Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
OS X on x86 Prototype Boxes
- Thread starter MacRumors
- Start date
- Sort by reaction score
I agree 100%.
Like it is said at MacEdition, it is purely an RnD effort to do test to make sure to have a back up plan so if moto just goes cold turkey on semiconductors, espeically since they are selling that devision anyway....
Who knows what happens.
I think apple is just using that as a plan probably to make a buffer between now and when they have full ppc970 machines shipping in case, like I said, if Moto drops the bucket.
We'll have to wait and see.
But like arn said, for apple to just allow you to download a full copy of MOSX for your x86? they would kill themselves. Just like the clone days when no one wanted to buy their cash cows, ie their hardware.
Maybe We're missing the point......
Perhaps We're missing another possiblility..... The machines wouldn't necessarily by X86 compatible. Maybe we'd be looking at Apple proprietary branded boxes, but with an Intel or AMD processor.
Also, this may be a speed and not a price issue.
Jim
Perhaps We're missing another possiblility..... The machines wouldn't necessarily by X86 compatible. Maybe we'd be looking at Apple proprietary branded boxes, but with an Intel or AMD processor.
Also, this may be a speed and not a price issue.
Jim
There's another reason why Apple won't go with AMD, and especially not with IA-64: AMD's future doesn't look so bright. For a while, they were beating Intel handily in the price/performance game, but lately, that's not so much the case. AMD is losing money; Intel is making it.
Why would Apple want to become dependent on yet another flaky supplier? If they do go X86, they might buy chips from AMD, but they'll be damned sure to stick to X86-32 and stay compatible with Intel's chips.
X86-32 might be an ugly architecture when measured by some theoretical computer science yardstick, but so what? In the real world, X86 chips cream everything else that's available at a price point that will make for an affordable personal computer.
My personal hope is that the 970 hits the street quickly and closes much or all of the performance gap. I can't see any scenario in which a shift to X86 (even if only in proprietary Apple boxes) is anything but terrible news for Apple, a last-ditch desperation move. The anger it's likely to inspire in developers and in many users will take a long time to heal. Even if they make the transition smooth as possible, it will make a mockery of their marketing efforts of the past several years.
Here's one other thought as to why Apple maintains a port of OS/X to X86: It's their doomsday device. If Apple's situation becomes obviously hopeless (this would be some time down the road, given the cash they have in the bank), they can open-source the whole operating system (not just Darwin). Suddenly, there's a good open source GUI available for unix to replace the ugliness that is X11, and open source unix may actually start to look like a viable competitor on the desktop. This would probably be fairly bad news for MS. Sure, at that point Apple would probably be dead, but by letting MS know that they will suffer significantly upon Apple's death, Apple has some leverage to ensure that MS continues to provide a decent level of support for the Mac platform. (e.g. They won't stop selling Office for the Mac.)
Not necessarily. I'd be willing to bet that Adobe did not write very much ASM- this is what Apple has been trying to get companies (and themselves) to NOT do in order to have a recompile be simple. The reason that any port between OS's (Linux, OS X, Windows), and even games between consoles and computers, takes a long time is due to the conversion of the underlying calls to the OS. Try this: Install Linux on a PC and Linux on a Mac. Install KDE on both. Then install KDE programs. The same base code will work on both machines- programmers (unless writing ASM, which does not happen as often as you'd think it does) write for the OS, not the processor. That means they're writing C/C++ code for the KDE/Windows/OS X windowing system. If Apple were to port OS X to an Intel, all that you would need is to recompile it and change any ASM that was in X (which has been kept to a minimum). Then just recompile PS7 for an X86- the compiler, not the programmer, deals with the changes in stuff like Endiness, SSE/Altivec extensions, etc. I guarantee you, if you look at software that is cross platform the only differences you'll find are the OS calls, not the main code. When the OS stays the same, the OS functions are the same- it's up to the OS to account for differences in how a video card is addressed.
A thousand pardons to Leathal here - I missed your name on your prior post. But my comments are still aimed at the other people here who are fixated on Intel == Windows == PC. The x86 chip in the box doesn't mean Windows will run on it.Originally posted by LethalWolfe
Um, it was a rhetorical question. Go back to my first post on the first page. To save you the trip, I said that if Apple went x86 they would keep it a closed system, just like it is now, except it would run a x86 chip and not PPC.
Yer preachin' to the choir eric_n_dfw. Now, what were you sayin' 'bout readin' the entire thread?
Lethal
Actually, I take that back. There is another thread w/this topic that I posted in and and basically said the same thing you did. Cheers.
In response to another comment that someone said about x86 being technoligically about to hit a wall, I used to think the same thing. http://www.mackido.com was stating it years ago. But Intel and AMD seem to keep being able to stretch more and more life out of that dog. Apple cannot (and, most likely IS not) burry it's head in the sand hoping that the x86 will finally run out of gas and Mototolla (or IBM) will speed past them due to the better scalability of their PowerPC's.
Never gonna happen
Apple will never switch to x86 based processors when there is life in their IBM 970's and the new G4's. New macs aren't THAT much slower than PC's.
Apple is much better off siding with the linux/unix geeks at IBM and using the 970's architecture to change their mobo's. Why change the mobo's, and require a recompile to run ANY app on a x86 chip? At least with the 970, it will run any existing PowerPC based app. Sure, unless it's recompiled it won't run as fast or efficiently as it can, but it'd RUN. Apple is in too precarious a situation with developers right now to force them to change their apps a second time. I know some people may say that they could run hardware emulation of the PowerPC on an Athlon, but that ain't viable-- they're already losing the speed war and that won't help.
They also don't want to lose face with their Velocity Engine marketing. The 970 has an AltiVec compatible processing unit-- which Apple smartly marketed and abstracted to be the "Velocity Engine." So no marketing changes are required. They also want to maintain the MHz myth marketing, too. At 1.8 GHz the 970 will scream. Keep in mind that by that time the Wintel world will be dealing with MHz woes-- their new chips will only run at 1 GHz.
When the 970 goes into the pro line, the faster G4's will go in the iMac, iBook, and TiBook. It will all come together nicely and Apple will have a 2 processor family of machines again (G4 and G5). I know some state that the 970 rumors are just rumors right now. Well, yes, perhaps. But I'd bet the farm it's a certainty. In fact, I'll be much more shocked if anything else but a 970 based Mac occured by mid/late 2003.
If Apple ever had to resort to their 11th hour, x86 option, it will only work on proprietary Apple enabled hardware that only Apple sells. People need to get it out of their head that you will be able to buy a crappy $600 machine and load OS X on it. That will never happen. Apple will just switch their mobo's, add some proprietary firmware, and slap dual Athlons in a Quicksilver case. Apple's main selling point is ease of use-- they would lose ALL differentiation if the comsumer had to deal with mismatched 3rd party core components and low-end, poorly constructed hardware.
Lastly, but MOST IMPORTANTLY, is OS X. OS X is the future on the desktop. While there are speed issues with existing apps and existing builds of OS X, things will get better soon. Chimera on a DP machine screams compared to IE or Mozilla. A lot of the perceived slowdown is due to software not being written for the core competancies of OS X. Additionally, the basic features of OS X that redraw the screen and resize windows are currenly undergoing optimization. What does all of this mean?
The 970 will come out about the same time as OS 10.3.5 or 10.4... which will have marked speed improvement in core areas. It will be a speed-mature build. The 970 will also have raw performace well in excess of 5x the current capabilities. Couple these two designs together and Apple will have completed their software and hardware design paradigm.
Apple will never switch to x86 based processors when there is life in their IBM 970's and the new G4's. New macs aren't THAT much slower than PC's.
Apple is much better off siding with the linux/unix geeks at IBM and using the 970's architecture to change their mobo's. Why change the mobo's, and require a recompile to run ANY app on a x86 chip? At least with the 970, it will run any existing PowerPC based app. Sure, unless it's recompiled it won't run as fast or efficiently as it can, but it'd RUN. Apple is in too precarious a situation with developers right now to force them to change their apps a second time. I know some people may say that they could run hardware emulation of the PowerPC on an Athlon, but that ain't viable-- they're already losing the speed war and that won't help.
They also don't want to lose face with their Velocity Engine marketing. The 970 has an AltiVec compatible processing unit-- which Apple smartly marketed and abstracted to be the "Velocity Engine." So no marketing changes are required. They also want to maintain the MHz myth marketing, too. At 1.8 GHz the 970 will scream. Keep in mind that by that time the Wintel world will be dealing with MHz woes-- their new chips will only run at 1 GHz.
When the 970 goes into the pro line, the faster G4's will go in the iMac, iBook, and TiBook. It will all come together nicely and Apple will have a 2 processor family of machines again (G4 and G5). I know some state that the 970 rumors are just rumors right now. Well, yes, perhaps. But I'd bet the farm it's a certainty. In fact, I'll be much more shocked if anything else but a 970 based Mac occured by mid/late 2003.
If Apple ever had to resort to their 11th hour, x86 option, it will only work on proprietary Apple enabled hardware that only Apple sells. People need to get it out of their head that you will be able to buy a crappy $600 machine and load OS X on it. That will never happen. Apple will just switch their mobo's, add some proprietary firmware, and slap dual Athlons in a Quicksilver case. Apple's main selling point is ease of use-- they would lose ALL differentiation if the comsumer had to deal with mismatched 3rd party core components and low-end, poorly constructed hardware.
Lastly, but MOST IMPORTANTLY, is OS X. OS X is the future on the desktop. While there are speed issues with existing apps and existing builds of OS X, things will get better soon. Chimera on a DP machine screams compared to IE or Mozilla. A lot of the perceived slowdown is due to software not being written for the core competancies of OS X. Additionally, the basic features of OS X that redraw the screen and resize windows are currenly undergoing optimization. What does all of this mean?
The 970 will come out about the same time as OS 10.3.5 or 10.4... which will have marked speed improvement in core areas. It will be a speed-mature build. The 970 will also have raw performace well in excess of 5x the current capabilities. Couple these two designs together and Apple will have completed their software and hardware design paradigm.
Re: Parallel Universe?
An Apple server based on OSX and x86 processors exists and is practical now. The main benefit is it can run vertical market apps with no recompiling.
No reason why Apple could not make such a beast that cannot run other OS's at all or that only runs OSX if an Apple custom chip is present.
If the servers with G4/osx are any indication of future trends an x86 server could take significant market share in markets with an x86 bias based on practical reasons not religious zealotry.
Rocketman
I doubt Steve would do it though.
An Apple server based on OSX and x86 processors exists and is practical now. The main benefit is it can run vertical market apps with no recompiling.
No reason why Apple could not make such a beast that cannot run other OS's at all or that only runs OSX if an Apple custom chip is present.
If the servers with G4/osx are any indication of future trends an x86 server could take significant market share in markets with an x86 bias based on practical reasons not religious zealotry.
Rocketman
I doubt Steve would do it though.
That's almost precisely what I have in mind. Technical issues aside, Apple's biggest problem with releasing OSX for generic Intel is the economy of lost hardware sales. Okay then, how about a "limited" version of OSX with just the Cocoa APIs? This would be a relatively low-risk proposition: hardware sales would be protected, and it and might create a small base of OSX on x86 users -- developers who'd be new to the Mac, I would think, mainly.
Given the limited nature of the product, Apple could downplay the importance of the initiative, so it could disappear quietly if it didn't succeed.
I keep hearing people moan about how they'd try OSX "if only it was available for x86." This is a way for Apple to test those waters without all the potential downsides of releasing a complete version OSX for generic Intel. Hey, the tech press would eat it up!
I may be wrong. But the first post said something about OS X and Final Cut Pro running on a x86 system. Maybe all this stuff is for editing systems in the TV and Movie industry. Apple has been making a big push into this industry with some big software buys. I think that they might be putting together mid and high end editing boxes that can be used on PC's and Mac's to compete with AVID technologies. Just a guess though.
Ref: OS/2
Seamless Windows compatibility invites developers (or, more precisely, the upper management that determine where developer resources go) to ignore your platform altogether.
Re: Whatever...
I don't follow.
blakespot
What is going to happen? Someone might also install Linux on their x86 Mac, or an X Server on their Mac and run KDX?
I don't follow.
blakespot
Re: Re: Whatever...
He altered the KDX screenshot to show a "hacked version" of OS X being distributed on KDX (first line under Server Description). That's the point he was trying to make.
Of course, it wouldn't be quite that simple. It would still only support the processor/chipset combos that Apple provided, unless the "distributors" supplied additional code to support other combinations.
He altered the KDX screenshot to show a "hacked version" of OS X being distributed on KDX (first line under Server Description). That's the point he was trying to make.
Of course, it wouldn't be quite that simple. It would still only support the processor/chipset combos that Apple provided, unless the "distributors" supplied additional code to support other combinations.
Apple is already a staple in the TV and Movie industry. At least in hardware. FCP and DVDSP are starting to make a software staple too.
For post production moving to PCs is a step backwards, IMO. Even when you have an IBM workstation that is certified to by Avid it will still tank on you more than a 6 year old Mac running 6 year old software.
Lethal
Apple was making plenty of money from the OS licensees. Search through the macintouch.com archives: Ric Ford nailed down the numbers pretty conclusively.
The problem was that the cloners tuned out to be much, much better at getting new hardware out the door than Apple was. It's ironic that you mentioned the StarMax sucking, because it was in fact the StarMax line that forced the issue: Motorola was ready to ship the StarMax 6000, their first G3-powered machine, months before Apple's PowerMac G3 (the original beige one) was going to be ready, and it was a faster machine with more PCI slots to boot. And PowerComputing and Daystar weren't far behind.
(Motorola shipped a bunch of final pre-release SM6000s to MacWorld magazine and other reviewers before Apple pulled the plug; I often wonder if there wasn't a whole warehouse of them that got dropped into a landfill, and if any managed to filter out into collectors' hands.)
Apple's own hardware division could only earn money in the absence of any competition. Nobody was going to buy an Apple PowerMac 9600 in favor of a StarMax 6000 no matter how much Motorola paid for an OS license. So, faced with a choice between closing down their hardware division (which up until that point had been the primary revenue source for the company) and killing the cloners, Apple killed the cloners.
Now, this may have been the right decision for Apple in the long run: it would have been exponentially more difficult to pull of the migration to OSX when they weren't in charge of the hardware platform, and had to convince a dozen licensees to make the change. But that doesn't change the fact that Apple's stated reasons for ending OS licensing were nothing but lies.
it's amazing how people don't study history
Doctor Memory has a good memory. In fact, the first Apple G3-based machines came essentially a FULL YEAR after Motorola shipped these G3-equipped Starmax clones to Macworld.
Power Computing had also developed a G3-based machine (which I had been planning to buy) that blew the doors off its PowerTower Pro and anything Apple offered.
The PowerTower Pro was the Power Computing equivalent of the 9600, and it outperformed it. Ditto the top end Starmax.
The real reason Apple killed clones was loss of revenue. Now, the real reason they lost revenue is: They couldn't ****ing compete with other computers on hardware.
Basically, this goes back to Steve Jobs in general; he's pathetic at actually competing with other companies. He is brilliant at coming up with visionary ideas, but he leads the company into this culture of isolation that is unhealthy for Apple's long-term competitiveness. The last numbers I saw were that the clone makers (in aggregate, including Power Computing, APS, UMAX, Motorola, DayStar and the smaller ones) had managed to capture 10-15% of Apple's hardware market. (And rightly so; they were shipping superior products.)
The *real* reason Motorola has "dropped the ball", IMO, is because Steve Jobs ****ed them over. I think Motorola is just handing his ass back to him, and I think he deserves it. (He ****ed over Power Computing customers (like me), too, by buying the company and then disbanding support for their machines.)
The problem is, it's us who get screwed in the end.
The only reason I have continued to use Apple's products over the years is OS 9 and, now, OS X. Software has always been Apple's real strength, though you can make a great case for design and durability of their products, too. Therefore, I'd be happy to see Apple return to the "limited cloning" arrangements like they had. Clearly, other companies were better at milking the PowerPC architecture's performance.
Now, to X86: Clearly, from these lessons, you can deduce that Apple won't ship OS X for commodity hardware unless they are basically ready to go out of business and the owners want to sell their shares at a profit to shareholders. ("Apple shifting to Intel! They will surely profit now! I must buy shares!"...then, a month later: "Apple has announced that they have been acquired outright by AOL/Time Warner at a very competitive value, and all the rich people made a lot of money, while OS X development will now wither and die, because AOL will lay off half of the workers and restructure the company.")
(I think Steve Jobs will do this once he's bored with running Apple, and finished running it into the ground. He's more concerned with his personal legacy than Apple, IMO.)
But, I'd still wager on the 970 as the next CPU for Macs. I think maintaining a current X86 build is just good business sense in case there is some radical shift in the computing landscape...(like, say, if current X86 CPUs were suddenly available for like $20 apiece or something, thus radically shifting the economies of scale in the marketplace).
It's called "hedging your bets", and it's what Microsoft has done brilliantly over the years.
Doctor Memory has a good memory. In fact, the first Apple G3-based machines came essentially a FULL YEAR after Motorola shipped these G3-equipped Starmax clones to Macworld.
Power Computing had also developed a G3-based machine (which I had been planning to buy) that blew the doors off its PowerTower Pro and anything Apple offered.
The PowerTower Pro was the Power Computing equivalent of the 9600, and it outperformed it. Ditto the top end Starmax.
The real reason Apple killed clones was loss of revenue. Now, the real reason they lost revenue is: They couldn't ****ing compete with other computers on hardware.
Basically, this goes back to Steve Jobs in general; he's pathetic at actually competing with other companies. He is brilliant at coming up with visionary ideas, but he leads the company into this culture of isolation that is unhealthy for Apple's long-term competitiveness. The last numbers I saw were that the clone makers (in aggregate, including Power Computing, APS, UMAX, Motorola, DayStar and the smaller ones) had managed to capture 10-15% of Apple's hardware market. (And rightly so; they were shipping superior products.)
The *real* reason Motorola has "dropped the ball", IMO, is because Steve Jobs ****ed them over. I think Motorola is just handing his ass back to him, and I think he deserves it. (He ****ed over Power Computing customers (like me), too, by buying the company and then disbanding support for their machines.)
The problem is, it's us who get screwed in the end.
The only reason I have continued to use Apple's products over the years is OS 9 and, now, OS X. Software has always been Apple's real strength, though you can make a great case for design and durability of their products, too. Therefore, I'd be happy to see Apple return to the "limited cloning" arrangements like they had. Clearly, other companies were better at milking the PowerPC architecture's performance.
Now, to X86: Clearly, from these lessons, you can deduce that Apple won't ship OS X for commodity hardware unless they are basically ready to go out of business and the owners want to sell their shares at a profit to shareholders. ("Apple shifting to Intel! They will surely profit now! I must buy shares!"...then, a month later: "Apple has announced that they have been acquired outright by AOL/Time Warner at a very competitive value, and all the rich people made a lot of money, while OS X development will now wither and die, because AOL will lay off half of the workers and restructure the company.")
(I think Steve Jobs will do this once he's bored with running Apple, and finished running it into the ground. He's more concerned with his personal legacy than Apple, IMO.)
But, I'd still wager on the 970 as the next CPU for Macs. I think maintaining a current X86 build is just good business sense in case there is some radical shift in the computing landscape...(like, say, if current X86 CPUs were suddenly available for like $20 apiece or something, thus radically shifting the economies of scale in the marketplace).
It's called "hedging your bets", and it's what Microsoft has done brilliantly over the years.
if apple makes an os for the pc, apple is dead.
The only thing apple needs is good marketing and a good campaign to instruct people, thou this would be very expensive, let us think of a way of helping,hehhehe,do you think we could find an answer?
Let this be our assingment this weekend.
The only thing apple needs is good marketing and a good campaign to instruct people, thou this would be very expensive, let us think of a way of helping,hehhehe,do you think we could find an answer?
Let this be our assingment this weekend.
Apple and the attack of the Mac clones!!!
Though Apple is partly to blame on killing the Mac clones, I feel that old excuse is somewhat lame... all those clone makers could have easily licensed BeOS from Be Inc. which was running exclusively on PowerPC at the time and well almost ready for primetime, so the clone makers are as much to blame for being stupid with their get rich quick scheme, instead of working for the market (idiots)!
Though Apple is partly to blame on killing the Mac clones, I feel that old excuse is somewhat lame... all those clone makers could have easily licensed BeOS from Be Inc. which was running exclusively on PowerPC at the time and well almost ready for primetime, so the clone makers are as much to blame for being stupid with their get rich quick scheme, instead of working for the market (idiots)!
Re: Apple and the attack of the Mac clones!!!
What the hell are you talking about?
"Almost ready for primetime" doesn't cut any mustard in the real market. BeOS was a beautifully designed system, but it had no applications. People don't buy computers to beta test new OSes, they buy them to get work done. And in 1997, you simply could not do that on the BeOS. Netscape? Not available. Word? Not available. Excel? Not available. Quark XPress? Hah! Photoshop? As if.
And as it happens, PowerComputing, the largest of the clone makers, did offer machines with BeOS preinstalled. It didn't save them.
What the hell are you talking about?
"Almost ready for primetime" doesn't cut any mustard in the real market. BeOS was a beautifully designed system, but it had no applications. People don't buy computers to beta test new OSes, they buy them to get work done. And in 1997, you simply could not do that on the BeOS. Netscape? Not available. Word? Not available. Excel? Not available. Quark XPress? Hah! Photoshop? As if.
And as it happens, PowerComputing, the largest of the clone makers, did offer machines with BeOS preinstalled. It didn't save them.
My 2¢...I think Apple has been moving themselves in to a platform non-chip specific since SJ has came into office. However, in order to move the entire community he had to make it a viable platform in the first place, and to move the developers and the users to this platform w/o ripping to shreds AAPL. That had to take a lot of time, money, secrecy, and an iron grip on the community in order to do this, that is why they killed off the clones, and why we are only now seeing the spoils of there work after so many years.
SJ even at time said that "Rhapsody" (the OS X codeword at that time) was going to work on the Itanium, as said in a scene by ATAT,
This movement had to be done slowly and in many steps...
First, all the apps would have to written in a language that would be OS specific, rather than OS and hardware specific. This would also have to be intriguing and palpable to the developers, enter Cocoa.
Second the current Apps that are in development would have to be able to work in the same environment as the new Cocoa language and the very new environment that will move to a non specific hardware environment, and all done w/o breaking the apps and pissing off the developers in the process, all done with less than 20% of their current code would have to change, enter Carbon.
Thirdly this OS that would have to be portable, the portable part would be threw the open-sourced kernel, the kernel would fetch instructions given out by the corresponding layer on top that will not be hardware specific but rather to the kernel that can be changed to suit to the hardware necessities. Open sourced so a computer maker can use this OS threw the fastest, coolest, and cheapest, hardware be it x86, PPC, x86-64, PPC-64... ect. , And the shell on top would be something sold threw Apple, enter Darwin and its shell Aqua.
All of this would have to be done, while continuing relationships w/ companies, such as that in Redmond to continue the development of Office: Mac. Apple is also welcoming the communities that are still growing such as that from the Linux and Open source movements to be able to easily port over their projects to Apple's software, such as Apache.
I think most of this has played out very well; all of the developers have taken the road path that Apple has set out, all writing in to the Carbon language for OS X. Now Apple is telling them to take the next logical step, which would be for them to write into Cocoa and be fully OS X specific, and most developers should be on this path by now. I think this why Quark decided at the last minute of changing all its code to Cocoa to suit this, and why it is taking so long for a port of their software. Once most of the big apps start coming out into Cocoa then they can bring out OS X for everybody. The only problem would be that the Carbon apps that are out now, are still requesting PPC specific hardware, and they will not be able to work or at least not as well as Cocoa apps, being that it might not be a PPC chip inside the computer the OS is on.
Nevertheless, I don't think that an x86 port is going to happen with in '03. Its more likely in '04, by then a lot of Cocoa third party apps will be out, OS X will be more of a main stay in the computer community, the Macintoshes will be looking a lot more tasty w/ the possibility of the PPC 970, and Apple will be looking less like they are running away from the PPC. Even in 04 though I think they will first just call it something like OS X x86 Server, so it wont scare the be-jesus out of MS enough to drop Mac-BU and lose the number one reasons why people can switch to Mac, namely Office.
This is of course IMHO, and I could be just drinking too much coffee, but I think this they way Apple has been moving. Apple loves the PPC, they have even said this in a interview lately (I cant find it right now), and it allows them to have the best and fastest Laptops and the coolest form-factors for their computers, but the real speed and money is threw third party computer dealers and the biggest chip makers. Apple can continue to make money on the hardware side of things threw people wanting to use legacy software on new computers that wont work on the x86, from making really cool looking computers, and the fastest Laptops from using PPCs, and by saleing great software to more than just 5% of the market, and even might pull some of the big cash from MS by getting in the pants of corporations.
SJ even at time said that "Rhapsody" (the OS X codeword at that time) was going to work on the Itanium, as said in a scene by ATAT,
This movement had to be done slowly and in many steps...
First, all the apps would have to written in a language that would be OS specific, rather than OS and hardware specific. This would also have to be intriguing and palpable to the developers, enter Cocoa.
Second the current Apps that are in development would have to be able to work in the same environment as the new Cocoa language and the very new environment that will move to a non specific hardware environment, and all done w/o breaking the apps and pissing off the developers in the process, all done with less than 20% of their current code would have to change, enter Carbon.
Thirdly this OS that would have to be portable, the portable part would be threw the open-sourced kernel, the kernel would fetch instructions given out by the corresponding layer on top that will not be hardware specific but rather to the kernel that can be changed to suit to the hardware necessities. Open sourced so a computer maker can use this OS threw the fastest, coolest, and cheapest, hardware be it x86, PPC, x86-64, PPC-64... ect. , And the shell on top would be something sold threw Apple, enter Darwin and its shell Aqua.
All of this would have to be done, while continuing relationships w/ companies, such as that in Redmond to continue the development of Office: Mac. Apple is also welcoming the communities that are still growing such as that from the Linux and Open source movements to be able to easily port over their projects to Apple's software, such as Apache.
I think most of this has played out very well; all of the developers have taken the road path that Apple has set out, all writing in to the Carbon language for OS X. Now Apple is telling them to take the next logical step, which would be for them to write into Cocoa and be fully OS X specific, and most developers should be on this path by now. I think this why Quark decided at the last minute of changing all its code to Cocoa to suit this, and why it is taking so long for a port of their software. Once most of the big apps start coming out into Cocoa then they can bring out OS X for everybody. The only problem would be that the Carbon apps that are out now, are still requesting PPC specific hardware, and they will not be able to work or at least not as well as Cocoa apps, being that it might not be a PPC chip inside the computer the OS is on.
Nevertheless, I don't think that an x86 port is going to happen with in '03. Its more likely in '04, by then a lot of Cocoa third party apps will be out, OS X will be more of a main stay in the computer community, the Macintoshes will be looking a lot more tasty w/ the possibility of the PPC 970, and Apple will be looking less like they are running away from the PPC. Even in 04 though I think they will first just call it something like OS X x86 Server, so it wont scare the be-jesus out of MS enough to drop Mac-BU and lose the number one reasons why people can switch to Mac, namely Office.
This is of course IMHO, and I could be just drinking too much coffee, but I think this they way Apple has been moving. Apple loves the PPC, they have even said this in a interview lately (I cant find it right now), and it allows them to have the best and fastest Laptops and the coolest form-factors for their computers, but the real speed and money is threw third party computer dealers and the biggest chip makers. Apple can continue to make money on the hardware side of things threw people wanting to use legacy software on new computers that wont work on the x86, from making really cool looking computers, and the fastest Laptops from using PPCs, and by saleing great software to more than just 5% of the market, and even might pull some of the big cash from MS by getting in the pants of corporations.
Re: Maybe We're missing the point......
YES! Ive been thinking this ALL ALONG! Im glad someone posted it. Let me second this opinion.
YES! Ive been thinking this ALL ALONG! Im glad someone posted it. Let me second this opinion.
For most professionals, 10 is enough. Hell, T1 isn't even close to 10, only at 1.544Mbps.
What the hell do you need 1000Mbps for? If you're a regular home user, you won't have anything above 1.5Mbps on cable/DSL, so it doesn't frickin matter. It's just a showoff/marketing thing.
100Mbps is also plenty for most corporate networks.
Besides, you can simply pop in a PCI gigabit ethernet for a couple more bucks, not a big deal.
Come on, be realistic. Why waste on gigabit ethernet when it' not needed.
By the way, they already made OSes for the PC. Darwin, Mac OS (Startrek or whatever), Rhapsody (early Mac OS X), and finally Mac OS X (as far as we've heard). If you count NeXT, count NeXT OS.
Also, the Macintosh *is* a PC--personal computer, isn't that what Apple sparked the revolution of anyway.
Mmmmmph. Remember what the Mac's core market is: video, sound and graphics editing. The difference between 10, 100 and 1000baseT ethernet is not so academic when you're tossing multi-gigabyte quicktime files around on a day to day basis.
Comparing your LAN speed to your internet connection is apples-and-oranges: you use them for different things, and you don't need anywhere near as much performance on your WAN link as on your LAN. My company is only connected to the internet by a T1, but we most certainly use gigabit ethernet to connect our PowerMac video editing stations to the file servers, and if something faster were available we'd use that too.
That said, yeah, buying a gigabit PCI card for a Dell is only going to set you back another $75, so it's not a huge pricing issue.