MAC OX S For Intel?
- Thread starter dvigue
- Start date
sfish
Fark Lover
This is strictly IMO, but I do think it's been created and continues to be kept up with the PPC-native version of X. If Motorola ever goes under and IBM doesn't fit Apple's vision (either by inability or corporate incompatability), I think we'll be surprised how quickly the Intel version of Mac OS hits shelves.
ElDiabloConCaca
U.S.D.A. Prime
I don't think an Intel version of Mac OS X has been developed. You can get the UNIX underpinings of OS X for an Intel machine, though, sans beautiful interface and all. Go grab the Darwin ISO images from the developer's section of Apple.com and go crazy on your Windows box...
Apple has been a hardware and software company, and one reason the Mac OS is so elegant to use is Apple's strategic integration of their hardware and software. I doubt they'd port their operating system to another platform without the threat of losing their hardware section of business. If, for some reason, all the Apple computers simultaneously blow up and cease to exist, I think Apple would think about moving to an Intel-platform, but until then, I think there's little hope an entire operating system would run on two different machines.
Apple has been a hardware and software company, and one reason the Mac OS is so elegant to use is Apple's strategic integration of their hardware and software. I doubt they'd port their operating system to another platform without the threat of losing their hardware section of business. If, for some reason, all the Apple computers simultaneously blow up and cease to exist, I think Apple would think about moving to an Intel-platform, but until then, I think there's little hope an entire operating system would run on two different machines.
As long as the PowerPC family of processors stay ahead of Intel/AMD processors (and the G5 promises that very well), there won't be an Intel compatible version of OS X.
The first two developer releases of Mac OS X Server 1.0 (Rhapsody DR1 and Rhapsody DR2) were available for PC Compatibles back then. Remember that OpenStep was ported to the PPC then, and it *was* running on Intel hardware.
I guess the Intel port is still happily alive. It doesn't include the Classic environment, but surely contains Cocoa and maybe even Carbon to some extent. But it'd be an internal project only with much less activity as long as the need isn't there to provide the world with the software.
Cocoa apps are just a recompile away from running on a OS X for PC Compatibles. Carbon apps: Dunno.
But right now: There is none.
The first two developer releases of Mac OS X Server 1.0 (Rhapsody DR1 and Rhapsody DR2) were available for PC Compatibles back then. Remember that OpenStep was ported to the PPC then, and it *was* running on Intel hardware.
I guess the Intel port is still happily alive. It doesn't include the Classic environment, but surely contains Cocoa and maybe even Carbon to some extent. But it'd be an internal project only with much less activity as long as the need isn't there to provide the world with the software.
Cocoa apps are just a recompile away from running on a OS X for PC Compatibles. Carbon apps: Dunno.
But right now: There is none.
Apple, motorola, and IBM developed the PPC chip to show how the benefits of a RISC processor can apply to desktop computing. apple was _almost_ the first computer to try this. i understand the acorn had one earlier. RISC processing is a movement in processors that has proven very effective for server platforms in the last 15 years. All platforms use it these days. let me list them: Sun, sparc; SGI,MIPS;HP HP-RISC;IBM,PPC;DEC/Compaq,alpha.
wait, did i say everyone? there is one architecture that doesn t:Intel. This is the only architecture that doesn t use RISC processing. complex instruciton sets were great in the 70s when silicon chips were at a premium. now that we have more memory than we know what to do with, this paradigm doesn t make any sense.
If you need proof of how RISC processing is better, look at how every computer manufacturer uses it for their servers. or look at how AMD beats faster clocked intel machines. AMD uses a RISC virtual machine with intel emulation. Transmeta does the same thing with their crusoe chip, and their chips use 1/10th the energy of intel. or just look at the way apple computers beat intel in benchmarks.
It wouldn t make any sense for apple to switch to intel. if motorola went under, apple can get it s PPC chips from IBM. IBM has some great machines using PPC. to switch to intel would mean that the OS would have to be completely rewritten. Intel is a small-endian architecture and PPC is big-endian, and it will take a long time to report it. it will never happen.
wait, did i say everyone? there is one architecture that doesn t:Intel. This is the only architecture that doesn t use RISC processing. complex instruciton sets were great in the 70s when silicon chips were at a premium. now that we have more memory than we know what to do with, this paradigm doesn t make any sense.
If you need proof of how RISC processing is better, look at how every computer manufacturer uses it for their servers. or look at how AMD beats faster clocked intel machines. AMD uses a RISC virtual machine with intel emulation. Transmeta does the same thing with their crusoe chip, and their chips use 1/10th the energy of intel. or just look at the way apple computers beat intel in benchmarks.
It wouldn t make any sense for apple to switch to intel. if motorola went under, apple can get it s PPC chips from IBM. IBM has some great machines using PPC. to switch to intel would mean that the OS would have to be completely rewritten. Intel is a small-endian architecture and PPC is big-endian, and it will take a long time to report it. it will never happen.
Someone wrote: 'All chip makers use RISC. All? No, there's one (Intel) that still makes CISC chips.'
Also: 'All operating systems are UNIX-based. All? No, there's one (Microsoft) that still uses a proprietary base for their operating systems.'
The future is RISC. The future is UNIX. But Intel/AMD and Microsoft rule the Desktop. But who the f* read Asterix and was with the Romans?
I guess I got that totally mixed up now. We should get the attitude that in fact 'the world' (including us) uses RISC and UNIX. And that Wintel is an island that tends to be incompatible.
Also: 'All operating systems are UNIX-based. All? No, there's one (Microsoft) that still uses a proprietary base for their operating systems.'
The future is RISC. The future is UNIX. But Intel/AMD and Microsoft rule the Desktop. But who the f* read Asterix and was with the Romans?
I guess I got that totally mixed up now. We should get the attitude that in fact 'the world' (including us) uses RISC and UNIX. And that Wintel is an island that tends to be incompatible.
.dev.lqd
Angry Member
The article on arstechnica I referenced makes a pretty good case for their not really being many 'RISC' processors anymore. Back in the day there was a big difference- chipmakers were like, "Hey- instead of adding more ish to do more we could try and improve performance by making the ish we have do more." I'm straining for a metaphor- but my mind fails me now.
Nowadays, there aren't really many RISC or CISC processors- almost all the major microprocessors, especially the mainstream ones for desktop PC's, are what one might call "post risc"- where there's been a melding of the two schools of processor kung fu. CISC elements in the G4, RISC type ish bleeding into the K7... etc.
Even the P4 isn't 'CISC' purely, although it's certainly easier to argue that it's more CISC than the G4 and even the K7. It's kinda like "yeah- that band used to be good- back when they were punk... now they're all ska and I just don't dig it at all".
Then again, that's just my take on it, but really read the ArsTechnica article- it's very very good.
-stephen
Nowadays, there aren't really many RISC or CISC processors- almost all the major microprocessors, especially the mainstream ones for desktop PC's, are what one might call "post risc"- where there's been a melding of the two schools of processor kung fu. CISC elements in the G4, RISC type ish bleeding into the K7... etc.
Even the P4 isn't 'CISC' purely, although it's certainly easier to argue that it's more CISC than the G4 and even the K7. It's kinda like "yeah- that band used to be good- back when they were punk... now they're all ska and I just don't dig it at all".
Then again, that's just my take on it, but really read the ArsTechnica article- it's very very good.
-stephen
serpicolugnut
OS X Supreme Being
Even if Apple shifted their OS to run on Intel/AMD processors, they would make it proprietary and only run on hardware they sold. Apple is a complete solutions provider, and wants to control the whole widget.
I doubt Apple would abandon the PPC. Apple has the option to buy the PPC assets from Motorola come January. They have 6 months to do it if they want. I bet they will.
I doubt Apple would abandon the PPC. Apple has the option to buy the PPC assets from Motorola come January. They have 6 months to do it if they want. I bet they will.
Anyway, the architecture is quite different.
Especially when it comes to putting the graphics up on the screen....
if you put a Quickdraw emulator on a high speed bus accessable to the CPU then you can have your mac on a Wintel Box.
One of the big problems for windows has been that the GUI has to be processed by the main CPU.
Now that the CPU is getting so fast, this is turning out to be less of a hit on "apparent" performance.
And at the same time, you can keep your coffee warm...a true "desktop device".
I have also heard that we can soon have pentium processors for the kitchen, the laundry and the bathroom. In the latter case, one might note, that after 30 years, the thing would be toilet trained anyway.
I digress...
anyway, as far as I know, the code to quickdraw is still the biggest secret in the universe and since Quartz and Carbon are derivitive decendents, it is not likely that they will be making an appearance on windows real soon, though who knows what that little piece of steath software known as QT.dll has embedded?
Especially when it comes to putting the graphics up on the screen....
if you put a Quickdraw emulator on a high speed bus accessable to the CPU then you can have your mac on a Wintel Box.
One of the big problems for windows has been that the GUI has to be processed by the main CPU.
Now that the CPU is getting so fast, this is turning out to be less of a hit on "apparent" performance.
And at the same time, you can keep your coffee warm...a true "desktop device".
I have also heard that we can soon have pentium processors for the kitchen, the laundry and the bathroom. In the latter case, one might note, that after 30 years, the thing would be toilet trained anyway.
I digress...
anyway, as far as I know, the code to quickdraw is still the biggest secret in the universe and since Quartz and Carbon are derivitive decendents, it is not likely that they will be making an appearance on windows real soon, though who knows what that little piece of steath software known as QT.dll has embedded?
something occurred to me after i wrote my response to this thread: sun ported their OS to intel. and sun uses RISC bigendian processor. so i guess the technical difficulties are not too great. OS X could be done, and since the kernel is already done, it might not be too hard.
there still remain 2 issues with this: microsoft s OS supports an enormnous database of hardware. if apple ported OSX to the PC, it would have to worry about drivers for all the myriads of PC hardware components. right now apple only has to worry about supporting apple s hardware. it s the same with sun. i ve tried Solaris on intel. my video card and ethernet cards didn t work. you haver to have a hardware package endorsed by sun if you want a functional system. it is an attempt by sun to let users access the most popular UNIX on the market at much lower costs. sun s hardware is vcery expensive. i think they are mostly tyrying to make it easy for existing NT admins change over easily. sun is the most popular server OS, but windows server OSes have been getting popular. i guess OS X opens a new door of possiblities for server solutions, but i think it is a very iffy question, how that will turn out.
the last point is the difference in philosophies between the two platforms. with a PC you have a huge amount of choice for every component in your system. these choices include the really crappy brands. you can ignorantly put together a system with a 2 GHz processor, and 100MHz system bus, 5400 rpm IDE hard drive, 64M RAM at 100MHz, and you will never see your processors speed. you can get an ethernet card that won t properly negotian it s IRQ with the system board, and fail to work. then you can watch windows 98 struggle to make it all work. oops windows doesn t have a driver for your video card? sorry. or you can get apple to sell you a machine where the OS is taylored to your hardware. and with apple, you know you re getting good hardware. new technologies like firewire(1394), wireless (802.11), and gigabit ethernet (i don t know the ieee no for that one! how embarrassing!).
there was a time when apple lisenced out it s architecture, for about 3 years, about 10 years ago. then you could buy clones from UMAX and motorola, prices were cheaper, and you had more options, including the option to get cheaper hardware. but steve came back and took care of that situation. apple software on apple hardware. it has it s pros and cons, but i for one like what apple is doing these days.
there still remain 2 issues with this: microsoft s OS supports an enormnous database of hardware. if apple ported OSX to the PC, it would have to worry about drivers for all the myriads of PC hardware components. right now apple only has to worry about supporting apple s hardware. it s the same with sun. i ve tried Solaris on intel. my video card and ethernet cards didn t work. you haver to have a hardware package endorsed by sun if you want a functional system. it is an attempt by sun to let users access the most popular UNIX on the market at much lower costs. sun s hardware is vcery expensive. i think they are mostly tyrying to make it easy for existing NT admins change over easily. sun is the most popular server OS, but windows server OSes have been getting popular. i guess OS X opens a new door of possiblities for server solutions, but i think it is a very iffy question, how that will turn out.
the last point is the difference in philosophies between the two platforms. with a PC you have a huge amount of choice for every component in your system. these choices include the really crappy brands. you can ignorantly put together a system with a 2 GHz processor, and 100MHz system bus, 5400 rpm IDE hard drive, 64M RAM at 100MHz, and you will never see your processors speed. you can get an ethernet card that won t properly negotian it s IRQ with the system board, and fail to work. then you can watch windows 98 struggle to make it all work. oops windows doesn t have a driver for your video card? sorry. or you can get apple to sell you a machine where the OS is taylored to your hardware. and with apple, you know you re getting good hardware. new technologies like firewire(1394), wireless (802.11), and gigabit ethernet (i don t know the ieee no for that one! how embarrassing!).
there was a time when apple lisenced out it s architecture, for about 3 years, about 10 years ago. then you could buy clones from UMAX and motorola, prices were cheaper, and you had more options, including the option to get cheaper hardware. but steve came back and took care of that situation. apple software on apple hardware. it has it s pros and cons, but i for one like what apple is doing these days.
I cannot believe that Apple would concurrently develop an OS X build for Intel processors. They have too much to do making it function the way we macheads want on their own (Apple) hardware. Add the myriad of combinations of hardware one could assemble with an Intel mobo and you have a real challenge. We are still waiting for many device drivers for X, could you imagine what it would be like for an Intel processor?
It isn't going to happen, let it die!
It isn't going to happen, let it die!
the list of supported platforms is really pretty small
AGP graphics basically unsupported.
basically roll back the clock a year or more on which platform you can actually install it on.
like the old days of linux buy your machine carefully...
and the BIOS on my machine would not stop putting some plug-and play info into the mix
(well it was not on the supported list now I know why)
I removed the offending entries from the list,
and the install proceeded, but I couldn't ever boot from it...
(though I crashed out of the installer shell and things were basically working....)
So, maybe, as lethe says, some NT admin could...but only some.
So, finally windows has a good bit of effort in supporting most things that people can buy and when you look at the specs on the hardware it looks like it should just float off the desk.
But then you install windows and it just becomes this sort of OK machine....
The unfortunate thing is ...that seems to be what people are satisfied with.
<b> Insanely Great is only for Mac fanatics.
</b>
If someone built a cool windowmanager for Intel Darwin which could utilize the QD routines without bogging down the main cpu. (read add-on technology which is not on the PCI bus and also not just software emulation) then you could have your OSX for any platform...
until then I am sure also, Apple doesn't want the responsibility.
AGP graphics basically unsupported.
basically roll back the clock a year or more on which platform you can actually install it on.
like the old days of linux buy your machine carefully...
and the BIOS on my machine would not stop putting some plug-and play info into the mix
(well it was not on the supported list now I know why)
I removed the offending entries from the list,
and the install proceeded, but I couldn't ever boot from it...
(though I crashed out of the installer shell and things were basically working....)
So, maybe, as lethe says, some NT admin could...but only some.
So, finally windows has a good bit of effort in supporting most things that people can buy and when you look at the specs on the hardware it looks like it should just float off the desk.
But then you install windows and it just becomes this sort of OK machine....
The unfortunate thing is ...that seems to be what people are satisfied with.
<b> Insanely Great is only for Mac fanatics.
</b>
If someone built a cool windowmanager for Intel Darwin which could utilize the QD routines without bogging down the main cpu. (read add-on technology which is not on the PCI bus and also not just software emulation) then you could have your OSX for any platform...
until then I am sure also, Apple doesn't want the responsibility.
i read that article on arstechnica. the one about the G4e vs P4. it was good. even more educational for me was the article it linked to. the RISC vs CISC article.
i guess that i have to admit, i m no expert in computer architecture, and was a victim of media propoganda on the RISC vs CISC matter. apparently the terms no longer even make sense...
i guess that i have to admit, i m no expert in computer architecture, and was a victim of media propoganda on the RISC vs CISC matter. apparently the terms no longer even make sense...
Dak RIT
Missing Intelligence
Actually there was an article a while back (even before OS X DP3) by an engineer at Apple that outlined what would actually be necessary to port OS X to run on tan x86 machine... very little modification at all would actually be necessary on the upper layers of OS X for x86... for those of you that know much about programming... OS X is a true object-oriented OS, and I'm sure the engineers at Apple took full advantage of OOP to make it easy to port to different architectures.
The majority of the work that would need to be done to bring OS X to anything else would be mainly in the lower levels, specifically Darwin. O wait, Darwin is kept up-to-date for the x86 through a nice little open source project And even more intriguing... a large percentage of applications compiled to run on OS X on the PPC would also work just as well on OS X on any other architecture.
Realistically I doubt OS X would be brought to a different architecture from the PPC without a drastic change in either the availability of PPC chips or a large shift in Apple's business model. The later at first glance seems unlikely... although I personally think that's exactly what Apple is (slowly) working on doing right now. Although I don't want to try to make any predictions or spread rumors... a lot can change in a short period of time, so who really knows.
Even so, it's interesting to know the option does exist... I think there's little doubt right now that if OS X was released as open source tomorrow, distributed freely, and available for the x86, that MacOS X would dominate the consumer market in as little as a 1 year time span (in other words... if Apple can successfully shift its business model... it holds a trump card capable of toppling the M$ behemoth... although that's just rumor fodder).
Cheers,
Dak
The majority of the work that would need to be done to bring OS X to anything else would be mainly in the lower levels, specifically Darwin. O wait, Darwin is kept up-to-date for the x86 through a nice little open source project
And even more intriguing... a large percentage of applications compiled to run on OS X on the PPC would also work just as well on OS X on any other architecture.Realistically I doubt OS X would be brought to a different architecture from the PPC without a drastic change in either the availability of PPC chips or a large shift in Apple's business model. The later at first glance seems unlikely... although I personally think that's exactly what Apple is (slowly) working on doing right now. Although I don't want to try to make any predictions or spread rumors... a lot can change in a short period of time, so who really knows.
Even so, it's interesting to know the option does exist... I think there's little doubt right now that if OS X was released as open source tomorrow, distributed freely, and available for the x86, that MacOS X would dominate the consumer market in as little as a 1 year time span (in other words... if Apple can successfully shift its business model... it holds a trump card capable of toppling the M$ behemoth... although that's just rumor fodder
).Cheers,
Dak
RacerX
Old Rhapsody User
This always brings us back to a time when there actually was an Apple OS for Intel based systems. As someone who uses Rhapsody for Intel on a daily basis, who has studied the history and searched for any compatible applications, I can tell you that we are not going to see an Intel version of Mac OS X any time in the near future. One of the key factors in the death of Rhapsody for Intel was developer indifference. Between the time Apple released the Rhapsody Developer Release (Rhapsody 5.0) and when Apple killed the Intel version just before releasing Mac OS X Server 1.0 (Rhapsody 5.3) the ratio of PPC apps to Intel apps was at least 2 to 1. What is so surprising about that is that you only needed to click a radio button to make any app run on both before the final build. Even more interesting is the fact that most of the developers were former OpenStep developers who had been working with OPENSTEP 4.2 on Intel based systems. With half as many native apps and no blue box environment, Rhapsody for Intel was DOA. In fact, the Rhapsody client release (Rhapsody 5.2) never made it to the public because of poor developer responce.
Apple would not be able to port "classic" and they would have a very hard time getting developers to rework their apps to run on Intel based systems. And Apple is not about to go down the same road they did with Rhapsody for Intel any time in the near future.
Apple would not be able to port "classic" and they would have a very hard time getting developers to rework their apps to run on Intel based systems. And Apple is not about to go down the same road they did with Rhapsody for Intel any time in the near future.
I've used Rhapsody DR2 for PC Compatibles myself. At the time, many software packages were out compiled as FAT (somesoftware_1x_PI.tar.gz), others were available in two separate packages.
But also it was at a very early stage. And developers on the PowerPC side were also reluctant to develop for Rhapsody - that's why Steve Jobs had to introduce Carbon to pave the way for all the major Macintosh developers who didn't want to learn Cocoa.
My guess is still that the PC Compatible project (or rather the 'other platforms' project) is happily alive, that it's only some steps behind the Mac OS X on PowerPC development.
Why that? Because this gives Apple the freedom to choose. And it needs that. Sometimes the AIM group seems so flaky that I already expected Apple to jump out of it and go someplace else. But Apple can't abandon the PowerPC just yet. Apps must at least be carbonated and should be rewritten in Cocoa for such a step.
But also it was at a very early stage. And developers on the PowerPC side were also reluctant to develop for Rhapsody - that's why Steve Jobs had to introduce Carbon to pave the way for all the major Macintosh developers who didn't want to learn Cocoa.
My guess is still that the PC Compatible project (or rather the 'other platforms' project) is happily alive, that it's only some steps behind the Mac OS X on PowerPC development.
Why that? Because this gives Apple the freedom to choose. And it needs that. Sometimes the AIM group seems so flaky that I already expected Apple to jump out of it and go someplace else. But Apple can't abandon the PowerPC just yet. Apps must at least be carbonated and should be rewritten in Cocoa for such a step.