It's supposed to be everything the Power4 was, just a whole lot faster and far more scaleable;
Well, it certainly isn't supposed to be a whole lot faster. The Power4 is a completely different beast. However, it is supposed to be far more scalable; some have attributed this to the lack of need for the 970 to be as bulletproof as the Power4.
No. Power4 is not what Apple is using, the 970 is. Too many differences for reliable application testing, but close enough probably for quite a bit of Darwin kernel work.
I would say exactly the opposite, actually. The user mode execution model is usually the thing which changes the least, whereas the supervisor mode and special function registers often go through subtle changes between processor versions and generations. I doubt Apple would have provided POWER4s to 3rd parties, but its not unprecedented -- back when the first PowerMacs were coming (in 93-94) developers had to have IBM RS/6000s to develop software for the new Macs. The lack of AltiVec in the POWER4, however, would limit how much real optimization work could be done and it would mostly be a matter of getting compilers and environments working properly. Early work on the 64-bit APIs would be a good candidate for development on POWER4 hardware. Some of the cache differences could be explored and researched as well.
Why do you think it's a so far fetched idea for Apple to have OSX today running on computers sporting Power-processors?
But what you actually wrote was that they had been doing this for several years. They're still stuggling to get a stable/mature MacOS X, I can't see them expending major developer time on every processsor out there.
Besides which, and this is what is becoming really quite monotonous on this board, you just pulled that piece of non-information out of the air. It's not even informed opinion, because you admit yourself you don't know what you're talking about. I'm all for pissing in the wind, but first you at least have to have a drink - some of the people here don't even seem to have ever held a glass in their hand.
But what you actually wrote was that they had been doing this for several years. They're still stuggling to get a stable/mature MacOS X, I can't see them expending major developer time on every processsor out there.
This is hardly "every processor out there". People talk about x86 ports, for heaven's sake! A port to POWER4 would take less than 1/10th the effort of an x86 port. Apple has had a good idea that the 970 was in their future since 2000 and the first remotely close hardware that they could actually lay their hands on was the POWER4 delivered in (IIRC) late 2001. This is hardly a stretch, it would be an obvious and logical step for Apple to take.
This is hardly "every processor out there". People talk about x86 ports, for heaven's sake! A port to POWER4 would take less than 1/10th the effort of an x86 port.... This is hardly a stretch, it would be an obvious and logical step for Apple to take.
But NExTStep ran on X86, so it's hardly outside the realms of reason to start with. And of course there was YellowBox, which was the X86 abstraction layer (or was it red box!? - whatever). I really don't see the logic in porting to a chip that in fact you *know* you're not going to use!?
I also don't see that, having had a hand in designing the chip (must have, right), why you'd be wanting to mess around with things you could not be implementing on, when you'd have prototypes aplenty long before you could even think about going into production?
I really don't see the logic in porting to a chip that in fact you *know* you're not going to use!?
No, they're not going to use that exact chip, but they are going to use the 970, which is closely related to the POWER series. The POWER4 has the same 64-bit ISA they're going to use with the 970, so that would give them some time to practice writing 64 bit powerpc code. Seems perfectly logical to me.
I don't know how early in the development process Apple would get 970 prototypes from IBM, but if they have been playing with the POWER4 for over a year before laying their hands on the 970, they could have written a lot of 64-bit code. Then when the 970 finally arrived, they could concentrate most of their efforts on optimizing the code for the 970.
We know that Apple is doing marklar as a backup plan if the PowerPC fails to deliver. So they are using a lot of resources on a processor that maybe NEVER will see the light of day in a mac. I imagine doing this is taking up way more resources than it would take to "port" over to the POWER4. And they could use code written for the POWER4 and use it directly on the 970 when it arrived at campus. Pure logic isn't it
But what you actually wrote was that they had been doing this for several years.
No I didn't. I said things like "they should have" and "they might have" witch is consistent with me not really knowing. Programmer on the other hand seems to know a lot, and since he's backing me up, I'm fairly confident that my assumptions are correct.
Quote:
I'm all for pissing in the wind, but first you at least have to have a drink - some of the people here don't even seem to have ever held a glass in their hand.
Piss all you like but please do not piss on me.
I follow these boards very carefully and I've learned a lot from some very competent people. I've read every article there is to read concerning the future of PowerPC's i general and 970 specifically, so I consider myself at least enough educated as to have an oppinion. I don't know the ins and outs of porting operating systems to other plattforms but when the likes of IBM representatives and Programmer says that it's a farily small piece of work, I think it's very reasonable for Apple to have done so.. even if it's just to evaluate the plattform.
In all fairness, we don't know for ceratin that apple is going to use the 970, it's just very plausible. But.. since Apple is not stupid, they must do at least some homework before making a very large strategic decision like this, and one lesson to learn is to evaluate the 64 bit version of the PowerPC ISA. Apple have had the opportunity to do this for several years, and I really think they have done this. I expect them to have done that. Everything else would be just plain wrong.
Everything you have written is just outright conjecture - as you write there's not even any confirmation that Apple is going to use the 970.
I also don't have any idea what your criteria is for judging valid posts is, someone called "programmer" must surely know what they are talking about?
So a poster called Steve Jobs would have you enraptured?
You say you've read alot of stuff here, but what's the objective value of posts - very low at the moment if you ask me, at least half the threads started are based on *no actual information*.
So, has Apple been using Power4- or Power5-based boards in secret machines in secret labs...? Wrong question. The question is, why would they do that? Your proposition, with no programming knowledge whatsoever, is that these chips are similar to a 970 - that we can acknowledge is not even confirmed as an Apple chip.
This is just plain nonsense.
You may as well argue that Apple has been developing Motorola 8500-based boards, because that's our promised "G5" route - which is pretty much universally held to be something of a cancelled project.
Do you want to argue that too?
I really don't see Apple investing great efforts into making their OS 64-bit until they start putting real, live product out. The reasons and the historical precedents are obvious: look how long it took them to produce a PPC native Finder. Apple doesn't have the resources to throw around on dead-end projects - the status of the WWDC should throw a great deal of light on the problems they are having trying to move projects along.
In theory you are correct, but for a Power4 box to be useful in practice for application developers the whole OS would have to be running relatively clean for a significant time (beta quality) including any required drivers/.kexts, something I doubt highly.
As for the early RS/6000s, it is my impression that due to the CHRP efforts at the time their architecture was significantly closer to an Apple PPC implementation than a current IBM Power4 would be to an Apple 970 setup.
My expectation is that Apple would use a POWER4 or two internally to get their OS up and running so that they can be sure to have this base level of software ready for the 970s when they arrive. Once the 970s are there and the hardware guys get them into working boxes, then they can start handing out boxes to key 3rd party developers. Getting an early version of an OS, especially one as portable and modern as MacOSX / Darwin, up and running on new hardware is not a huge amount of effort and doesn't require a large number of engineers -- getting it to a shippable state is but for development they don't have to take it that far. The work they did would be directly applicable to the 970, however, allowing them to accelerate their timetable as much as possible. Adapting PowerPC code to a new flavour of PowerPC would be a lot easier than back-porting all of the changes made to Darwin since it lived on x86 long ago. I'm not saying they haven't done both, but it seems like a stretch to say that they'd ignore the chance to use a POWER4 machine as an early prototype because its too much work!
And for anybody who still doesn't think Apple is going to use the 970... give me a break! Do you need it tattooed on SJ's forehead or something? If somebody told you a meteor was going to land you probably wouldn't believe it until it cratered your house.
The use of RS/6000 boxes by Apple pre-dated the CHRP project(s). The main reason for it was that was where the original POWER compiler & dev environment ran. It was not a happy solution, and it didn't last very long. Fortunately Metrowerks came out of the woodwork and saved the day.
In all fairness, we don't know for ceratin that apple is going to use the 970, it's just very plausible. But.. since Apple is not stupid, they must do at least some homework before making a very large strategic decision like this, and one lesson to learn is to evaluate the 64 bit version of the PowerPC ISA. Apple have had the opportunity to do this for several years, and I really think they have done this. I expect them to have done that. Everything else would be just plain wrong.
You are so right. Computer manufacturers and OS vendors don't just randomly switch CPUs overnight. For Apple to settle on the 970 means that they spent an extraordinary amount of time and money on evaluating the alternatives.
G4 development was obviously lagging as far back as the 500mhz plateau. At this point, executives and board-members, with any degree competence, had to be worried about the future of the platform. The only reasonable course of action would have been to investigate how hard it would be to switch manufactures and/or CPU architectures.
Their investigation could have been purely speculative...
Yet, my money is on Apple actually having played around with the 64bit PPC ISA.
I also don't have any idea what your criteria is for judging valid posts is, someone called "programmer" must surely know what they are talking about?
Yeah, my oppinion about peoples competence is based on their name.
Quote:
So, has Apple been using Power4- or Power5-based boards in secret machines in secret labs...? Wrong question. The question is, why would they do that?
Well, if you actually have followed this threead and others you'd probably know why. Short recap: The 970 is well suited for Apple to use. It is a PowerPC processor. It is 64-bit. One of the major obstacles in Apples deployment of the 970 is its 64-bitness. OSX must be 64-bit aware. If Apple are to use 970 a really great idea would be to get some years head start in developing OSX for 64-bit. Even if they don't intend to use 970, they must have done some evaluation of 64-bit PowerPCs. The Power3 and Power4 processors have been avaiable for several years. They are PowerPC processors. They are 64 bit. OSX is very portable. Porting OSX to the Power platform wouldn't be that hard, nor that expensive.
Quote:
Your proposition, with no programming knowledge whatsoever, is that these chips are similar to a 970
Where on earth did you come to the conclusion that I have no programming knowledge whatsoever? I have in fact quite some experience. What I don't have is experience with porting operating sysems to other plattforms, who among us have? The OSX kernel runs on x86 hardware as we speak (This is a fact, if you didn't know that), large parts of the kernel have earlier been know to run on several other platforms (this is also a fact), and the developer tools needed for a 64-bit PowerPC port of OSX are available and have been available for several years.
The conclusion I came to is that there is absolutely no reason whatsoever for Apple not to have done a Power4-port of OSX in the past. None.
Quote:
This is just plain nonsense.
You seem to be the only one to disagree. What do you base your conclusions on?
Quote:
You may as well argue that Apple has been developing Motorola 8500-based boards, because that's our promised "G5" route - which is pretty much universally held to be something of a cancelled project.
As far as we know the 8500 never materialized, it might never even existed. But again..if it did, and if it indeed was a 64-bit PowerPC processor, I would've expected Apple to have a 64-bit OSX port based on Power3/4 ready in that case also.
But.. just to carry on arguing. The 970 DO exist, this is a fact, witch the whole 8500-thing never was.
Quote:
I really don't see Apple investing great efforts into making their OS 64-bit until they start putting real, live product out.
You seem to have missed that Apple could have real live 64-bit OSX machines running for several years by now. One of the reasons why we havn't seen them at CompUSA is that they'd be insanely expensive. One other is that IBM kept the Power3/4 processors for themseleves. But with the 970 it's compeltely different. IBM is pitching the 970 to the embedden market (this is a fact) and they are going to licence it to anyone who want to build one themseleves in the future (this is also a fact).
Porting the nesessary foundation of OSX to a 64-bit platform wouldn't take much effort, everyone agrees on that except you. Even IBM have stated that it would take much for Apple to do so. A handfull of developers a couple of months is really cheap considering.
Starting development of the operating system of a completely new platform the moment the hardware is ready is just stupid. In this case it is so insanely stupid considering that Apple have absolutely no problems getting a flying start at development. Really, what are you thinking?
Quote:
The reasons and the historical precedents are obvious: look how long it took them to produce a PPC native Finder.
Porting something from 68k (a CISC platform) to PowerPC (a RISC platform) is an undertaking in a completely different league. If you had the slightest idea what this entire discussion is about you'd know that. But.. since you obviously don't have a clue I must make a little parallell:
Apple does have the nessessary foundation of OSX running on x86, a completely different architecture. To our knowlege they don't have any future plans for commercial development of it, but they have it up and running anyway. Why? Partly just because they can. And if they can do THAT, porting OSX to the only 64 bit PowerPC platform readily available today (i can buy one if I had the money) would really be a walk in the park.
Quote:
Apple doesn't have the resources to throw around on dead-end projects
But they do, and they must! They have 4.3 billion dollars in their treasure chest just sitting there. Every company risk developing products wicth just might not get out of their labs, otherwise they are dead.
Despite a year of education on the part of Moki, Programmer and others, there are still f***wit posts with regard to the 970.
Changing to different but faster CPUs happens all the time. Pentium 3 to Athlon and Pentium 4. G3 to G4.
The 64-bit capability of the PowerPC 970 will not interfere at all with Apple's OS development.
Apple probably looked at all the alternatives and decided that a desktop POWER derivative was the best... the alternative is indecision and bankruptcy. Coming to a quick (but not too quick) decision is good management.
Apple have great engineers working for them. They never had problems developing chipsets in the past, what's different with the 970? Of course Apple will be ready with a computer to use the 970 in.
Barto
It gets to the point where you are either a retard or a troll
I would like to clear up the MPC8500 issue. It never existed. Never. The 7500 did exist, at least on Motorola's roadmap. That was a desktop G5, with a G5 modular core. The 7500 disappeared from Motorola's roadmap in November 2001 (presumably after being cancelled). MOSR and TheReg then got suckered in by a source who claimed that an "8500" existed. The "8" designation does exist, and refers to communication chips. I'm sure everyone here wants a Mac with the power of a mid-range router?
Comments
Originally posted by RBR
It's supposed to be everything the Power4 was, just a whole lot faster and far more scaleable;
Well, it certainly isn't supposed to be a whole lot faster. The Power4 is a completely different beast. However, it is supposed to be far more scalable; some have attributed this to the lack of need for the 970 to be as bulletproof as the Power4.
Originally posted by AirSluf
No. Power4 is not what Apple is using, the 970 is. Too many differences for reliable application testing, but close enough probably for quite a bit of Darwin kernel work.
I would say exactly the opposite, actually. The user mode execution model is usually the thing which changes the least, whereas the supervisor mode and special function registers often go through subtle changes between processor versions and generations. I doubt Apple would have provided POWER4s to 3rd parties, but its not unprecedented -- back when the first PowerMacs were coming (in 93-94) developers had to have IBM RS/6000s to develop software for the new Macs. The lack of AltiVec in the POWER4, however, would limit how much real optimization work could be done and it would mostly be a matter of getting compilers and environments working properly. Early work on the 64-bit APIs would be a good candidate for development on POWER4 hardware. Some of the cache differences could be explored and researched as well.
Originally posted by Henriok
Why do you think it's a so far fetched idea for Apple to have OSX today running on computers sporting Power-processors?
But what you actually wrote was that they had been doing this for several years. They're still stuggling to get a stable/mature MacOS X, I can't see them expending major developer time on every processsor out there.
Besides which, and this is what is becoming really quite monotonous on this board, you just pulled that piece of non-information out of the air. It's not even informed opinion, because you admit yourself you don't know what you're talking about. I'm all for pissing in the wind, but first you at least have to have a drink - some of the people here don't even seem to have ever held a glass in their hand.
Originally posted by Clive
But what you actually wrote was that they had been doing this for several years. They're still stuggling to get a stable/mature MacOS X, I can't see them expending major developer time on every processsor out there.
This is hardly "every processor out there". People talk about x86 ports, for heaven's sake! A port to POWER4 would take less than 1/10th the effort of an x86 port. Apple has had a good idea that the 970 was in their future since 2000 and the first remotely close hardware that they could actually lay their hands on was the POWER4 delivered in (IIRC) late 2001. This is hardly a stretch, it would be an obvious and logical step for Apple to take.
Originally posted by Programmer
This is hardly "every processor out there". People talk about x86 ports, for heaven's sake! A port to POWER4 would take less than 1/10th the effort of an x86 port.... This is hardly a stretch, it would be an obvious and logical step for Apple to take.
But NExTStep ran on X86, so it's hardly outside the realms of reason to start with. And of course there was YellowBox, which was the X86 abstraction layer (or was it red box!? - whatever). I really don't see the logic in porting to a chip that in fact you *know* you're not going to use!?
I also don't see that, having had a hand in designing the chip (must have, right), why you'd be wanting to mess around with things you could not be implementing on, when you'd have prototypes aplenty long before you could even think about going into production?
Originally posted by Clive
I really don't see the logic in porting to a chip that in fact you *know* you're not going to use!?
No, they're not going to use that exact chip, but they are going to use the 970, which is closely related to the POWER series. The POWER4 has the same 64-bit ISA they're going to use with the 970, so that would give them some time to practice writing 64 bit powerpc code. Seems perfectly logical to me.
I don't know how early in the development process Apple would get 970 prototypes from IBM, but if they have been playing with the POWER4 for over a year before laying their hands on the 970, they could have written a lot of 64-bit code. Then when the 970 finally arrived, they could concentrate most of their efforts on optimizing the code for the 970.
We know that Apple is doing marklar as a backup plan if the PowerPC fails to deliver. So they are using a lot of resources on a processor that maybe NEVER will see the light of day in a mac. I imagine doing this is taking up way more resources than it would take to "port" over to the POWER4. And they could use code written for the POWER4 and use it directly on the 970 when it arrived at campus. Pure logic isn't it
Originally posted by Clive
But what you actually wrote was that they had been doing this for several years.
No I didn't. I said things like "they should have" and "they might have" witch is consistent with me not really knowing. Programmer on the other hand seems to know a lot, and since he's backing me up, I'm fairly confident that my assumptions are correct.
I'm all for pissing in the wind, but first you at least have to have a drink - some of the people here don't even seem to have ever held a glass in their hand.
Piss all you like but please do not piss on me.
I follow these boards very carefully and I've learned a lot from some very competent people. I've read every article there is to read concerning the future of PowerPC's i general and 970 specifically, so I consider myself at least enough educated as to have an oppinion. I don't know the ins and outs of porting operating systems to other plattforms but when the likes of IBM representatives and Programmer says that it's a farily small piece of work, I think it's very reasonable for Apple to have done so.. even if it's just to evaluate the plattform.
In all fairness, we don't know for ceratin that apple is going to use the 970, it's just very plausible. But.. since Apple is not stupid, they must do at least some homework before making a very large strategic decision like this, and one lesson to learn is to evaluate the 64 bit version of the PowerPC ISA. Apple have had the opportunity to do this for several years, and I really think they have done this. I expect them to have done that. Everything else would be just plain wrong.
Originally posted by NETROMac
Pure logic isn't it
I take it the smiley's for sarcasm, sounds like pure bollocks to me.
Originally posted by Henriok
Everything else would be just plain wrong.
Everything you have written is just outright conjecture - as you write there's not even any confirmation that Apple is going to use the 970.
I also don't have any idea what your criteria is for judging valid posts is, someone called "programmer" must surely know what they are talking about?
So a poster called Steve Jobs would have you enraptured?
You say you've read alot of stuff here, but what's the objective value of posts - very low at the moment if you ask me, at least half the threads started are based on *no actual information*.
So, has Apple been using Power4- or Power5-based boards in secret machines in secret labs...? Wrong question. The question is, why would they do that? Your proposition, with no programming knowledge whatsoever, is that these chips are similar to a 970 - that we can acknowledge is not even confirmed as an Apple chip.
This is just plain nonsense.
You may as well argue that Apple has been developing Motorola 8500-based boards, because that's our promised "G5" route - which is pretty much universally held to be something of a cancelled project.
Do you want to argue that too?
I really don't see Apple investing great efforts into making their OS 64-bit until they start putting real, live product out. The reasons and the historical precedents are obvious: look how long it took them to produce a PPC native Finder. Apple doesn't have the resources to throw around on dead-end projects - the status of the WWDC should throw a great deal of light on the problems they are having trying to move projects along.
Originally posted by AirSluf
In theory you are correct, but for a Power4 box to be useful in practice for application developers the whole OS would have to be running relatively clean for a significant time (beta quality) including any required drivers/.kexts, something I doubt highly.
As for the early RS/6000s, it is my impression that due to the CHRP efforts at the time their architecture was significantly closer to an Apple PPC implementation than a current IBM Power4 would be to an Apple 970 setup.
My expectation is that Apple would use a POWER4 or two internally to get their OS up and running so that they can be sure to have this base level of software ready for the 970s when they arrive. Once the 970s are there and the hardware guys get them into working boxes, then they can start handing out boxes to key 3rd party developers. Getting an early version of an OS, especially one as portable and modern as MacOSX / Darwin, up and running on new hardware is not a huge amount of effort and doesn't require a large number of engineers -- getting it to a shippable state is but for development they don't have to take it that far. The work they did would be directly applicable to the 970, however, allowing them to accelerate their timetable as much as possible. Adapting PowerPC code to a new flavour of PowerPC would be a lot easier than back-porting all of the changes made to Darwin since it lived on x86 long ago. I'm not saying they haven't done both, but it seems like a stretch to say that they'd ignore the chance to use a POWER4 machine as an early prototype because its too much work!
And for anybody who still doesn't think Apple is going to use the 970... give me a break! Do you need it tattooed on SJ's forehead or something? If somebody told you a meteor was going to land you probably wouldn't believe it until it cratered your house.
The use of RS/6000 boxes by Apple pre-dated the CHRP project(s). The main reason for it was that was where the original POWER compiler & dev environment ran. It was not a happy solution, and it didn't last very long. Fortunately Metrowerks came out of the woodwork and saved the day.
Originally posted by Henriok
In all fairness, we don't know for ceratin that apple is going to use the 970, it's just very plausible. But.. since Apple is not stupid, they must do at least some homework before making a very large strategic decision like this, and one lesson to learn is to evaluate the 64 bit version of the PowerPC ISA. Apple have had the opportunity to do this for several years, and I really think they have done this. I expect them to have done that. Everything else would be just plain wrong.
You are so right. Computer manufacturers and OS vendors don't just randomly switch CPUs overnight. For Apple to settle on the 970 means that they spent an extraordinary amount of time and money on evaluating the alternatives.
G4 development was obviously lagging as far back as the 500mhz plateau. At this point, executives and board-members, with any degree competence, had to be worried about the future of the platform. The only reasonable course of action would have been to investigate how hard it would be to switch manufactures and/or CPU architectures.
Their investigation could have been purely speculative...
Yet, my money is on Apple actually having played around with the 64bit PPC ISA.
Originally posted by Bigc
Must be that time of the month...
Originally posted by Clive
I also don't have any idea what your criteria is for judging valid posts is, someone called "programmer" must surely know what they are talking about?
Yeah, my oppinion about peoples competence is based on their name.
So, has Apple been using Power4- or Power5-based boards in secret machines in secret labs...? Wrong question. The question is, why would they do that?
Well, if you actually have followed this threead and others you'd probably know why. Short recap: The 970 is well suited for Apple to use. It is a PowerPC processor. It is 64-bit. One of the major obstacles in Apples deployment of the 970 is its 64-bitness. OSX must be 64-bit aware. If Apple are to use 970 a really great idea would be to get some years head start in developing OSX for 64-bit. Even if they don't intend to use 970, they must have done some evaluation of 64-bit PowerPCs. The Power3 and Power4 processors have been avaiable for several years. They are PowerPC processors. They are 64 bit. OSX is very portable. Porting OSX to the Power platform wouldn't be that hard, nor that expensive.
Your proposition, with no programming knowledge whatsoever, is that these chips are similar to a 970
Where on earth did you come to the conclusion that I have no programming knowledge whatsoever? I have in fact quite some experience. What I don't have is experience with porting operating sysems to other plattforms, who among us have? The OSX kernel runs on x86 hardware as we speak (This is a fact, if you didn't know that), large parts of the kernel have earlier been know to run on several other platforms (this is also a fact), and the developer tools needed for a 64-bit PowerPC port of OSX are available and have been available for several years.
The conclusion I came to is that there is absolutely no reason whatsoever for Apple not to have done a Power4-port of OSX in the past. None.
This is just plain nonsense.
You seem to be the only one to disagree. What do you base your conclusions on?
You may as well argue that Apple has been developing Motorola 8500-based boards, because that's our promised "G5" route - which is pretty much universally held to be something of a cancelled project.
As far as we know the 8500 never materialized, it might never even existed. But again..if it did, and if it indeed was a 64-bit PowerPC processor, I would've expected Apple to have a 64-bit OSX port based on Power3/4 ready in that case also.
But.. just to carry on arguing. The 970 DO exist, this is a fact, witch the whole 8500-thing never was.
I really don't see Apple investing great efforts into making their OS 64-bit until they start putting real, live product out.
You seem to have missed that Apple could have real live 64-bit OSX machines running for several years by now. One of the reasons why we havn't seen them at CompUSA is that they'd be insanely expensive. One other is that IBM kept the Power3/4 processors for themseleves. But with the 970 it's compeltely different. IBM is pitching the 970 to the embedden market (this is a fact) and they are going to licence it to anyone who want to build one themseleves in the future (this is also a fact).
Porting the nesessary foundation of OSX to a 64-bit platform wouldn't take much effort, everyone agrees on that except you. Even IBM have stated that it would take much for Apple to do so. A handfull of developers a couple of months is really cheap considering.
Starting development of the operating system of a completely new platform the moment the hardware is ready is just stupid. In this case it is so insanely stupid considering that Apple have absolutely no problems getting a flying start at development. Really, what are you thinking?
The reasons and the historical precedents are obvious: look how long it took them to produce a PPC native Finder.
Porting something from 68k (a CISC platform) to PowerPC (a RISC platform) is an undertaking in a completely different league. If you had the slightest idea what this entire discussion is about you'd know that. But.. since you obviously don't have a clue I must make a little parallell:
Apple does have the nessessary foundation of OSX running on x86, a completely different architecture. To our knowlege they don't have any future plans for commercial development of it, but they have it up and running anyway. Why? Partly just because they can. And if they can do THAT, porting OSX to the only 64 bit PowerPC platform readily available today (i can buy one if I had the money) would really be a walk in the park.
Apple doesn't have the resources to throw around on dead-end projects
But they do, and they must! They have 4.3 billion dollars in their treasure chest just sitting there. Every company risk developing products wicth just might not get out of their labs, otherwise they are dead.
Changing to different but faster CPUs happens all the time. Pentium 3 to Athlon and Pentium 4. G3 to G4.
The 64-bit capability of the PowerPC 970 will not interfere at all with Apple's OS development.
Apple probably looked at all the alternatives and decided that a desktop POWER derivative was the best... the alternative is indecision and bankruptcy. Coming to a quick (but not too quick) decision is good management.
Apple have great engineers working for them. They never had problems developing chipsets in the past, what's different with the 970? Of course Apple will be ready with a computer to use the 970 in.
Barto
It gets to the point where you are either a retard or a troll
I wonder what time and energy we should spend convincing these trolls though.
I would like to clear up the MPC8500 issue. It never existed. Never. The 7500 did exist, at least on Motorola's roadmap. That was a desktop G5, with a G5 modular core. The 7500 disappeared from Motorola's roadmap in November 2001 (presumably after being cancelled). MOSR and TheReg then got suckered in by a source who claimed that an "8500" existed. The "8" designation does exist, and refers to communication chips. I'm sure everyone here wants a Mac with the power of a mid-range router?
Barto