I havent read the entire thread so please bear with me:
Moki: Wouldn't going to Intel mean that you should rewrite a lot of your software or do the OS "protect" the app completly from the CPU architecture. Writing code especially for altivec suggest the rewrite option but I really don't know <img src="confused.gif" border="0">
Going to intel (I assume you mean IA-32) would mean that would would have to recompile all you software. If you use AltiVec (perhaps we should start saying VMX ), you would need to rewrite it in SSE2 or remove the AltiVec code for a performance decease. If you use assembly, you would need to rewrite it painstakingly (but hey, you probably enjoy pain if you code in assembly).
Most pro software and games are optimized for a CPU in their code. The code would have to be re-optimized for similar performance.
If Mac OS X for IA-32 emulated a G4, it would be slow. Very slow. Apparently worse than Virtual PC for Macs.
More seriously, IA-32 would remove Apple's performance differention. Right now, Apple has a negative performance differention. However, if Apple finds PPCs with positive performance differention, then Microsoft will be banished forever from the pro market, as Apple will (for once) have kick-ass hardware AND software.
I don't have a major problem with the technical aspects of _doing_ the switch, even though it would cause too much developer defection (it's way too close to the _last_ migration). I'll grant that it's possible.
My only question is this:
Do a hypothetical price/performance estimate on where Apple's hardware would be relative to Dell.
No matter what specs Apple puts on an x86 box, were I Dell I would ship the _IDENTICAL_ box and discount it to be a loss leader. Intel and MS might even pitch in with special discounts.
If ProgramX is your work's focus (Photoshop say), and ProgramX runs on a $2000 Dual 2.43 GHz Dell at precisely the same speed as it runs on a $2600 Dual 2.43 GHz Apple-x86, which one do you buy? And if Dell says 'Gosh, $200 off', Intel says '$100 off' and MS says 'Free OS for switchers'... That's another $500ish off.
30 million CPUs shipped in a typical quarter.
1 million shipped by Apple.
1000000*500=$500 million a quarter in lost profits for Dell/Intel/MS. Comes closer to $50 million a piece after you take off the margin on the original box. $1500 v $2600.
It's one thing to spend 'extra' on something that is clearly different. It's something else to spend 'extra' to buy precisely the same hardware in a prettier/more functional box.
At the moment, there are things that have run faster (recently) on ppcs than x86s. There's always a niche to crawl into and optimize the bejeezus out of Shake or whatever. But if it's _identical_ hardware, the OS has to be that much snazzier, or the program has to only run on the Apple-box.
'They'd sell more boxes' isn't a reason for the Apple price to drop from $2600. Intel shipped 30 million x86 chips last quarter, Apple shipped 800,000ish. Apple's marketshare would need to go through the freaking ROOF to allow any sort of price-break-through-economies-of-scale to beat where Dell is now.
If they stopped OS development, sure, price could come down some. Does that sound sane?
If they stopped working on the design so much, sure, the price could come down some. But even at 0$ spent on case design, Dell would be out-designing Apple & still undercutting them on price. [It helps when you _really_ outsell everyone else.] Dell outdesigning Apple would cause an instant stroke on the part of Steve.
Name 10 computer manufacturers/OS manufacturers that switched/merged to x86.
Name one that made the switch a profitable one.
Name one that's still in the Fortune 5000. Two if you consider the 'Itanium2' a profitable successful product.
Now, a Shake monster machine ($12,000+) might have an x86 coprocessor or something, but competing in the desktop x86 market would require much stronger arguments than anything I've seen thus far.
IBM adds Vector Unit to new chips or AMD starts fabbing ppcs both seem inherently more sensible.
The one thing that is being forgotten is that Apple sell their computers at a much higher margin than PC makers. The disparity in wholesale cost is not as large as we all think. This price difference rest soley on the shoulders of Apple. I think current Powermacs should be priced higher because of the superior casing but not hundreds of dollars.
[quote]Do a hypothetical price/performance estimate on where Apple's hardware would be relative to Dell.
<hr></blockquote>
This only matters if both the Apple and Dell computers run Mac OS X. Most of us buy Macs, not because we like the hardware better, but because of the Mac OS.
If Apple switched to x86, it could ensure that the Mac OS only ran on Apple branded boxes. Thus whatever price/performance Dell offers doesn't matter. At worst, we'd be where we are today; just that we have an x86 is inside the box instead of a PowerPC.
Personally, I don't see how a $3000 dual 1 Ghz G4 is worse for Apple than a $3000 2.6 Ghz P4 with DDR. Apple would charge more for it than a Wintel 2.6 GHz P4, but at least we'd have much better performance than we do now; and still be able to run Mac OS X.
Over the past 5 or 6 years, Apple has always been missing one crucial part of the equation. For a long time, they had fast machines, but a crappy OS (IMHO). Now Mac OS X is starting to mature, but the hardware is relatively slow. Thus their sales have been largely constant (less than 1 million Macs a quarter since 1995). If Apple could have both a great OS, and great hardware; I think that would be a good first step to regaining some of the market they lost in the mid 90s.
If Apple were to switch, there's a lot of work they have to do. Along with fixing all the endian issues in their own code and frameworks and redoing the runtime model, they have to write a Power PC emulator for x86. If they want Classic to be able to function, they'd better rewrite the 68k emulator for x86 as well.
For Apple's sake, I hope they can find someone else to make a Power PC compatible CPU for them (IBM or otherwise). I have doubts as to whether developers would be happy about the switch, and whether customers would be willing to buy another round of upgrades. If it ever came to a switch to x86, Apple would have to be a little generous with the old bank account. They should fund some of the application development for x86, so that developers can afford to give free or discounted upgrades to the x86 version.
<strong>I havent read the entire thread so please bear with me:
Moki: Wouldn't going to Intel mean that you should rewrite a lot of your software or do the OS "protect" the app completly from the CPU architecture. Writing code especially for altivec suggest the rewrite option but I really don't know </strong><hr></blockquote>
Not really. Off of the top of my head, I can't think of too many places I'd need to change things in most of our products to get them to cross-compile, assuming a full-featured Carbon framework was available for another architecture.
Moki, you would spend the money on a X86-Apple Box, buy a new version of OS X86, and new apps (assuming that software writers would support two ports for X86)?
I think this is different from the 68k jump. I say this because when that jump was made, you could still run your old apps. I have heard here that a X86 can not emulate a PPC (too many registers or something to that effect). So all legacy apps would be gone!
Plus if the machine was able to duel boot, as a developer, why would you write an app for both WinXP and for OS X? Just tell the person to boot into WinXP to run the app. Or someone would do a little translating to WINE (now we are on X86 boxes) and you could run the WinXP title without having to reboot. We would loose even more developers.
Also, how long before a hack came out to allow OS X86 to boot on a home built PC? Look at the XBox situation. There is already a group trying to get Linux to run on it (and given time, I bet they will succeed). It would be the same thing. Given time, any one would be able to install OS X, and there goes Apple's Hardware sales.
As a consumer, if you saw 2 machines, same speed, why would you buy the more expensive one? The hardcore Apple people would buy Apple (maybe), but I would think that Apple wouldn't gain any market share this way.
Too many of you are thinking with your "I want the fastest machine" hats on. You need to look at it from all standpoints (that of the consumers as well). When you do that, it really doesn't make sense for Apple.
But who really knows. Apple has a plan, and none of us know it, that is what is exciting.
Ben
PS Didn't someone bring this up at the last shareholders meeting, and Steve basicly shot him down? Now if they did go through with this move, couldn't they (Apple) get in some trouble for not telling the shareholders the truth?
[quote]I think this is different from the 68k jump. I say this because when that jump was made, you could still run your old apps. I have heard here that a X86 can not emulate a PPC (too many registers or something to that effect). So all legacy apps would be gone!
<hr></blockquote>
It's not impossible, just more difficult. I mean, in the worst case, you can always store the contents of the additional registers in RAM. Ridiculously slow? Yes. But not impossible.
Heck, with DDR 2, you have a 533 MHz bus theoretically more than 4 times as fast as the current MPX PC133 implementation. It may not be too horrible.
[quote]Plus if the machine was able to duel boot, as a developer, why would you write an app for both WinXP and for OS X? Just tell the person to boot into WinXP to run the app. Or someone would do a little translating to WINE (now we are on X86 boxes) and you could run the WinXP title without having to reboot. We would loose even more developers.
<hr></blockquote>
If the machines were able to run Windows, I agree, there would be little motivation for many developers to write a native Mac OS X version. Which is why Apple probably wouldn't allow XP or any other version of Windows to be installed on these machines out of the box - though you can bet that someone like Connectix would have a PC "emulator" (not really an emulator anymore) that lets you use Windows.
OK, there are two things that Neff's prediction is based on:
1) Motorola's and IBM's PPC processor development will continue to see performance enhancements at the current rate over the next 1-2 years.
2) That x86 chips from Intel and or AMD will continue to see performance improvements at the current rate over the next 1-2 years.
If both of these conditions are true at the end of the next two years, then yes it is possible that Apple will invest the significant effort to develop chipsets and motherboards that support Mac OS X running on an x86 processor. But right now, that's a long shot.
A lot can happen in two years to make Andrew Neff look very stupid. (Among other things, the HP-Compaq merger could turn out to be a dud.) The PPC architecture could very well leapfrog the Intel set again. And it could even stay out in front -- depending on the Itanium's performance.
While CNet is touting Neff's oracular prowess, let's not forget that he has been wide of the mark more often than not. At the same time he first suggested the HP-Compaq merger in 2001, he also suggested Dell should buy Gateway and IBM's PC division, which is not going to happen. And in March of that year he floated a rumor that Apple was being put on the Auction block by it's board of directors. All in all, I think Neff is less of a prophet and more of a mountebank.
<strong>Any x86 venture by Apple will not include any kind of PPC emulation. Just erase that from your mind as a precondition.</strong><hr></blockquote>
Erasing precondition...
Error, inconsistent logic...
> Add millions of users who won't switch to mac/x86 & Apple going bankrupt...
hmmm - why is IBM using PPC chips in it's high performance servers if x86 is superior?????? x86 is ugly old. not even intel itself plans to use x86 chips forever (Itanium?)
<strong>Any x86 venture by Apple will not include any kind of PPC emulation. Just erase that from your mind as a precondition.</strong><hr></blockquote>
Then add new FSB timing, cache architectures and completely rewritten Open Firmware. Motherboard changes across the entire hardware line? Big or little endian I/O is just scratching the surface, here. The way Apple designs systems, this would require a massive, long term effort and would be unlikely to be accomplished in any reasonable time frame unless Apple were already at work at it. Apple's research budget (which would probably double or triple) doesn't reflect such an effort.
It'd take at least four years to do. (And that's if they had help from Intel/AMD.) It's not something Apple would undertake for the fun of it, and it's not anything that will happen anytime soon.
Apple is never going to replace the current PPC platform with x86 (emphasis on x86).
they may introduce a new line that features x86, if they feel like having a super-workstation for their movie stuff or whatever, but even that would be a bad PR move.
Moving to PPC from 68k was so massively different to moving from PPC to x86, because PPC was a whole new architecture with a lot of life and potential left in it. x86 on the other hand is about to slowly dissappear from the market within the net 5 to 10 years.
Moving to x86 now would be the most stupid thing Apple could possibly do. But then again they have proven often enough to be fond of worst case solutions, so I probably shouldn't be so certain.
I read a lot of posts and one thing that's apparent is that Apple is always being compared to the PC OEMs; especially Dull.. errrr uh Dell ;-)
One thing many people over look is the fact that Dell does little in the way of R&D. They don'd design their own mobos, supporting chips, cases, processors or anything else -- including the operating system.
What Dell and many other PC OEMs do is buy parts and assemble a system, brand it as their own and sell it to the consumer. I think if Michael Dell had an original idea his head would instantly cave in.
Macs cost more, but Why? If people actually took the time to find the answer to that question, they wouldn't be asking it in the first place.
Apple designs the motherboards, the supporting chips, the ROMs, the case, the power supply -- to some extent the processor and of course an operating system that's easily on par with anything farting out of Redmond. Let's see Dell design their own OS and give M$ some competition.. Yeah, my thoughts too... They couldn't do it and embarrass the hell out of themselves.
On another note, where is Dell contributions with respect to Industry standards or even with crossplatform software like QuickTime. Where is their iPod etc. ? The list can go on and on for quite some time.
I can't believe the cost argument still comes into play when comparing Apple to the rest of the PC OEMs. Imagine if auto manufacturers decided to take the same rout as Dell when building and marketing their cars. That is.. Take the R&D costs out of the automaker's equation.
The point that is being missed here is all that PPC developers (Apple, IBM, MOT, Sony, etc0 are heading for higher bandwidth and multiprocessors. The x86 options are currently on MHz and single bandwidth processors. Apple is taking the other path. The fact they are buying up high end video software companies is to get software fully up to multiprocessors (it seems to me).
As I have said before, Starship Enterprise ain't gonna be run on single 100 GHz processors, everything is going to be multiprocessors passing information around to each other.
Processor speed is definitely a factor, but that is currently a manufacturing problem with IBM/MOT. Bandwidth and multiprocessors will be the key and I hope that is where Apple is headed to.
Comments
Please explain, what is GP-UL?
Thanks.
Moki: Wouldn't going to Intel mean that you should rewrite a lot of your software or do the OS "protect" the app completly from the CPU architecture. Writing code especially for altivec suggest the rewrite option but I really don't know <img src="confused.gif" border="0">
Most pro software and games are optimized for a CPU in their code. The code would have to be re-optimized for similar performance.
If Mac OS X for IA-32 emulated a G4, it would be slow. Very slow. Apparently worse than Virtual PC for Macs.
More seriously, IA-32 would remove Apple's performance differention. Right now, Apple has a negative performance differention. However, if Apple finds PPCs with positive performance differention, then Microsoft will be banished forever from the pro market, as Apple will (for once) have kick-ass hardware AND software.
Barto
[ 08-03-2002: Message edited by: Barto ]</p>
My only question is this:
Do a hypothetical price/performance estimate on where Apple's hardware would be relative to Dell.
No matter what specs Apple puts on an x86 box, were I Dell I would ship the _IDENTICAL_ box and discount it to be a loss leader. Intel and MS might even pitch in with special discounts.
If ProgramX is your work's focus (Photoshop say), and ProgramX runs on a $2000 Dual 2.43 GHz Dell at precisely the same speed as it runs on a $2600 Dual 2.43 GHz Apple-x86, which one do you buy? And if Dell says 'Gosh, $200 off', Intel says '$100 off' and MS says 'Free OS for switchers'... That's another $500ish off.
30 million CPUs shipped in a typical quarter.
1 million shipped by Apple.
1000000*500=$500 million a quarter in lost profits for Dell/Intel/MS. Comes closer to $50 million a piece after you take off the margin on the original box. $1500 v $2600.
It's one thing to spend 'extra' on something that is clearly different. It's something else to spend 'extra' to buy precisely the same hardware in a prettier/more functional box.
At the moment, there are things that have run faster (recently) on ppcs than x86s. There's always a niche to crawl into and optimize the bejeezus out of Shake or whatever. But if it's _identical_ hardware, the OS has to be that much snazzier, or the program has to only run on the Apple-box.
'They'd sell more boxes' isn't a reason for the Apple price to drop from $2600. Intel shipped 30 million x86 chips last quarter, Apple shipped 800,000ish. Apple's marketshare would need to go through the freaking ROOF to allow any sort of price-break-through-economies-of-scale to beat where Dell is now.
If they stopped OS development, sure, price could come down some. Does that sound sane?
If they stopped working on the design so much, sure, the price could come down some. But even at 0$ spent on case design, Dell would be out-designing Apple & still undercutting them on price. [It helps when you _really_ outsell everyone else.] Dell outdesigning Apple would cause an instant stroke on the part of Steve.
Name 10 computer manufacturers/OS manufacturers that switched/merged to x86.
Name one that made the switch a profitable one.
Name one that's still in the Fortune 5000. Two if you consider the 'Itanium2' a profitable successful product.
Now, a Shake monster machine ($12,000+) might have an x86 coprocessor or something, but competing in the desktop x86 market would require much stronger arguments than anything I've seen thus far.
IBM adds Vector Unit to new chips or AMD starts fabbing ppcs both seem inherently more sensible.
<hr></blockquote>
This only matters if both the Apple and Dell computers run Mac OS X. Most of us buy Macs, not because we like the hardware better, but because of the Mac OS.
If Apple switched to x86, it could ensure that the Mac OS only ran on Apple branded boxes. Thus whatever price/performance Dell offers doesn't matter. At worst, we'd be where we are today; just that we have an x86 is inside the box instead of a PowerPC.
Personally, I don't see how a $3000 dual 1 Ghz G4 is worse for Apple than a $3000 2.6 Ghz P4 with DDR. Apple would charge more for it than a Wintel 2.6 GHz P4, but at least we'd have much better performance than we do now; and still be able to run Mac OS X.
Over the past 5 or 6 years, Apple has always been missing one crucial part of the equation. For a long time, they had fast machines, but a crappy OS (IMHO). Now Mac OS X is starting to mature, but the hardware is relatively slow. Thus their sales have been largely constant (less than 1 million Macs a quarter since 1995). If Apple could have both a great OS, and great hardware; I think that would be a good first step to regaining some of the market they lost in the mid 90s.
If Apple were to switch, there's a lot of work they have to do. Along with fixing all the endian issues in their own code and frameworks and redoing the runtime model, they have to write a Power PC emulator for x86. If they want Classic to be able to function, they'd better rewrite the 68k emulator for x86 as well.
For Apple's sake, I hope they can find someone else to make a Power PC compatible CPU for them (IBM or otherwise). I have doubts as to whether developers would be happy about the switch, and whether customers would be willing to buy another round of upgrades. If it ever came to a switch to x86, Apple would have to be a little generous with the old bank account. They should fund some of the application development for x86, so that developers can afford to give free or discounted upgrades to the x86 version.
[ 08-03-2002: Message edited by: PipelineStall ]</p>
<strong>I havent read the entire thread so please bear with me:
Moki: Wouldn't going to Intel mean that you should rewrite a lot of your software or do the OS "protect" the app completly from the CPU architecture. Writing code especially for altivec suggest the rewrite option but I really don't know
Not really. Off of the top of my head, I can't think of too many places I'd need to change things in most of our products to get them to cross-compile, assuming a full-featured Carbon framework was available for another architecture.
2 things:
-Apple is NEVER going to move to _x86_
-Apple is not moving to another CPU platform for at least another 2 years. Transiting on the OS _AND_ the CPU at the same time is NOT going to happen.
As for the feasibility: nothing is impossible, but not everything that is possible makes sense.
G-News
<strong>-Apple is NEVER going to move to _x86_
G-News</strong><hr></blockquote>
So certain are you...
<strong>
So certain are you...
Moki, you would spend the money on a X86-Apple Box, buy a new version of OS X86, and new apps (assuming that software writers would support two ports for X86)?
I think this is different from the 68k jump. I say this because when that jump was made, you could still run your old apps. I have heard here that a X86 can not emulate a PPC (too many registers or something to that effect). So all legacy apps would be gone!
Plus if the machine was able to duel boot, as a developer, why would you write an app for both WinXP and for OS X? Just tell the person to boot into WinXP to run the app. Or someone would do a little translating to WINE (now we are on X86 boxes) and you could run the WinXP title without having to reboot. We would loose even more developers.
Also, how long before a hack came out to allow OS X86 to boot on a home built PC? Look at the XBox situation. There is already a group trying to get Linux to run on it (and given time, I bet they will succeed). It would be the same thing. Given time, any one would be able to install OS X, and there goes Apple's Hardware sales.
As a consumer, if you saw 2 machines, same speed, why would you buy the more expensive one? The hardcore Apple people would buy Apple (maybe), but I would think that Apple wouldn't gain any market share this way.
Too many of you are thinking with your "I want the fastest machine" hats on. You need to look at it from all standpoints (that of the consumers as well). When you do that, it really doesn't make sense for Apple.
But who really knows. Apple has a plan, and none of us know it, that is what is exciting.
Ben
PS Didn't someone bring this up at the last shareholders meeting, and Steve basicly shot him down? Now if they did go through with this move, couldn't they (Apple) get in some trouble for not telling the shareholders the truth?
[ 08-03-2002: Message edited by: kupan787 ]</p>
<hr></blockquote>
It's not impossible, just more difficult. I mean, in the worst case, you can always store the contents of the additional registers in RAM. Ridiculously slow? Yes. But not impossible.
Heck, with DDR 2, you have a 533 MHz bus theoretically more than 4 times as fast as the current MPX PC133 implementation. It may not be too horrible.
[quote]Plus if the machine was able to duel boot, as a developer, why would you write an app for both WinXP and for OS X? Just tell the person to boot into WinXP to run the app. Or someone would do a little translating to WINE (now we are on X86 boxes) and you could run the WinXP title without having to reboot. We would loose even more developers.
<hr></blockquote>
If the machines were able to run Windows, I agree, there would be little motivation for many developers to write a native Mac OS X version. Which is why Apple probably wouldn't allow XP or any other version of Windows to be installed on these machines out of the box - though you can bet that someone like Connectix would have a PC "emulator" (not really an emulator anymore) that lets you use Windows.
1) Motorola's and IBM's PPC processor development will continue to see performance enhancements at the current rate over the next 1-2 years.
2) That x86 chips from Intel and or AMD will continue to see performance improvements at the current rate over the next 1-2 years.
If both of these conditions are true at the end of the next two years, then yes it is possible that Apple will invest the significant effort to develop chipsets and motherboards that support Mac OS X running on an x86 processor. But right now, that's a long shot.
A lot can happen in two years to make Andrew Neff look very stupid. (Among other things, the HP-Compaq merger could turn out to be a dud.) The PPC architecture could very well leapfrog the Intel set again. And it could even stay out in front -- depending on the Itanium's performance.
While CNet is touting Neff's oracular prowess, let's not forget that he has been wide of the mark more often than not. At the same time he first suggested the HP-Compaq merger in 2001, he also suggested Dell should buy Gateway and IBM's PC division, which is not going to happen. And in March of that year he floated a rumor that Apple was being put on the Auction block by it's board of directors. All in all, I think Neff is less of a prophet and more of a mountebank.
<strong>Any x86 venture by Apple will not include any kind of PPC emulation. Just erase that from your mind as a precondition.</strong><hr></blockquote>
Erasing precondition...
Error, inconsistent logic...
> Add millions of users who won't switch to mac/x86 & Apple going bankrupt...
I'm sorry, Dave. I can't allow that.
[the air is pumped out of the room]
hmmm - why is IBM using PPC chips in it's high performance servers if x86 is superior?????? x86 is ugly old. not even intel itself plans to use x86 chips forever (Itanium?)
<strong>Any x86 venture by Apple will not include any kind of PPC emulation. Just erase that from your mind as a precondition.</strong><hr></blockquote>
Then add new FSB timing, cache architectures and completely rewritten Open Firmware. Motherboard changes across the entire hardware line? Big or little endian I/O is just scratching the surface, here. The way Apple designs systems, this would require a massive, long term effort and would be unlikely to be accomplished in any reasonable time frame unless Apple were already at work at it. Apple's research budget (which would probably double or triple) doesn't reflect such an effort.
It'd take at least four years to do. (And that's if they had help from Intel/AMD.) It's not something Apple would undertake for the fun of it, and it's not anything that will happen anytime soon.
Apple is never going to replace the current PPC platform with x86 (emphasis on x86).
they may introduce a new line that features x86, if they feel like having a super-workstation for their movie stuff or whatever, but even that would be a bad PR move.
Moving to PPC from 68k was so massively different to moving from PPC to x86, because PPC was a whole new architecture with a lot of life and potential left in it. x86 on the other hand is about to slowly dissappear from the market within the net 5 to 10 years.
Moving to x86 now would be the most stupid thing Apple could possibly do. But then again they have proven often enough to be fond of worst case solutions, so I probably shouldn't be so certain.
Say hello to the Crappletosh.
G-News
One thing many people over look is the fact that Dell does little in the way of R&D. They don'd design their own mobos, supporting chips, cases, processors or anything else -- including the operating system.
What Dell and many other PC OEMs do is buy parts and assemble a system, brand it as their own and sell it to the consumer. I think if Michael Dell had an original idea his head would instantly cave in.
Macs cost more, but Why? If people actually took the time to find the answer to that question, they wouldn't be asking it in the first place.
Apple designs the motherboards, the supporting chips, the ROMs, the case, the power supply -- to some extent the processor and of course an operating system that's easily on par with anything farting out of Redmond. Let's see Dell design their own OS and give M$ some competition.. Yeah, my thoughts too... They couldn't do it and embarrass the hell out of themselves.
On another note, where is Dell contributions with respect to Industry standards or even with crossplatform software like QuickTime. Where is their iPod etc. ? The list can go on and on for quite some time.
I can't believe the cost argument still comes into play when comparing Apple to the rest of the PC OEMs. Imagine if auto manufacturers decided to take the same rout as Dell when building and marketing their cars. That is.. Take the R&D costs out of the automaker's equation.
--
Ed M.
As I have said before, Starship Enterprise ain't gonna be run on single 100 GHz processors, everything is going to be multiprocessors passing information around to each other.
Processor speed is definitely a factor, but that is currently a manufacturing problem with IBM/MOT. Bandwidth and multiprocessors will be the key and I hope that is where Apple is headed to.
Forget the MHz race.
<strong>
Erasing precondition...
Error, inconsistent logic...
> Add millions of users who won't switch to mac/x86 & Apple going bankrupt...
I'm sorry, Dave. I can't allow that.
[the air is pumped out of the room]</strong><hr></blockquote>
I never said Apple would "switch" to x86 -- options, remember? Switching from one ISA to another doesn't gain you options.