<strong>If apple is a hardware company then why is the majority of its R&D on software? If they are a hardware company then why do they make the operating system?</strong><hr></blockquote>
And why does Epson still make inkjets when they sell them at a loss and make the money on the ink? And please guess what gets more R&D spending, the inkjets or the ink?
[quote]<strong>If they are just a hardware company they wouldn't be any better than dell. Just another slap some compenents together and lets make another nasty, excuse to check e-mail and play video games PC.</strong><hr></blockquote>
And how would that differentiate them? Could the OS make their hardware more appealing perhaps? What a novel idea.
[quote]<strong>But I bet you that with recent developments in apples advertising, and their overall emphasis on how OS X makes things easy, that they will push software more over than hardware. Look at the recent services such as .mac and having us upgrade to 10.2, its simply because they are finding their software will be their new beer ticket.</strong><hr></blockquote>
Perhaps you should look at that balance sheet. Apple will follow the money and the balance sheet says HARDWARE=$$$. DotMac is unproven and will probably sink. I just don't see the value in it. And Apple just can't live on the upgrades to the MacOS.
ok ok... Lets not be so close-minded in our aspirations. I have a couple of questions for you all. Do you use because of its seamless integration with its hardware? Or because of the Mac OS? Do you use it for ease-of-use? or because of its particular processor its on?
Coming from a marketing background, I'm curious to hear what other people think. Lets think of it as the business.. of the business.
-and-
X86 will never be on Mac, its dead even PC manufacturers are preparing for whats around the corner. When you say the bucks are in hardware they aren't necessarily anymore. Earnings from hardware are at their lowest they've ever been. Profit margins are slim, and resorting to construction overseas for a cheap labor force is because of this.
I personally think that like the art of doing art, that its not what brand of canvas you use, its what you paint on it... that the computer business is similiar. Look at the industry, everyone moving to subscription services, and more contractual business. Retailers now sell warranty plans because of the slim 10-20$ margin they make on sales, but they don't mention what they lose for stocking it for so long. The business is changing, and apple sees this..
Did you even read the half dozen messages above that explain why you have made several invalid assumptions? Moki's proposed direction has nothing to do with clones or Apple licensing the OS. It only has to do with Apple introducing an additional line of machines which happen to use an x86 processor.</strong><hr></blockquote>
What possible good would putting Mac OS X on this old fashion achitecture ( which I bet Intel wishes they could move on from ) do? A waste of time and money if you ask me.
<strong>oops I messed up on that one question I meant to say
"do you use Mac OS because of its seamless integration"...
Curious to know what, and why you use Mac OS.
Why should it be on X86?</strong><hr></blockquote>
I think a lot of people want Mac OS X on x86 purely because 2.2 is a bigger number than 1.0. In reality, the performance gap is less than it seems, but we're feeling the pinch now so everybody thinks other processors must be the answer. A much better answer is faster PowerPCs.
I agree Motorola needs a kick up the arse, and I think they are getting it from Jobs. First he mentions Apple likes CPU options, then a few weeks later IBM announces a seemingly ideal processor just after the AIM alliance ends.
<strong>ok ok... Lets not be so close-minded in our aspirations. I have a couple of questions for you all. Do you use because of its seamless integration with its hardware? Or because of the Mac OS? Do you use it for ease-of-use? or because of its particular processor its on?</strong><hr></blockquote>
All of the above. IMO it's simply the best overall solution, and far superior to the alternatives in its integration.
You did miss one other potential reason, tho', and it's another one of mine, actually.
Using a Mac means that I don't have to have any dealings with M$ - ever - unless I choose to do so.
Their behavior has been sometimes questionable, sometimes reprehensible, but almost always unethical (even when barely legal), and this DOES affect my purchasing decision to some degree. It also follows that I do care what proccessor I use because of this. Intel simply has not behaved in any more ethical a manner over the years, and therefore I have no desire to deal with them either.
What possible good would putting Mac OS X on this old fashion achitecture ( which I bet Intel wishes they could move on from ) do? A waste of time and money if you ask me.</strong><hr></blockquote>
People who have discounted the enourmous power that the entrenchment of the x86 architecture has (including Intel) have historically been misguided to do so. Yes, the x86 instruction set is ridiculous, and Intel thought they'd be switching to a new ISA years ago, but the market didn't want a new chip -- they wanted incremental improvements, and they got it.
Especially with AMD's new 64 bit extensions to the x86 architecture, I think it is safe to say we'll be seeing it for longer than most people expect at this point. Getting the market to switch to the Itanium is going to be a monumental task for Intel -- not impossible, but it'll take an awful lot of arm twisting to make it happen.
Again, I don't see Apple *switching* to x86, I'd imagine they might make custom x86 boxes so that their xplat OS is finally utilizing its xplat abilities. Why? Because some companies simply will not purchase *anything* that isn't an x86-based computer -- this gives them a foot in the door where otherwise they'd have no point of entry.
Developed in parallel with a killer PPC line, and it can be seen as merely a way to expand the market penetration of the Mac. Most of the parts needed to make such a beast can just be picked off of the shelf -- the major R&D for Apple would be in software, not hardware.
It seemed a little wonky to me too, when I first digested it -- and maybe it'll never see the light of day -- but it is beginning to make more sense to me now. If Apple truly wants people to "switch", they need to break down as many of the barriers as they can.
Also, we all know that Apple has been at the mercy of Motorola for its processors for some time -- any responsible businessman would see this single point of failure, and do everything in their power to eliminate it.
<strong>Again, I don't see Apple *switching* to x86, I'd imagine they might make custom x86 boxes so that their xplat OS is finally utilizing its xplat abilities. Why? Because some companies simply will not purchase *anything* that isn't an x86-based computer -- this gives them a foot in the door where otherwise they'd have no point of entry.</strong><hr></blockquote>
Where this argument falls down for me is that a custom x86 box from Apple would not be an x86 ISA clone, that is, it wouldn't run Windows so why would any "x86 only" shop want to buy it anymore than the current Mac?
[ 08-08-2002: Message edited by: Tomb of the Unknown ]</p>
If apple made an x86 machine and sold it for the same margin it makes on present machines, wouldn't the x86 machine be even more expensive than the present towers? The p4 costs more than the g4, right? Where would any cost savings come from? Are they just going to use a standard x86 motherboard?
You think the "mhz gap" is bad, try the "price gap" on for size. What wintel user is going to switch when they see the exact same specs on the $2000 apple machine [1] as on their $1000 machine [2]?
Having a completely different architecture helps apple by shielding them a little from uneducated comparisions with the wintel world (like I made in the last paragraph). In other words: Apple computers cost more because they are better engineered. That's believable because the apple really is different on the inside. But if an Apple is just a p4 on an nForce motherboard (running OS X), how are you going to explain the huge markup in price? OS X only costs $129!
The different architecture lets them maintain big margins (compared to wintel clone makers) because there aren't any other ppc desktop machines to compare with them. Sure the mhz gap probably hurts sales a little, but I think the "mhz myth" response actually works on people. Apple machines are better (I think,) and worth the price. A little CISC/RISC mumbo jumbo can smooth over the mhz gap problem. But it's going to be hard to convice people that a stock wintel machine is worth a grand (or more) more just because it has an Apple ROM that will let it boot OS X.
Moki's dual lines maybe makes a little bit of sense, but I still don't see it.
<strong>Again, I don't see Apple *switching* to x86, I'd imagine they might make custom x86 boxes so that their xplat OS is finally utilizing its xplat abilities. Why? Because some companies simply will not purchase *anything* that isn't an x86-based computer -- this gives them a foot in the door where otherwise they'd have no point of entry.</strong><hr></blockquote>
I was going to say that you'd offered a flimsy reason for Apple to adopt x86 as an option, and then I remembered that yes, corporate IT can be that silly, and often is.
Let's say that an x86 (realistically, x86-64) based Mac is going to happen, for argument's sake. This is what I'm curious about: There are a lot of aspects of PC hardware that have hung on just as tenaciously as the x86 ISA (BIOS, IRQs, etc.) that Apple can discard because they're not going to make PCs. They'll want there to be more similarities than differences in their PPC and x86 mobos, for a large number of reasons. But the AMD and Mot and IBM CPUs use three different busses to talk to the rest of the board, and one of the CPUs is little-endian. How to get around this?
I don't know enough about hardware to state this definitively, but it seems like if Apple had the CPU module plug into a HT fabric they'd moot the bus issue. Endianness could be handled either by a bit of hardware judo, or by flipping the PPCs over to run little-endian, which they can do now, easily. :eek: (I hope this doesn't happen, because big-endian makes so much more sense and it's compatible with the big UNIXen, but I have to admit it's a tempting option. Damn Intel. )
The other variable, I suppose, is whether the CPU has an on-die memory controller - so far, the Mot G4s don't. If not, Apple would have to compensate for that (or the possibility that Mot goes to an on-die memory controller - that expects to hook up to a RapidIO bus) as well.
Am I on the right track here? Or, to mix metaphors, am I even in the right ballpark?
<strong>ok ok... Lets not be so close-minded in our aspirations. I have a couple of questions for you all. Do you use because of its seamless integration with its hardware? Or because of the Mac OS? Do you use it for ease-of-use? or because of its particular processor its on?</strong><hr></blockquote>
All of those but the last. But of course, a move to x86 will harm the ease-of-use since I cannot run my old software anymore. Or it will divide the Mac-userbase, creating a lot of trouble for me as well.
[quote]<strong>X86 will never be on Mac, its dead even PC manufacturers are preparing for whats around the corner.</strong><hr></blockquote>
x86-64 will probably win on the desktop, that's basically an extension to x86. I've seen no proof that a transition to x86-64 will somehow be easier.
[quote]<strong>When you say the bucks are in hardware they aren't necessarily anymore. Earnings from hardware are at their lowest they've ever been. Profit margins are slim, and resorting to construction overseas for a cheap labor force is because of this.</strong><hr></blockquote>
That has to do with the delays in fast new PowerMacs and OS X being a bit/awfully slow. There is no reason why we can't have a fast, fairly competitive PowerMac soon, except for Apple ****ing up of course. OS X is another story, but that has little to do with the hardware of course (except that a switch will take engineering resources that can't go into improving OS X).
[quote]<strong>I personally think that like the art of doing art, that its not what brand of canvas you use, its what you paint on it... that the computer business is similiar. Look at the industry, everyone moving to subscription services, and more contractual business. Retailers now sell warranty plans because of the slim 10-20$ margin they make on sales, but they don't mention what they lose for stocking it for so long. The business is changing, and apple sees this..</strong><hr></blockquote>
I've been hearing that services shit for over five years now. Customers like to buy. They want to own things. They will only pay subscription for things that make sense to them, like cable, e-mail or stuff like that. And even for many services they try very hard to avoid paying and flock to free services. There won't be a revolution with everything moving to services. It just doesn't make sense to be dependent on MS or Apple so you can use an OS or a word processor. Customers have resisted those things in the past and they will do so in the future.
Trust me, subscriptions will be limited to things that you already do online and that are worth the money (which is probably not that much for non-business users). The money will go to content-providers and developers, not to companies that create OS's. Apple can perhaps earn a bit of the money with their iApps, but it won't be hundreds of millions. Not unless they come up with some very, very clever stuff. I don't think they are that smart
<strong>Where this argument falls down for me is that a custom x86 box from Apple would not be an x86 ISA clone, that is, it wouldn't run Windows so why would any "x86 only" shop want to buy it anymore than the current Mac?</strong><hr></blockquote>
I agree. Few shops athat refuse to buy Macs are into x86, they standardize on Windows. Personally I believe that these companies have their head so far up their behind that they will always just buy the cheapest Dell/Compaq or whatever. They won't even consider something that doesn't run what 'everybody uses'.
its funny how you mention the cable service analogy. Look at how the internet started and its trend to national broadcasting. How broadcasting would be free and everyone would just have to buy a tv set to be able to watch somethen. I agree tho that customers don't want to have a subscription service. But then again look at how people at first refuses cable television, and look at it now?
But lets talk about the near future... WE NEED NEW POWERMACS I do feel though that if apple could supplement their income if they used Shake and such to their advantage and made POWERFUL software to use it. I feel we will see a big split now between consumer and pro hardware, but... it will be awesome.
Well, I'm sure it'll help someone -- I've been doing AltiVec code for some time now, and I absolutely love it. The number of apps that use it are still miniscule, though, and honestly, Apple doesn't push the technology to developers all that much.</strong><hr></blockquote>
I'd like to just slightly ammend that to 'The number of _shrinkwrapped_ apps....' though.
Science/engineering is replete with custom code & people willing to optimize the snot out of things. AV a very useful tool here. Counting 'apps' is a little tougher when the 'app' is a very small chunk of FFT code or whatever that's mixed with 47 other little blocks of code to get to a result.
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.
</strong><hr></blockquote>
(replying to a month old post)
i don't know why run time emulators are so popular. except for whole machine (eg, pc emulators like bochs) emulation, it's stupid to emulate. the smart thing to do is translate on the fly, from ppc to x86, and cache the resulting x86 binary. it could be tricky, caching just parts of the original binary that have been run. a better way would be, upon installation, or first run, translate the binar{y,ies} completely and cache the resulting native binary. when it's subsequently run, it will run at full speed.
and, regarding runtime emulation/interpretation, it can be FASTER than native execution. lookup hp's dynamo(sp). if that's the way they want to go.
as far as technical hurtles for a switch to x86, there are tons of x86 emulators. the fact that x86 is register starved (ie, has less regs than ppc) is not a problem. it just requires more overhead to schedule reg usage and (of course) use the stack a lot (which x86 does all the time, by nature, already).
Without Apple's consent, all they can do is make a clone PC that can run OS X, and then tell people to go out and buy OS X for x86... oops, Apple won't be offering that as a retail package. Hmm... now what?
However, let's say that some PC clones do manage to run OS X somehow... sure would be a nice gradual way for Apple to transition away from making their income mostly from hardware...</strong><hr></blockquote>
Who would think about buying an x86-based OSX computer? Two groups mainly: Those that want OSX on a fast computer, and those that want an x86-based computer and are considering switching to OSX instead of Windows.
The first group mainly consists of people that already use Macs; selling x86-based Macs is merely cannibalizing the G3/4/5 market.
The second group want to buy an x86-based computer because they want ISA(?)/Windows-compatibility. But an Apple x86-based computer will NOT be Windows-compatible (for obvious reasons); similarly, if the hardware differences between Apple's machine and the ISA-standard are enough to keep OSX from being able to run on generic PCs, MS would need to modify Windows to enable it to run on Apple's machine.... and THAT...
Some IT-dept. leaders may insist on buying x86 CPU-based computers only, for the sake of ease of repairs. Again, the necessarily unique features in Apple x86-computers mean that they will not be considered.
The uniqueness necessary in such machines rules out any savings from scale-of-production (they have to be produced in completely separate runs), and so x86-based Apple computers will not be able to compete on price, even if Apple chooses to lower its desired margin of 25%.
I fail to see any valid reasons for introducing a parallel line of Apple x86 computers. Could anyone help me out?
Comments
<strong>If apple is a hardware company then why is the majority of its R&D on software? If they are a hardware company then why do they make the operating system?</strong><hr></blockquote>
And why does Epson still make inkjets when they sell them at a loss and make the money on the ink? And please guess what gets more R&D spending, the inkjets or the ink?
[quote]<strong>If they are just a hardware company they wouldn't be any better than dell. Just another slap some compenents together and lets make another nasty, excuse to check e-mail and play video games PC.</strong><hr></blockquote>
And how would that differentiate them? Could the OS make their hardware more appealing perhaps? What a novel idea.
[quote]<strong>But I bet you that with recent developments in apples advertising, and their overall emphasis on how OS X makes things easy, that they will push software more over than hardware. Look at the recent services such as .mac and having us upgrade to 10.2, its simply because they are finding their software will be their new beer ticket.</strong><hr></blockquote>
Perhaps you should look at that balance sheet. Apple will follow the money and the balance sheet says HARDWARE=$$$. DotMac is unproven and will probably sink. I just don't see the value in it. And Apple just can't live on the upgrades to the MacOS.
Coming from a marketing background, I'm curious to hear what other people think. Lets think of it as the business.. of the business.
-and-
X86 will never be on Mac, its dead even PC manufacturers are preparing for whats around the corner. When you say the bucks are in hardware they aren't necessarily anymore. Earnings from hardware are at their lowest they've ever been. Profit margins are slim, and resorting to construction overseas for a cheap labor force is because of this.
I personally think that like the art of doing art, that its not what brand of canvas you use, its what you paint on it... that the computer business is similiar. Look at the industry, everyone moving to subscription services, and more contractual business. Retailers now sell warranty plans because of the slim 10-20$ margin they make on sales, but they don't mention what they lose for stocking it for so long. The business is changing, and apple sees this..
"do you use Mac OS because of its seamless integration"...
Curious to know what, and why you use Mac OS.
Why should it be on X86?
<strong>
Did you even read the half dozen messages above that explain why you have made several invalid assumptions? Moki's proposed direction has nothing to do with clones or Apple licensing the OS. It only has to do with Apple introducing an additional line of machines which happen to use an x86 processor.</strong><hr></blockquote>
What possible good would putting Mac OS X on this old fashion achitecture ( which I bet Intel wishes they could move on from ) do? A waste of time and money if you ask me.
<strong>oops I messed up on that one question I meant to say
"do you use Mac OS because of its seamless integration"...
Curious to know what, and why you use Mac OS.
Why should it be on X86?</strong><hr></blockquote>
I think a lot of people want Mac OS X on x86 purely because 2.2 is a bigger number than 1.0. In reality, the performance gap is less than it seems, but we're feeling the pinch now so everybody thinks other processors must be the answer. A much better answer is faster PowerPCs.
I agree Motorola needs a kick up the arse, and I think they are getting it from Jobs. First he mentions Apple likes CPU options, then a few weeks later IBM announces a seemingly ideal processor just after the AIM alliance ends.
Wheels within wheels...
Altivec is there, and much more
<strong>ok ok... Lets not be so close-minded in our aspirations. I have a couple of questions for you all. Do you use because of its seamless integration with its hardware? Or because of the Mac OS? Do you use it for ease-of-use? or because of its particular processor its on?</strong><hr></blockquote>
All of the above. IMO it's simply the best overall solution, and far superior to the alternatives in its integration.
You did miss one other potential reason, tho', and it's another one of mine, actually.
Using a Mac means that I don't have to have any dealings with M$ - ever - unless I choose to do so.
Their behavior has been sometimes questionable, sometimes reprehensible, but almost always unethical (even when barely legal), and this DOES affect my purchasing decision to some degree. It also follows that I do care what proccessor I use because of this. Intel simply has not behaved in any more ethical a manner over the years, and therefore I have no desire to deal with them either.
<strong>
What possible good would putting Mac OS X on this old fashion achitecture ( which I bet Intel wishes they could move on from ) do? A waste of time and money if you ask me.</strong><hr></blockquote>
People who have discounted the enourmous power that the entrenchment of the x86 architecture has (including Intel) have historically been misguided to do so. Yes, the x86 instruction set is ridiculous, and Intel thought they'd be switching to a new ISA years ago, but the market didn't want a new chip -- they wanted incremental improvements, and they got it.
Especially with AMD's new 64 bit extensions to the x86 architecture, I think it is safe to say we'll be seeing it for longer than most people expect at this point. Getting the market to switch to the Itanium is going to be a monumental task for Intel -- not impossible, but it'll take an awful lot of arm twisting to make it happen.
Again, I don't see Apple *switching* to x86, I'd imagine they might make custom x86 boxes so that their xplat OS is finally utilizing its xplat abilities. Why? Because some companies simply will not purchase *anything* that isn't an x86-based computer -- this gives them a foot in the door where otherwise they'd have no point of entry.
Developed in parallel with a killer PPC line, and it can be seen as merely a way to expand the market penetration of the Mac. Most of the parts needed to make such a beast can just be picked off of the shelf -- the major R&D for Apple would be in software, not hardware.
It seemed a little wonky to me too, when I first digested it -- and maybe it'll never see the light of day -- but it is beginning to make more sense to me now. If Apple truly wants people to "switch", they need to break down as many of the barriers as they can.
Also, we all know that Apple has been at the mercy of Motorola for its processors for some time -- any responsible businessman would see this single point of failure, and do everything in their power to eliminate it.
<strong>Again, I don't see Apple *switching* to x86, I'd imagine they might make custom x86 boxes so that their xplat OS is finally utilizing its xplat abilities. Why? Because some companies simply will not purchase *anything* that isn't an x86-based computer -- this gives them a foot in the door where otherwise they'd have no point of entry.</strong><hr></blockquote>
Where this argument falls down for me is that a custom x86 box from Apple would not be an x86 ISA clone, that is, it wouldn't run Windows so why would any "x86 only" shop want to buy it anymore than the current Mac?
[ 08-08-2002: Message edited by: Tomb of the Unknown ]</p>
You think the "mhz gap" is bad, try the "price gap" on for size. What wintel user is going to switch when they see the exact same specs on the $2000 apple machine [1] as on their $1000 machine [2]?
Having a completely different architecture helps apple by shielding them a little from uneducated comparisions with the wintel world (like I made in the last paragraph). In other words: Apple computers cost more because they are better engineered. That's believable because the apple really is different on the inside. But if an Apple is just a p4 on an nForce motherboard (running OS X), how are you going to explain the huge markup in price? OS X only costs $129!
The different architecture lets them maintain big margins (compared to wintel clone makers) because there aren't any other ppc desktop machines to compare with them. Sure the mhz gap probably hurts sales a little, but I think the "mhz myth" response actually works on people. Apple machines are better (I think,) and worth the price. A little CISC/RISC mumbo jumbo can smooth over the mhz gap problem. But it's going to be hard to convice people that a stock wintel machine is worth a grand (or more) more just because it has an Apple ROM that will let it boot OS X.
Moki's dual lines maybe makes a little bit of sense, but I still don't see it.
The g4 is a good thing!
And I haven't been wrong (or right) yet.
[1] $2099 - Powermac 1 ghz g4 256meg/60gig/cdrw
[2] $948 - Dell Dimension 8200 2 ghz p4 256meg/80gig/cdrw
<strong>Again, I don't see Apple *switching* to x86, I'd imagine they might make custom x86 boxes so that their xplat OS is finally utilizing its xplat abilities. Why? Because some companies simply will not purchase *anything* that isn't an x86-based computer -- this gives them a foot in the door where otherwise they'd have no point of entry.</strong><hr></blockquote>
I was going to say that you'd offered a flimsy reason for Apple to adopt x86 as an option, and then I remembered that yes, corporate IT can be that silly, and often is.
Let's say that an x86 (realistically, x86-64) based Mac is going to happen, for argument's sake. This is what I'm curious about: There are a lot of aspects of PC hardware that have hung on just as tenaciously as the x86 ISA (BIOS, IRQs, etc.) that Apple can discard because they're not going to make PCs. They'll want there to be more similarities than differences in their PPC and x86 mobos, for a large number of reasons. But the AMD and Mot and IBM CPUs use three different busses to talk to the rest of the board, and one of the CPUs is little-endian. How to get around this?
I don't know enough about hardware to state this definitively, but it seems like if Apple had the CPU module plug into a HT fabric they'd moot the bus issue. Endianness could be handled either by a bit of hardware judo, or by flipping the PPCs over to run little-endian, which they can do now, easily. :eek: (I hope this doesn't happen, because big-endian makes so much more sense and it's compatible with the big UNIXen, but I have to admit it's a tempting option. Damn Intel.
The other variable, I suppose, is whether the CPU has an on-die memory controller - so far, the Mot G4s don't. If not, Apple would have to compensate for that (or the possibility that Mot goes to an on-die memory controller - that expects to hook up to a RapidIO bus) as well.
Am I on the right track here? Or, to mix metaphors, am I even in the right ballpark?
<strong>ok ok... Lets not be so close-minded in our aspirations. I have a couple of questions for you all. Do you use because of its seamless integration with its hardware? Or because of the Mac OS? Do you use it for ease-of-use? or because of its particular processor its on?</strong><hr></blockquote>
All of those but the last. But of course, a move to x86 will harm the ease-of-use since I cannot run my old software anymore. Or it will divide the Mac-userbase, creating a lot of trouble for me as well.
[quote]<strong>X86 will never be on Mac, its dead even PC manufacturers are preparing for whats around the corner.</strong><hr></blockquote>
x86-64 will probably win on the desktop, that's basically an extension to x86. I've seen no proof that a transition to x86-64 will somehow be easier.
[quote]<strong>When you say the bucks are in hardware they aren't necessarily anymore. Earnings from hardware are at their lowest they've ever been. Profit margins are slim, and resorting to construction overseas for a cheap labor force is because of this.</strong><hr></blockquote>
That has to do with the delays in fast new PowerMacs and OS X being a bit/awfully slow. There is no reason why we can't have a fast, fairly competitive PowerMac soon, except for Apple ****ing up of course. OS X is another story, but that has little to do with the hardware of course (except that a switch will take engineering resources that can't go into improving OS X).
[quote]<strong>I personally think that like the art of doing art, that its not what brand of canvas you use, its what you paint on it... that the computer business is similiar. Look at the industry, everyone moving to subscription services, and more contractual business. Retailers now sell warranty plans because of the slim 10-20$ margin they make on sales, but they don't mention what they lose for stocking it for so long. The business is changing, and apple sees this..</strong><hr></blockquote>
I've been hearing that services shit for over five years now. Customers like to buy. They want to own things. They will only pay subscription for things that make sense to them, like cable, e-mail or stuff like that. And even for many services they try very hard to avoid paying and flock to free services. There won't be a revolution with everything moving to services. It just doesn't make sense to be dependent on MS or Apple so you can use an OS or a word processor. Customers have resisted those things in the past and they will do so in the future.
Trust me, subscriptions will be limited to things that you already do online and that are worth the money (which is probably not that much for non-business users). The money will go to content-providers and developers, not to companies that create OS's. Apple can perhaps earn a bit of the money with their iApps, but it won't be hundreds of millions. Not unless they come up with some very, very clever stuff. I don't think they are that smart
<strong>Where this argument falls down for me is that a custom x86 box from Apple would not be an x86 ISA clone, that is, it wouldn't run Windows so why would any "x86 only" shop want to buy it anymore than the current Mac?</strong><hr></blockquote>
I agree. Few shops athat refuse to buy Macs are into x86, they standardize on Windows. Personally I believe that these companies have their head so far up their behind that they will always just buy the cheapest Dell/Compaq or whatever. They won't even consider something that doesn't run what 'everybody uses'.
But lets talk about the near future... WE NEED NEW POWERMACS
<strong>
Well, I'm sure it'll help someone -- I've been doing AltiVec code for some time now, and I absolutely love it. The number of apps that use it are still miniscule, though, and honestly, Apple doesn't push the technology to developers all that much.</strong><hr></blockquote>
I'd like to just slightly ammend that to 'The number of _shrinkwrapped_ apps....' though.
Science/engineering is replete with custom code & people willing to optimize the snot out of things. AV a very useful tool here. Counting 'apps' is a little tougher when the 'app' is a very small chunk of FFT code or whatever that's mixed with 47 other little blocks of code to get to a result.
<a href="http://www.forbes.com/technology/2002/08/09/0809apple.html?partner=newscom"; target="_blank">From Forbes</a>
<strong>
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.
</strong><hr></blockquote>
(replying to a month old post)
i don't know why run time emulators are so popular. except for whole machine (eg, pc emulators like bochs) emulation, it's stupid to emulate. the smart thing to do is translate on the fly, from ppc to x86, and cache the resulting x86 binary. it could be tricky, caching just parts of the original binary that have been run. a better way would be, upon installation, or first run, translate the binar{y,ies} completely and cache the resulting native binary. when it's subsequently run, it will run at full speed.
and, regarding runtime emulation/interpretation, it can be FASTER than native execution. lookup hp's dynamo(sp). if that's the way they want to go.
as far as technical hurtles for a switch to x86, there are tons of x86 emulators. the fact that x86 is register starved (ie, has less regs than ppc) is not a problem. it just requires more overhead to schedule reg usage and (of course) use the stack a lot (which x86 does all the time, by nature, already).
<strong>
Without Apple's consent, all they can do is make a clone PC that can run OS X, and then tell people to go out and buy OS X for x86... oops, Apple won't be offering that as a retail package. Hmm... now what?
However, let's say that some PC clones do manage to run OS X somehow... sure would be a nice gradual way for Apple to transition away from making their income mostly from hardware...</strong><hr></blockquote>
Who would think about buying an x86-based OSX computer? Two groups mainly: Those that want OSX on a fast computer, and those that want an x86-based computer and are considering switching to OSX instead of Windows.
The first group mainly consists of people that already use Macs; selling x86-based Macs is merely cannibalizing the G3/4/5 market.
The second group want to buy an x86-based computer because they want ISA(?)/Windows-compatibility. But an Apple x86-based computer will NOT be Windows-compatible (for obvious reasons); similarly, if the hardware differences between Apple's machine and the ISA-standard are enough to keep OSX from being able to run on generic PCs, MS would need to modify Windows to enable it to run on Apple's machine.... and THAT...
Some IT-dept. leaders may insist on buying x86 CPU-based computers only, for the sake of ease of repairs. Again, the necessarily unique features in Apple x86-computers mean that they will not be considered.
The uniqueness necessary in such machines rules out any savings from scale-of-production (they have to be produced in completely separate runs), and so x86-based Apple computers will not be able to compete on price, even if Apple chooses to lower its desired margin of 25%.
I fail to see any valid reasons for introducing a parallel line of Apple x86 computers. Could anyone help me out?
engpjp
<strong>
Could anyone help me out?
engpjp</strong><hr></blockquote>
No, all the arguments pro & con have already been made. Just read the past posts in the numerous threads to help you formulate an opinion.