It wasn't a serious argument, and then everyone got on their case about it.
You're speculating way too much. Its ARM, there's no question about that.
Why ARM? What's the proof? the chip tear down showed it was fabbed by Samsung, and it was likely 45nm. Chipworks didn't seem to say anything about it being ARM or Power Architecture.
If there are no questions it being ARM - where's the definitive proof? I agree it could be, but it seems it's nice to support the argument of "there's no question it's ARM"
E.g. If it's ARM A8 - They seem to have upgraded the on-chip memory controller to 64 bits wide.
Does the die size and process width actually match up to be an ARM8? Why the delay to OS 4 coming to the iPad? Why Apple's spat with Flash, etc recently? They want all code developed in XCode - and they likely have their reasons.
A chip to chip comparison with a Hummingbird would really clear things up. Apple could been involved in a customisation of an A8 - but the A4 doesn't seem stock Cortex A8 - we agree on that, right? With all Apple's secrecy, it could well be that Samsung may have omitted in roadmaps, a chip it was developing with Apple, and that it is Cortex, just something we don't yet know about - is there more than just the Cortex-A9 MPCore?
Why ARM? What's the proof? the chip tear down showed it was fabbed by Samsung, and it was likely 45nm. Chipworks didn't seem to say anything about it being ARM or Power Architecture.
If there are no questions it being ARM - where's the definitive proof?
Occam's razor, the balance of probabilities, all point to it being a Samsung ARM based Hummingbird with half the memory left out.
Occam's razor, the balance of probabilities, all point to it being a Samsung ARM based Dragonfly with half the memory left out.
So your hard proof, is a meta-theoretical principle that the simplest solution is usually the correct one? Wiki: "In the scientific method, Occam's razor is not considered an irrefutable principle of logic, and certainly not a scientific result."
Why would they drop half the memory? Any place Samsung is saying they're shipping this chip for products in Q2 2010?
What did Jobs say at WWDC once?
"Now, let's go to a big topic - Transitions. Let's talk about transitions. The Mac in it's history has had two major transitions so far.
I - 68k to Power PC. And that transition happened about 10 years ago in the mid 90s. I wasn't here then, but the team then did a great job, from everything I hear. And the Power PC set Apple up for the next decade. It was a good move.
The second major transition though has been even bigger, and that was from OS 9 to OS X that we just finished a few years ago, the early part of this decade. This was a brain transplant, and even though these OS vary in name only by one, they are worlds apart in their technology. OS X is the most advanced OS in the planet, and it has set Apple up for the next 20 years.
Today it's time to begin a 3rd transition. We want to constantly be making the best computers for you and the rest of our users, so it's time for a 3rd transition, and yes, it's true! We are going to begin the transition from the Power PC to Intel processors and we are going to begin it for you now, and for our customers next year.
Now, why are we going to do this, right? Didn't we just get through going from OS 9 to OS X? Isn't the business great right now? Why do we want another transition? Because we want to be making the best computers for our customers looking forward.
.... The most important reasons are that as we look ahead, though we have great products right now, and we've got some great Power PC products still yet to come, as we look ahead, we can envision some amazing products we want to build for you, and we don't know how to build them with the future Power PC roadmap, and that's why we're going to do this.
When we look at Intel, they've got great performance yes, but they've got something else that's very important for us - just as important as performance, is power consumption. And the way we look at it, is performance per Watt. For 1 Watt of power, how much performance do you get. And when we look at the future roadmaps projected out mid-2006 and beyond, what we see is that the Power PC gives us sort of 15 units of performance per Watt, but the Intel Roadmap for the future gives us 70.
And so this tells us what we have to do. Now, this is not going to be a transition that happens overnight.It's going to happen over a period of a few years. "
...
"Two major challenges -The 1st one - making Mac OS X sing on Intel processors. Now, i have something to tell you today, Mac OS X has been leading a secret double life - for the past 5 years.
There have been rumours to this effect...we've had teams doing the Just In Case Scenario
Our rules have been, that our designs for OS X must be processor independent, and that every project must be built for both the Power PC and Intel processors. So for today for the first time, I can confirm the rumours that every release of Mac OS X has been compiled for both Power PC and Intel. This has been going on for the last 5 years. Just In Case. So Mac OS X is cross platform by design.
So Jobs', Apple's main criteria were roadmap, performance, power consumption - Performance per Watt.
There is talk that the A4 is actually a PPC dual core, not ARM at all. If that is the case, then they could have used existing PA Semi designs with modifications.
One thing is clear - Apple wants to differentiate itself on hardware speed + efficiency.
This would have come out in the SDK by now. This would mean from a software perspective supporting both ARM (iPhone/iPod) and iPad (PPC), would be extremely challenging and Apple would need to do something similar to what Apple did with the Mac PPC to Intel change.
Apple are cheap, they got Samsung to drop half the memory to save money and to allow the nonsense that is being promulgated all over the place that the A4 has a significant amount of Apple intellectual property in it. I doubt it has Any Apple IP in it. All it is, is a custom, lower spec version of the hummingbird.
I am not interested in supplying you with the 'proof' you keep demanding. I didn't say I had hard proof, I just think I am more likely to be right than you.
As for Samsung shipping the chip, I am not sure about that but they will be using millions of them in their S8500 Wave and i9000 S Galaxy smart phones which are due for release very shortly. They will probably be using the chip in the BADA S8200 phone as well.
I think this all points to Apple having wasted an awful lot of money buying PA Semi. I think they bought the company, most of the talent left, leaving a portfolio of outdated IP that was soon superseded in performance by the work Intensity did for Samsung. Oops!
This would have come out in the SDK by now. This would mean from a software perspective supporting both ARM (iPhone/iPod) and iPad (PPC), would be extremely challenging and Apple would need to do something similar to what Apple did with the Mac PPC to Intel change.
Is there a way to check the SDK for the iPad, to rule this out?
Guessing Apple moved to LPDDR2 to boost the memory bandwidth? (something other SoC makers have yet to do?
Agree it's cheap to drop the memory - Safari will suffer on the iPad as a result for starters.
Fair enough - You're likely right - just seeing if you could rule out Power Architecture.
I think this all points to Apple having wasted an awful lot of money buying PA Semi. I think they bought the company, most of the talent left, leaving a portfolio of outdated IP that was soon superseded in performance by the work Intensity did for Samsung. Oops!
It's two years since Apple bought PA Semi, and the article says it takes three years for a new product to be designed and produced, so PA Semi designs will hit the streets next year most likely.
Intrinsity will bring in a lot of experience to complement the PA Semi designs, so maybe PA Semi identified the technology as being useful to their designs, and Apple bought it. Expect to see PA Semi + Intrinsity efforts in two or three years from now.
The A4 is just a variant of an Intrinsity design, so that was a bonus. Also because that design was an evolution of the existing iPhone 3GS processor, so software changes were minimal. Higher L2 cache and a wider memory bus provide good performance boosts.
PA Semi are most likely working on the dual-core Cortex A9 based System-on-Chip with SGX 543 graphics for next year's iPad, AppleTV, etc.
Apple are cheap, they got Samsung to drop half the memory to save money and to allow the nonsense that is being promulgated all over the place that the A4 has a significant amount of Apple intellectual property in it. I doubt it has Any Apple IP in it. All it is, is a custom, lower spec version of the hummingbird.
I am not interested in supplying you with the 'proof' you keep demanding. I didn't say I had hard proof, I just think I am more likely to be right than you.
As for Samsung shipping the chip, I am not sure about that but they will be using millions of them in their S8500 Wave and i9000 S Galaxy smart phones which are due for release very shortly. They will probably be using the chip in the BADA S8200 phone as well.
I think this all points to Apple having wasted an awful lot of money buying PA Semi. I think they bought the company, most of the talent left, leaving a portfolio of outdated IP that was soon superseded in performance by the work Intensity did for Samsung. Oops!
What is to say they bought PASemi for the talent? PASemi had a nice portfolio of IP to leverage. Anyone who has done A&M knows that staffing bleed is a natural part of the process - and plan for it. There are synergies involved with the acquisition of PASemi with the existing in-house ASIC team. And with the addition of Intrinsity they can produce some very interesting road-maps.
Good to see another Austin based tech firm succeed. Austin, TX - little Silicon Valley.
Don't move to Austin. It's hot. The food is terrible. It's a bunch of rednecks. It's expensive. The music scene sucks and there are no hills, lakes, or water anywhere around. Please just stay where you are.
Seriously how cool is it that Intrinsity's HQ is listed on Google as being on "Farm to Market Road 2244"...only in Austin...Hook 'Em Horns!
Why ARM? What's the proof? the chip tear down showed it was fabbed by Samsung, and it was likely 45nm. Chipworks didn't seem to say anything about it being ARM or Power Architecture.
If there are no questions it being ARM - where's the definitive proof? I agree it could be, but it seems it's nice to support the argument of "there's no question it's ARM"
E.g. If it's ARM A8 - They seem to have upgraded the on-chip memory controller to 64 bits wide.
First of all, Technomicon isn't known as being reliable. I tend to dismiss anything they have to say. His speculation was gobblygook.
As far as edn.com goes, he didn't even get the other memory supplier correct. In addition to Samsung, it's Toshiba. He's not even sure that there's a compass chipset inside (yes, there is), which is well known.
If these are the two sources, I think you'd better drop them.
9quote0
Does the die size and process width actually match up to be an ARM8? Why the delay to OS 4 coming to the iPad? Why Apple's spat with Flash, etc recently? They want all code developed in XCode - and they likely have their reasons.[/quote]
Die size doesn't have to be the same if there's a difference on the chip itself. If word is correct, and Apple removed unnecessary parts, then they could be going for better wafer and die yield with smaller dies.
Process width? Do you understand what you just asked? You've already stated that it was probably 45nm. From the x-rays, it IS 45nm. It wouldn't be 65 at that speed and power draw. And is surely isn't less.
Quote:
A chip to chip comparison with a Hummingbird would really clear things up. Apple could been involved in a customisation of an A8 - but the A4 doesn't seem stock Cortex A8 - we agree on that, right? With all Apple's secrecy, it could well be that Samsung may have omitted in roadmaps, a chip it was developing with Apple, and that it is Cortex, just something we don't yet know about - is there more than just the Cortex-A9 MPCore?
It certainly seems to be a Hummingbird. That's the only thing that makes sense. Intrinsity and Samsung have the Hummingbird. Apple's chip supplier is Samsung. Apple bought Intrinsity.
If they were using a PPC, they would have to have the entire OS run in emulation. Really, its not likely they've got another secret program to run their OS's back on PPC. When Apple went to Intel, he said that "Power was over". I believe that. I can't think of a single reason to resurrect it.
So your hard proof, is a meta-theoretical principle that the simplest solution is usually the correct one? Wiki: "In the scientific method, Occam's razor is not considered an irrefutable principle of logic, and certainly not a scientific result."
Why would they drop half the memory? Any place Samsung is saying they're shipping this chip for products in Q2 2010?
What did Jobs say at WWDC once?
"Now, let's go to a big topic - Transitions. Let's talk about transitions. The Mac in it's history has had two major transitions so far.
I - 68k to Power PC. And that transition happened about 10 years ago in the mid 90s. I wasn't here then, but the team then did a great job, from everything I hear. And the Power PC set Apple up for the next decade. It was a good move.
The second major transition though has been even bigger, and that was from OS 9 to OS X that we just finished a few years ago, the early part of this decade. This was a brain transplant, and even though these OS vary in name only by one, they are worlds apart in their technology. OS X is the most advanced OS in the planet, and it has set Apple up for the next 20 years.
Today it's time to begin a 3rd transition. We want to constantly be making the best computers for you and the rest of our users, so it's time for a 3rd transition, and yes, it's true! We are going to begin the transition from the Power PC to Intel processors and we are going to begin it for you now, and for our customers next year.
Now, why are we going to do this, right? Didn't we just get through going from OS 9 to OS X? Isn't the business great right now? Why do we want another transition? Because we want to be making the best computers for our customers looking forward.
.... The most important reasons are that as we look ahead, though we have great products right now, and we've got some great Power PC products still yet to come, as we look ahead, we can envision some amazing products we want to build for you, and we don't know how to build them with the future Power PC roadmap, and that's why we're going to do this.
When we look at Intel, they've got great performance yes, but they've got something else that's very important for us - just as important as performance, is power consumption. And the way we look at it, is performance per Watt. For 1 Watt of power, how much performance do you get. And when we look at the future roadmaps projected out mid-2006 and beyond, what we see is that the Power PC gives us sort of 15 units of performance per Watt, but the Intel Roadmap for the future gives us 70.
And so this tells us what we have to do. Now, this is not going to be a transition that happens overnight.It's going to happen over a period of a few years. "
...
"Two major challenges -The 1st one - making Mac OS X sing on Intel processors. Now, i have something to tell you today, Mac OS X has been leading a secret double life - for the past 5 years.
There have been rumours to this effect...we've had teams doing the Just In Case Scenario
Our rules have been, that our designs for OS X must be processor independent, and that every project must be built for both the Power PC and Intel processors. So for today for the first time, I can confirm the rumours that every release of Mac OS X has been compiled for both Power PC and Intel. This has been going on for the last 5 years. Just In Case. So Mac OS X is cross platform by design.
So Jobs', Apple's main criteria were roadmap, performance, power consumption - Performance per Watt.
You didn't give a single reason why it wouldn't be an ARM chip in all of that. You thought that his argument was just inference, but yours was much more so.
Comments
It wasn't a serious argument, and then everyone got on their case about it.
You're speculating way too much. Its ARM, there's no question about that.
Why ARM? What's the proof? the chip tear down showed it was fabbed by Samsung, and it was likely 45nm. Chipworks didn't seem to say anything about it being ARM or Power Architecture.
If there are no questions it being ARM - where's the definitive proof? I agree it could be, but it seems it's nice to support the argument of "there's no question it's ARM"
E.g. If it's ARM A8 - They seem to have upgraded the on-chip memory controller to 64 bits wide.
Are there any other ARM A8's with this?
http://www.technomicon.com/iPad_Saga...10_p_IA49.html
http://www.edn.com/blog/400000040/post/1230053723.html
Does the die size and process width actually match up to be an ARM8? Why the delay to OS 4 coming to the iPad? Why Apple's spat with Flash, etc recently? They want all code developed in XCode - and they likely have their reasons.
A chip to chip comparison with a Hummingbird would really clear things up. Apple could been involved in a customisation of an A8 - but the A4 doesn't seem stock Cortex A8 - we agree on that, right? With all Apple's secrecy, it could well be that Samsung may have omitted in roadmaps, a chip it was developing with Apple, and that it is Cortex, just something we don't yet know about - is there more than just the Cortex-A9 MPCore?
Why ARM? What's the proof? the chip tear down showed it was fabbed by Samsung, and it was likely 45nm. Chipworks didn't seem to say anything about it being ARM or Power Architecture.
If there are no questions it being ARM - where's the definitive proof?
Occam's razor, the balance of probabilities, all point to it being a Samsung ARM based Hummingbird with half the memory left out.
Occam's razor, the balance of probabilities, all point to it being a Samsung ARM based Dragonfly with half the memory left out.
So your hard proof, is a meta-theoretical principle that the simplest solution is usually the correct one? Wiki: "In the scientific method, Occam's razor is not considered an irrefutable principle of logic, and certainly not a scientific result."
Why would they drop half the memory? Any place Samsung is saying they're shipping this chip for products in Q2 2010?
What did Jobs say at WWDC once?
"Now, let's go to a big topic - Transitions. Let's talk about transitions. The Mac in it's history has had two major transitions so far.
I - 68k to Power PC. And that transition happened about 10 years ago in the mid 90s. I wasn't here then, but the team then did a great job, from everything I hear. And the Power PC set Apple up for the next decade. It was a good move.
The second major transition though has been even bigger, and that was from OS 9 to OS X that we just finished a few years ago, the early part of this decade. This was a brain transplant, and even though these OS vary in name only by one, they are worlds apart in their technology. OS X is the most advanced OS in the planet, and it has set Apple up for the next 20 years.
Today it's time to begin a 3rd transition. We want to constantly be making the best computers for you and the rest of our users, so it's time for a 3rd transition, and yes, it's true! We are going to begin the transition from the Power PC to Intel processors and we are going to begin it for you now, and for our customers next year.
Now, why are we going to do this, right? Didn't we just get through going from OS 9 to OS X? Isn't the business great right now? Why do we want another transition? Because we want to be making the best computers for our customers looking forward.
.... The most important reasons are that as we look ahead, though we have great products right now, and we've got some great Power PC products still yet to come, as we look ahead, we can envision some amazing products we want to build for you, and we don't know how to build them with the future Power PC roadmap, and that's why we're going to do this.
When we look at Intel, they've got great performance yes, but they've got something else that's very important for us - just as important as performance, is power consumption. And the way we look at it, is performance per Watt. For 1 Watt of power, how much performance do you get. And when we look at the future roadmaps projected out mid-2006 and beyond, what we see is that the Power PC gives us sort of 15 units of performance per Watt, but the Intel Roadmap for the future gives us 70.
And so this tells us what we have to do. Now, this is not going to be a transition that happens overnight.It's going to happen over a period of a few years. "
...
"Two major challenges -The 1st one - making Mac OS X sing on Intel processors. Now, i have something to tell you today, Mac OS X has been leading a secret double life - for the past 5 years.
There have been rumours to this effect...we've had teams doing the Just In Case Scenario
Our rules have been, that our designs for OS X must be processor independent, and that every project must be built for both the Power PC and Intel processors. So for today for the first time, I can confirm the rumours that every release of Mac OS X has been compiled for both Power PC and Intel. This has been going on for the last 5 years. Just In Case. So Mac OS X is cross platform by design.
So Jobs', Apple's main criteria were roadmap, performance, power consumption - Performance per Watt.
There is talk that the A4 is actually a PPC dual core, not ARM at all. If that is the case, then they could have used existing PA Semi designs with modifications.
One thing is clear - Apple wants to differentiate itself on hardware speed + efficiency.
This would have come out in the SDK by now. This would mean from a software perspective supporting both ARM (iPhone/iPod) and iPad (PPC), would be extremely challenging and Apple would need to do something similar to what Apple did with the Mac PPC to Intel change.
I am not interested in supplying you with the 'proof' you keep demanding. I didn't say I had hard proof, I just think I am more likely to be right than you.
As for Samsung shipping the chip, I am not sure about that but they will be using millions of them in their S8500 Wave and i9000 S Galaxy smart phones which are due for release very shortly. They will probably be using the chip in the BADA S8200 phone as well.
I think this all points to Apple having wasted an awful lot of money buying PA Semi. I think they bought the company, most of the talent left, leaving a portfolio of outdated IP that was soon superseded in performance by the work Intensity did for Samsung. Oops!
This would have come out in the SDK by now. This would mean from a software perspective supporting both ARM (iPhone/iPod) and iPad (PPC), would be extremely challenging and Apple would need to do something similar to what Apple did with the Mac PPC to Intel change.
Is there a way to check the SDK for the iPad, to rule this out?
Guessing Apple moved to LPDDR2 to boost the memory bandwidth? (something other SoC makers have yet to do?
Agree it's cheap to drop the memory - Safari will suffer on the iPad as a result for starters.
Fair enough - You're likely right - just seeing if you could rule out Power Architecture.
ARM/Samsung's roadmap? http://www.slashgear.com/arm-cpu-roa...-2012-2282728/
I think this all points to Apple having wasted an awful lot of money buying PA Semi. I think they bought the company, most of the talent left, leaving a portfolio of outdated IP that was soon superseded in performance by the work Intensity did for Samsung. Oops!
It's two years since Apple bought PA Semi, and the article says it takes three years for a new product to be designed and produced, so PA Semi designs will hit the streets next year most likely.
Intrinsity will bring in a lot of experience to complement the PA Semi designs, so maybe PA Semi identified the technology as being useful to their designs, and Apple bought it. Expect to see PA Semi + Intrinsity efforts in two or three years from now.
The A4 is just a variant of an Intrinsity design, so that was a bonus. Also because that design was an evolution of the existing iPhone 3GS processor, so software changes were minimal. Higher L2 cache and a wider memory bus provide good performance boosts.
PA Semi are most likely working on the dual-core Cortex A9 based System-on-Chip with SGX 543 graphics for next year's iPad, AppleTV, etc.
Apple are cheap, they got Samsung to drop half the memory to save money and to allow the nonsense that is being promulgated all over the place that the A4 has a significant amount of Apple intellectual property in it. I doubt it has Any Apple IP in it. All it is, is a custom, lower spec version of the hummingbird.
I am not interested in supplying you with the 'proof' you keep demanding. I didn't say I had hard proof, I just think I am more likely to be right than you.
As for Samsung shipping the chip, I am not sure about that but they will be using millions of them in their S8500 Wave and i9000 S Galaxy smart phones which are due for release very shortly. They will probably be using the chip in the BADA S8200 phone as well.
I think this all points to Apple having wasted an awful lot of money buying PA Semi. I think they bought the company, most of the talent left, leaving a portfolio of outdated IP that was soon superseded in performance by the work Intensity did for Samsung. Oops!
What is to say they bought PASemi for the talent? PASemi had a nice portfolio of IP to leverage. Anyone who has done A&M knows that staffing bleed is a natural part of the process - and plan for it. There are synergies involved with the acquisition of PASemi with the existing in-house ASIC team. And with the addition of Intrinsity they can produce some very interesting road-maps.
Good to see another Austin based tech firm succeed. Austin, TX - little Silicon Valley.
Don't move to Austin. It's hot. The food is terrible. It's a bunch of rednecks. It's expensive. The music scene sucks and there are no hills, lakes, or water anywhere around. Please just stay where you are.
Seriously how cool is it that Intrinsity's HQ is listed on Google as being on "Farm to Market Road 2244"...only in Austin...Hook 'Em Horns!
http://www.google.com/search?client=...UTF-8&oe=UTF-8
Why ARM? What's the proof? the chip tear down showed it was fabbed by Samsung, and it was likely 45nm. Chipworks didn't seem to say anything about it being ARM or Power Architecture.
If there are no questions it being ARM - where's the definitive proof? I agree it could be, but it seems it's nice to support the argument of "there's no question it's ARM"
E.g. If it's ARM A8 - They seem to have upgraded the on-chip memory controller to 64 bits wide.
Are there any other ARM A8's with this?
http://www.technomicon.com/iPad_Saga...10_p_IA49.html
http://www.edn.com/blog/400000040/post/1230053723.html
First of all, Technomicon isn't known as being reliable. I tend to dismiss anything they have to say. His speculation was gobblygook.
As far as edn.com goes, he didn't even get the other memory supplier correct. In addition to Samsung, it's Toshiba. He's not even sure that there's a compass chipset inside (yes, there is), which is well known.
If these are the two sources, I think you'd better drop them.
9quote0
Does the die size and process width actually match up to be an ARM8? Why the delay to OS 4 coming to the iPad? Why Apple's spat with Flash, etc recently? They want all code developed in XCode - and they likely have their reasons.[/quote]
Die size doesn't have to be the same if there's a difference on the chip itself. If word is correct, and Apple removed unnecessary parts, then they could be going for better wafer and die yield with smaller dies.
Process width? Do you understand what you just asked? You've already stated that it was probably 45nm. From the x-rays, it IS 45nm. It wouldn't be 65 at that speed and power draw. And is surely isn't less.
A chip to chip comparison with a Hummingbird would really clear things up. Apple could been involved in a customisation of an A8 - but the A4 doesn't seem stock Cortex A8 - we agree on that, right? With all Apple's secrecy, it could well be that Samsung may have omitted in roadmaps, a chip it was developing with Apple, and that it is Cortex, just something we don't yet know about - is there more than just the Cortex-A9 MPCore?
It certainly seems to be a Hummingbird. That's the only thing that makes sense. Intrinsity and Samsung have the Hummingbird. Apple's chip supplier is Samsung. Apple bought Intrinsity.
If they were using a PPC, they would have to have the entire OS run in emulation. Really, its not likely they've got another secret program to run their OS's back on PPC. When Apple went to Intel, he said that "Power was over". I believe that. I can't think of a single reason to resurrect it.
So your hard proof, is a meta-theoretical principle that the simplest solution is usually the correct one? Wiki: "In the scientific method, Occam's razor is not considered an irrefutable principle of logic, and certainly not a scientific result."
Why would they drop half the memory? Any place Samsung is saying they're shipping this chip for products in Q2 2010?
What did Jobs say at WWDC once?
"Now, let's go to a big topic - Transitions. Let's talk about transitions. The Mac in it's history has had two major transitions so far.
I - 68k to Power PC. And that transition happened about 10 years ago in the mid 90s. I wasn't here then, but the team then did a great job, from everything I hear. And the Power PC set Apple up for the next decade. It was a good move.
The second major transition though has been even bigger, and that was from OS 9 to OS X that we just finished a few years ago, the early part of this decade. This was a brain transplant, and even though these OS vary in name only by one, they are worlds apart in their technology. OS X is the most advanced OS in the planet, and it has set Apple up for the next 20 years.
Today it's time to begin a 3rd transition. We want to constantly be making the best computers for you and the rest of our users, so it's time for a 3rd transition, and yes, it's true! We are going to begin the transition from the Power PC to Intel processors and we are going to begin it for you now, and for our customers next year.
Now, why are we going to do this, right? Didn't we just get through going from OS 9 to OS X? Isn't the business great right now? Why do we want another transition? Because we want to be making the best computers for our customers looking forward.
.... The most important reasons are that as we look ahead, though we have great products right now, and we've got some great Power PC products still yet to come, as we look ahead, we can envision some amazing products we want to build for you, and we don't know how to build them with the future Power PC roadmap, and that's why we're going to do this.
When we look at Intel, they've got great performance yes, but they've got something else that's very important for us - just as important as performance, is power consumption. And the way we look at it, is performance per Watt. For 1 Watt of power, how much performance do you get. And when we look at the future roadmaps projected out mid-2006 and beyond, what we see is that the Power PC gives us sort of 15 units of performance per Watt, but the Intel Roadmap for the future gives us 70.
And so this tells us what we have to do. Now, this is not going to be a transition that happens overnight.It's going to happen over a period of a few years. "
...
"Two major challenges -The 1st one - making Mac OS X sing on Intel processors. Now, i have something to tell you today, Mac OS X has been leading a secret double life - for the past 5 years.
There have been rumours to this effect...we've had teams doing the Just In Case Scenario
Our rules have been, that our designs for OS X must be processor independent, and that every project must be built for both the Power PC and Intel processors. So for today for the first time, I can confirm the rumours that every release of Mac OS X has been compiled for both Power PC and Intel. This has been going on for the last 5 years. Just In Case. So Mac OS X is cross platform by design.
So Jobs', Apple's main criteria were roadmap, performance, power consumption - Performance per Watt.
You didn't give a single reason why it wouldn't be an ARM chip in all of that. You thought that his argument was just inference, but yours was much more so.