code base portable.
Go have a look at the open source stuff out there! Now, with Apple switching to Intel, what is the reason to keep compatibility with other processors and OSes? But they DO support other processors and OSes! And Apple is using open source heavily. Darwin - the core of the OS - is open sourced!
Now, go to some scientific projects and you will find that some guys are using/porting FORTRAN libraries from the 60-ies! (FORTRAN is a programming language, for those of you less computer educated ).
And now you are telling me that Apple is going to abandon code portability because Intel Rules?
Edit: Oh, and now, when Apple is spreading MacOS to a wide range of devices, they will drop portability, "Because it's a waste of time and money, and they will have no use for it."?
You got it wrong from the beginning.
When was OS X ever being sold as an x86 OS before Apple completely switched over?
Never!
You can't say what you did, because it's just wrong.
Apple inherited an x86 OS with NEXT, but despite flirting with the Yellow Box for a short time, never released it.
The Intel based OS they had running in their labs was there purely for a bet that they would likely have to move to Intel at some point, which, of course, was exactly what they did.
After you fix your error, you can try to re-write your post so that it reflects it.
I am afraid you still don't get it! Do you understand that Apple currently supports 5 (FIVE) processor architectures? Will they bet there will be no changes in processor architectures the next couple of decades? Because the legacy code in OS X goes before NEXT, there is some older UNIX stuff there, so may be more than 2 decades.
And Yellow Box is another story - it is about support of Apple frameworks under Windows, not about supporting Intel. Apple never made Yellow box widely available, but it was there for a while for OEM use, to fill the gap for some software vendors running on NEXT before OS X was there. But thats not the point...
Oh, and did you notice that it took Apple much longer to add full 64 bit support than to switch to Intel? I don't want to go into a long discussion here on what it takes to support more processor architectures but I believe that most developers, including those working at Apple on Mac OS, don't really care to make special optimizations for a specific processor. They rely on the compiler (Apple is using third party compilers) and, in rare cases, on some highly optimized libraries.
That's absurd! No customer should ever have to pay for more than the product. It's not as though Apple was the only customer for the chips, even though they were the largest. This was up to IBM. If they wanted to sell their chips to a wider audience (and remember they used those chips in their own servers) then they had the responsibility to do their own investments. IBM was pushing the idea of a ciommunity of PPC users. It never got off the ground. If Apple had to bribe IBM, by paying them an extra hundred million or two each year, to help pay for R&D, Apple's costs would have risen too high, and then you would have complained about Apple's even higher pricing.
Apple was IBM's only real customer other than its own servers. Apple would have had to pay some development costs in order to continue development of the 970 series and beyond. When it came right down to it, staying with the PowerPC was a bigger pain than it was worth.
Comments
Sorry for my bad English capabilities but do you agree with me or not? No product released since the switch sofar is a real "child" of the switch. Everything could have been realeased in PowerPC version (maybe a little latter
@Mr. Me:
Sorry for my bad English capabilities but do you agree with me or not? No product released since the switch sofar is a real "child" of the switch. Everything could have been realeased in PowerPC version (maybe a little latter
Apple is a public corporation with a fiduciary responsibility to return a profit to its shareholders. But forced to make a choice, I would have to disagree with you. To be sure, Intel's standard support chips will help Apple cut costs in its production of servers, laptops, and servers. But cutting costs is not the best way to earn a profit. The best way is to create new products which create new markets. Apple is product-driven. The switch to Intel enabled products which were not possible with the PowerPC. For example, it is not possible to design a laptop based on the G5 (PPC 970). The very thought of a G5-based iPhone may cause the sick and elderly to die laughing.
There are many who believe that the switch to Intel is irrevocable. Hence, they believe, Apple will drop support for everything else and concentrate solely on the x86. Well, Apple is using processors other than the x86 in the iPhone and Apple TV. By keeping MacOS X processor-agnostic, Apple will be able to develop other new and innovative products based on the OS and whichever processor is most appropriate.
Neither performance nor watt where the reason for the switch. It was money, that's all. Through the switch Apple doesn't have to design their own chips, motherboards ,etc. And they don't have to pay for the cpu development. Intel does all that for Apple and they probably get a special price too. IBM can build more or less any CPU (see POWER6) you want, but you have to pay for that (like Sony, MS and Nintendo did). Apple didn't want to. As both didn't really needed each other, they seperated. Same with Freescale. Apple would use McDonalds CPUs if they thought they could make more money with them.
There were several factors. It wasn't just any one.
Performance was one.
energy efficiency was two.
New chip feature sets was three.
Manufacturer reliabilty was four.
The advantage of Intel's vast board design and manufacturering abilities was five.
x86 developers was six.
All of the above would lead Apple to build products at prices they felt they couldn't build before, and along the way, attract more developers who otherwise didn't want to bring products to a new chip family, using tools they weren't familiar with, as WELL as a new OS.
There were several factors. It wasn't just any one.
Performance was one.
energy efficiency was two.
New chip feature sets was three.
Manufacturer reliabilty was four.
The advantage of Intel's vast board design and manufacturering abilities was five.
x86 developers was six.
All of the above would lead Apple to build products at prices they felt they couldn't build before, and along the way, attract more developers who otherwise didn't want to bring products to a new chip family, using tools they weren't familiar with, as WELL as a new OS.
I would have to add to that by saying six reasons was just the tip of the iceberg. They had more reasons than those, but they all played a part in their decision to switch, and it really wasn't that big of a decision once you add them all up. Apple was right to leave. THey would have had a hard time growing with a manufacturer like IBM who couldn't even supply sufficient amounts of processors for existing buyers let alone new ones.
As soon as developers move to XCode, compiling universal applications is simple. It took Adobe CS3 a bit longer because they used different tools and (probably) some of their code base was not portable enough. The open source projects usually pay big attention to portability (even if performance is the price to pay), and this is not likely to change.
I am sure Apple is going to keep their code portable in the future. Now, when they improved portability with Intel transition and Leopard, why break it?
With full 64 bit support in Leopard, universal binaries may contain code for up to 4 processor architectures: 32 bit Intel and PPC and 64 bit Intel and PPC. The info on Apples Leopard pages suggests that they are going to have ONE version of Leopard, not 4. For comparison, Windows has different versions for 64-bit. That is, Apples Frameworks and libraries are going to be universal (4 architectures) as well.
As soon as developers move to XCode, compiling universal applications is simple. It took Adobe CS3 a bit longer because they used different tools and (probably) some of their code base was not portable enough. The open source projects usually pay big attention to portability (even if performance is the price to pay), and this is not likely to change.
I am sure Apple is going to keep their code portable in the future. Now, when they improved portability with Intel transition and Leopard, why break it?
Apple is interested in moving on from old computers and they have made that obvious. They are not making versions of Aperture compatible with every old Mac, same goes for Motion, and other applications. Even leopard doesn't have GUI features on G4's that are available on G5's. Apple isn't going to make Universal binary versions of every bit of software forever. They may do it in house, but they wont keep releasing them. Apple is interested in selling new computers.. not seeing how long you can get by on a 486 processor.
With full 64 bit support in Leopard, universal binaries may contain code for up to 4 processor architectures: 32 bit Intel and PPC and 64 bit Intel and PPC. The info on Apples Leopard pages suggests that they are going to have ONE version of Leopard, not 4. For comparison, Windows has different versions for 64-bit. That is, Apples Frameworks and libraries are going to be universal (4 architectures) as well.
As soon as developers move to XCode, compiling universal applications is simple. It took Adobe CS3 a bit longer because they used different tools and (probably) some of their code base was not portable enough. The open source projects usually pay big attention to portability (even if performance is the price to pay), and this is not likely to change.
I am sure Apple is going to keep their code portable in the future. Now, when they improved portability with Intel transition and Leopard, why break it?
Because it's a waste of time and money, and they will have no use for it.
As soon as developers move to XCode, compiling universal applications is simple. It took Adobe CS3 a bit longer because they used different tools and (probably) some of their code base was not portable enough.
Your reason sounds right to me -- Adobe no doubt had been using older tools for a long time. Let's hope Adobe converted everything to xCode this time, making future modifications easier.
There were several factors. It wasn't just any one.
Performance was one.
energy efficiency was two.
Well IBM can build a suitable CPU for every customer but Apple? No. The real difference Apple didn't want to pay for it, that's why we never saw a mobile G5.
New chip feature sets was three.
Name one.
Manufacturer reliabilty was four.
Ok maybe.
The advantage of Intel's vast board design and manufacturering abilities was five.
Which only save them the money to design it on their own.
x86 developers was six.
If you build a universal, do you care about x86 or PPC ?
All of the above would lead Apple to build products at prices they felt they couldn't build before, and along the way, attract more developers who otherwise didn't want to bring products to a new chip family, using tools they weren't familiar with, as WELL as a new OS.
Name those magic products from "attracted" developers.
The Power 6 may be fanastic but for the price of say 4 of those Apple could buy many many more Intel CPUs and still make a good rendering farm. If there is no G3 support in 10.5 I would not expect any PPC support in 10.6 nor would I care even if I still like my G4 tower.
No way, the PPC door is shut.
The Power 6 may be fanastic but for the price of say 4 of those Apple could buy many many more Intel CPUs and still make a good rendering farm. If there is no G3 support in 10.5 I would not expect any PPC support in 10.6 nor would I care even if I still like my G4 tower.
good point, but apple would never have used the POWER6 in one of their products, but we wont get into that, its been/being discussed elsewhere in the forum.
as regards to ppc support in the future, apple has to draw the line somewhere. while they pride themselves on being able to run there most up-to-date software on machines as old as 6 or 7 years, i think it is, they cant do so forever. im thinking:
10.5 = g3 support lost, everything else supported
10.6 = last release for g4 and g5, but would need pretty high specs on the g4 to run the OS, mainly being moderately good precessor, ram etc. might need other certain hardware specs such as a hd-dvd or blu-ray drive for installation as well, if apples made its mind up by then...
Well IBM can build a suitable CPU for every customer but Apple?
I don't think any serious observer claimed that IBM could not build a suitable processor for Apple. They do claim, however, that IBM had turned its attention elsewhere--until it was too late. If you remember, after Apple announced its switch to Intel, IBM tried to persuade it to change its decision.
No. The real difference Apple didn't want to pay for it, that's why we never saw a mobile G5.
....
Apple was a bad customer? How high did you have to reach up your butt to pull that one out? The PPC is not one processor, but a family of processors used by IBM and in numerous embedded applications. Every G5 (PPC 970) processor produced so much heat that it required its own custom-mated cooling unit. I am aware of no information that IBM ever tried to develop a processor using he PPC 970 ISA which was suitable for use in a laptop.
Well IBM can build a suitable CPU for every customer but Apple? No. The real difference Apple didn't want to pay for it, that's why we never saw a mobile G5.
IBM can build a suitable CPU for use in boxes that do not demand 'SPEED BOOSTS' every 6 to 9 months. Microsoft contracted with IBM for the 360 and guess what... the CPU that they used when building the very first box is the same speed as the CPU they are going to put into the box in 2008.
You think IBM can build a CPU for *everyone* but you're wrong, they can't... Even in the "power' line a CPU that would NEVER be usable in a box costing less than $10k / $20k (to start) doesn't get 'speed bumps' at anywhere near the frequency of 'normal desktop processors'.
IBM has a good CPU but they simply can't advance their CPUs at the pace demanded by desktop/laptop users.
Just my .02
Dave
P.S. Oh and I really don't mean **they can't** but simply they choose not to..
Well IBM can build a suitable CPU for every customer but Apple? No. The real difference Apple didn't want to pay for it, that's why we never saw a mobile G5.
And you are pointing out a big reason for going with Intel. Intel spreads out the cost of developing desktop and laptop CPUs over all PC makers.
Apple was at a big disadvantage with IBM, and Motorola, because Apple was the only customer for these PC CPUs. With Apple footing the bill for CPU development, and support chips too, Apple could not compete as well.
The only economic way Apple could have continued with IBM is if IBM had been willing to absorb most of the PPC development cost, as a marketing strategy to stay in the personal computer market with the PPC. IBM evidently did not want to follow this course.
Now, to address the question of this discussion, IBM might be willing to foot the bill to develop a competitive PPC cell phone CPU that Apple could use. Such a move would not surprise me at all. By the way, do we know what processor is in the iPhone?
I am sure Apple is going to keep their code portable in the future. Now, when they improved portability with Intel transition and Leopard, why break it?
Because it's a waste of time and money, and they will have no use for it.
Apple is interested in moving on from old computers and they have made that obvious. They are not making versions of Aperture compatible with every old Mac, same goes for Motion, and other applications. Even leopard doesn't have GUI features on G4's that are available on G5's. Apple isn't going to make Universal binary versions of every bit of software forever. They may do it in house, but they wont keep releasing them. Apple is interested in selling new computers.. not seeing how long you can get by on a 486 processor.
You guys miss the point entirely!
Look back at Mac OS history: first on Motorola 68k, then Intel, then PowerPC, than Intel again. The biggest problem for the last 2 transitions was Mac OS 9 legacy code, which was not portable. It was resolved with Carbon and dropping Classic for Intel. Now that the issues are fixed, why should Apple break them??? Of course, they are not going to release universal applications with binaries for every processor out there, only for subset of supported ones. But they *WILL* keep their code base portable.
Go have a look at the open source stuff out there! Now, with Apple switching to Intel, what is the reason to keep compatibility with other processors and OSes? But they DO support other processors and OSes! And Apple is using open source heavily. Darwin - the core of the OS - is open sourced!
Now, go to some scientific projects and you will find that some guys are using/porting FORTRAN libraries from the 60-ies! (FORTRAN is a programming language, for those of you less computer educated
And now you are telling me that Apple is going to abandon code portability because Intel Rules?
Edit: Oh, and now, when Apple is spreading MacOS to a wide range of devices, they will drop portability, "Because it's a waste of time and money, and they will have no use for it."?
Well IBM can build a suitable CPU for every customer but Apple? No. The real difference Apple didn't want to pay for it, that's why we never saw a mobile G5.
That's absurd! No customer should ever have to pay for more than the product. It's not as though Apple was the only customer for the chips, even though they were the largest. This was up to IBM. If they wanted to sell their chips to a wider audience (and remember they used those chips in their own servers) then they had the responsibility to do their own investments. IBM was pushing the idea of a ciommunity of PPC users. It never got off the ground. If Apple had to bribe IBM, by paying them an extra hundred million or two each year, to help pay for R&D, Apple's costs would have risen too high, and then you would have complained about Apple's even higher pricing.
Name one.
If you're so far out of it that you haven't been paying attention to what Intel is doing, and to what IBM hasn't been doing, you won't believe me anyway. Read some of the articles here about Intel.
Ok maybe.
Ok, definitely.
Which only save them the money to design it on their own.
Hello! What are we talking about?
If you build a universal, do you care about x86 or PPC ?
Read the posts here please. I'm not going to go over that again.
Name those magic products from "attracted" developers.
I'm not going to go looking for a lot of software. But this is an example of what I mean. We're seeing some heavy hitters in the Windows business space moving to the Mac because of the switch to Intel. I have others bookmarked, But I don't have time now to find them.
http://www.kx.com/news/press-releases/pr070417.php
You guys miss the point entirely!
Look back at Mac OS history: first on Motorola 68k, then Intel, then PowerPC, than Intel again. The biggest problem for the last 2 transitions was Mac OS 9 legacy code, which was not portable. It was resolved with Carbon and dropping Classic for Intel. Now that the issues are fixed, why should Apple break them??? Of course, they are not going to release universal applications with binaries for every processor out there, only for subset of supported ones. But they *WILL* keep their code base portable.
Go have a look at the open source stuff out there! Now, with Apple switching to Intel, what is the reason to keep compatibility with other processors and OSes? But they DO support other processors and OSes! And Apple is using open source heavily. Darwin - the core of the OS - is open sourced!
Now, go to some scientific projects and you will find that some guys are using/porting FORTRAN libraries from the 60-ies! (FORTRAN is a programming language, for those of you less computer educated
And now you are telling me that Apple is going to abandon code portability because Intel Rules?
Edit: Oh, and now, when Apple is spreading MacOS to a wide range of devices, they will drop portability, "Because it's a waste of time and money, and they will have no use for it."?
You got it wrong from the beginning.
When was OS X ever being sold as an x86 OS before Apple completely switched over?
Never!
You can't say what you did, because it's just wrong.
Apple inherited an x86 OS with NEXT, but despite flirting with the Yellow Box for a short time, never released it.
The Intel based OS they had running in their labs was there purely for a bet that they would likely have to move to Intel at some point, which, of course, was exactly what they did.
After you fix your error, you can try to re-write your post so that it reflects it.
You got it wrong from the beginning.
I am afraid you still don't get it! Do you understand that Apple currently supports 5 (FIVE) processor architectures? Will they bet there will be no changes in processor architectures the next couple of decades? Because the legacy code in OS X goes before NEXT, there is some older UNIX stuff there, so may be more than 2 decades.
And Yellow Box is another story - it is about support of Apple frameworks under Windows, not about supporting Intel. Apple never made Yellow box widely available, but it was there for a while for OEM use, to fill the gap for some software vendors running on NEXT before OS X was there. But thats not the point...
Oh, and did you notice that it took Apple much longer to add full 64 bit support than to switch to Intel? I don't want to go into a long discussion here on what it takes to support more processor architectures but I believe that most developers, including those working at Apple on Mac OS, don't really care to make special optimizations for a specific processor. They rely on the compiler (Apple is using third party compilers) and, in rare cases, on some highly optimized libraries.
That's absurd! No customer should ever have to pay for more than the product. It's not as though Apple was the only customer for the chips, even though they were the largest. This was up to IBM. If they wanted to sell their chips to a wider audience (and remember they used those chips in their own servers) then they had the responsibility to do their own investments. IBM was pushing the idea of a ciommunity of PPC users. It never got off the ground. If Apple had to bribe IBM, by paying them an extra hundred million or two each year, to help pay for R&D, Apple's costs would have risen too high, and then you would have complained about Apple's even higher pricing.
Apple was IBM's only real customer other than its own servers. Apple would have had to pay some development costs in order to continue development of the 970 series and beyond. When it came right down to it, staying with the PowerPC was a bigger pain than it was worth.