There are some things in the OS X 10.1.3 CD that I found interesting. First there is a /private/drivers/i386 folder. Furthermore, in this folder is the standard System.config file which is the same as the one included with Darwin x86 1.4.1, but there is also a .config file for a network install option that is not in the Darwin distribution.
There is also a /usr/standalone/i386 folder. This is where the files necessary for an el-torito bootable CD would be kept. The only thing missing from this folder is the boot.flp file which the bios uses to emulate a 2.88 MB floppy. I know Darwin runs on Intel, I have it installed on an old PII, and that this folder is needed for it, but why would Apple leave this on a CD meant for a PowerPC machine?
This made me wonder whether some of the files on the CD were built fat by accident. I decided to try to build my own x86 install CD. I knew it was a long shot, but what the hell.
I had access to a new iBook, so I built and installed the mkisofs package needed to build a Darwin install CD. I then copied to the iBook the 10.1.3 CD and the Darwin x86 CD. I then mixed and matched the contents of the two to make a new CD structure that included the Intel kernel and low level binaries in the /bin and /sbin folder and the /System/Installation and /Applications etc. from the 10.1.3 CD. I then made the custom CD ISO and burned it. It booted fine on my Athlon, loaded the kernel, and then began to try to run the rc.cdrom install script from 10.1.3. I got excited and thought I was on to something when it then began to give me error messages that the wrong CPU type was specified in a bunch of Framework files. It was not able to load the CDIS splash screen in the/System/Installation folder or the Installer.app in the Applications folder. While I was defeated in my own attempts, I found it interesting how easy it would be for Apple to build an x86 build of OS X. Now all I need is someone who knows how to decompile the necessary file on the 10.1.3 CD and build them for x86?
<strong>My friend 'in the know' if you could say that has been whispering this to me for about two weeks now. He has heard that Apple has a partnership with AMD. </strong><hr></blockquote>It's hard to imagine converting to a new chip, but maybe AMD will buy out Motorola's semiconductors and pick up the PowerPC.
although everyone seems to love AMD - people should be aware that they aren't making money, and havent for a while - there are perhaps better alternatives... Itanium2 is looking good today, and when transistor densities make the currently expensive Itanium2 into a econo chip, it could be very attractive..... intel is VERY determined to see this processor succeed.... and the engineers intel obtained from the Alpha group are on the job, they are old-timers, and they are top notch.
<strong>My friend 'in the know' if you could say that has been whispering this to me for about two weeks now. He has heard that Apple has a partnership with AMD. He has also heard that Apple plans to release boxes that Dual Boot.
The second one I am leery of but he swears on the first, and this is the same guy who told me about the solid state scroll on thhe new iPods about two weeks ago. I don't think Apple is looking to go x86, they would piss off too many developers. But a PowerPC made by AMD would work well.</strong><hr></blockquote>
d00d! If you could get your buddy to elaborate on this....it would be very cool!
Could this partnership with AMD have to do with a mobo chipset, and Apple will continue to use Moto as their CPU supplier? Or do you think this AMD-Apple partnership is about CPUs?
And, aren't all current Macs "Dual Boot", in that one can boot into OS X or OS 9? Or do you mean dual boot, as in, these new "Macs" will be able to boot either OS X, or Windows?
Or could this Dual Boot mean that one boots into OS X on PPC for older apps, but boots into OS X on x86 for newer apps?
Just curious.
What troubles me about all this is the following:
1. Mac developers, including Apple, have invested a great amount of resources in Altivec. From OS X to Photoshop to screen savers, a multitude of Mac code has been written with Altivec in mind. Now that the G4 is finally in both pro AND consumer Macs, a developer can write altivec code with the knowledge that ANY Mac user will benefit. Futhermore, altivec optimizations are now an incentive for all mac users to choose a particular piece of software over another. It would be a small tragedy if right when Altivec was finally gaining widespread adoption and support, Apple abandoned it.
2. The transition to OS X continues, and by the looks of things, it will be years before the Mac user base has fully migrated to OS X, with OS 9 native apps being relics from the past. This transition is costly to both Apple and developers, AND to Mac users. Mac users who switch to OS X are also compelled to replace all of their software with OS X native versions, if not immediately, then after a few months of using the classic environment. For Apple to undertake another major transition within the next few years would be dangerous. Not only would Apple risk alienating developers (some who have invested lots of time and money in Altivec optimizations, imagine all the Mac programmers who've taken the time to learn how to code for altivec), but more importantly, Apple could alienate the entire Mac user base. Asking them to buy all new software once is bad enough, but doing it again....I don't see many people sticking with the platform. The migration to OS X is confusing enough for most people, try throwing in a concomitant migration to a different CPU architecture and you suddenly have a bunch of confused Mac users who will see Windows as being more simple and cheaper.
3. Motorola DOES have some sort of "G5" ready, and IBM has the Power 5 which will supposedly be good for workstations. With options like these, why would Apple choose x86? Furthermore, x86 is supposedly near the end of it's life. I don't understand the technical reasons, but it's something I've read a few places. Not in the next few years, but certainly within the next decade. IF Apple were to choose a new architecture, wouldn't it be better to choose something that is new, with a long life ahead of it, rather than something in it's twilight years, with the end in sight, like x86?
I seriously doubt that Apple is gonna switch over to the x86 platform; anytime soon OR ever...!
The IBM Power 4/5 derivative sounds good... But the AMD fabbed / Apple designed G5 sounds more like the choice of a company (... a man...) that wants complete control over their 'widget'...
Couple this with the Apple / nVidia secret collaberation rumours, mixed in with GeForce5 / nForce GPUs & chipsets...
I looks to me that Apple is gonna take the horse by the reins, and steer it directly towards the high-end entertainment 3d market...
Now if the wankers would just hurry up and purchase Maya (and it's engineers/programmers...) from Alias|wavefront...!
<strong>There are some things in the OS X 10.1.3 CD that I found interesting. First there is a /private/drivers/i386 folder. Furthermore, in this folder is the standard System.config file which is the same as the one included with Darwin x86 1.4.1, but there is also a .config file for a network install option that is not in the Darwin distribution.
There is also a /usr/standalone/i386 folder. This is where the files necessary for an el-torito bootable CD would be kept. The only thing missing from this folder is the boot.flp file which the bios uses to emulate a 2.88 MB floppy. I know Darwin runs on Intel, I have it installed on an old PII, and that this folder is needed for it, but why would Apple leave this on a CD meant for a PowerPC machine?
This made me wonder whether some of the files on the CD were built fat by accident. I decided to try to build my own x86 install CD. I knew it was a long shot, but what the hell.
I had access to a new iBook, so I built and installed the mkisofs package needed to build a Darwin install CD. I then copied to the iBook the 10.1.3 CD and the Darwin x86 CD. I then mixed and matched the contents of the two to make a new CD structure that included the Intel kernel and low level binaries in the /bin and /sbin folder and the /System/Installation and /Applications etc. from the 10.1.3 CD. I then made the custom CD ISO and burned it. It booted fine on my Athlon, loaded the kernel, and then began to try to run the rc.cdrom install script from 10.1.3. I got excited and thought I was on to something when it then began to give me error messages that the wrong CPU type was specified in a bunch of Framework files. It was not able to load the CDIS splash screen in the/System/Installation folder or the Installer.app in the Applications folder. While I was defeated in my own attempts, I found it interesting how easy it would be for Apple to build an x86 build of OS X. Now all I need is someone who knows how to decompile the necessary file on the 10.1.3 CD and build them for x86?
Well, too bad it didn't work, but your attempt was pretty kewl.
No way any x86 platform is an option. Apple has been playing up the disadvantages of x86 for years, it would be major feast of crow to go x86 now. But, most other high performance cpus would be real possibilities. From IBM, we have Power line, from Intel and AMD we have Itanium (1 and 2) and Hammer (that is their 64bit chip right?). Other chips are options as well (MIPS, Sparc...), but as they are all from shrinking companies or product lines, they probably aren't good options as they would require Apple to the major promoter of the platform, even more than when they went PPC (remember at the time that Intel paid something like 90% of ad costs for companies that played the Intel jingle in their commercials. Apple tried for a similar deal with IBM and Moto, but was brushed off) With cpus from IBM, Intel or AMD, anyone else using the cpu in their machines, by default help promote Apple's choice of cpu.
Anyway, that was a tangent from the post I originally replied to, but didn't want to post another jsut to be on topic for the thread.
<strong>all I need is someone who knows how to decompile the necessary file on the 10.1.3 CD and build them for x86?
</strong><hr></blockquote>
Impossible. Still, almost every component of Mac OS X has a mention of i386, including AppleScript. This means that Apple is keeping x86-compatibility up-to-date. Of course, some required components are missing, such as frameworks with their libraries.
If ever Apple switches to x86, there will be no software/hardware PPC emulators. Software ones would be inefficient and utterly useless, hardware ones would be too expensive (consider a whole x86 CPU). If so, Apple probably will suggest 2 breeds of Mac OS X: one for PPC and the other for Intel/AMD. And yes, every app will have to be at least recompiled. And yes, every CPU-optimized app (such as Photoshop's AltiVec support) will have to be rewritten. Just remember that Photoshop, for example, already exists for x86 platform and is already optimized for it. Just mix Aqua for Mac OS version with mathematics' guts for Intel version and get your uber-new Photoshop for Mac OS X86. It's pretty simple in theory. If all system calls are ported properly, most apps will simply have to be recompiled.
One side-effect of this switch may be a huge load of other Windows-specific software made available for Mac OS X86.
Another side-effect is a terrible head-ache for developers supporting twice as many products and for you who, to know which CPU you are using, will 'ask your system administrator'
I'd hardly call the Itanium a 'good' chip. It's been plagued by so many set-backs and mhz stalls it would be the perfect successor to the PPC.</strong><hr></blockquote>
AMD's 64-bit chip is the "Hammer" (the x86-64 architecture). Intel's is the Itanium.
IF Apple moves to a new chip, I think they will aim for the path of least resistance. x86 does not fit here because of the amount of work/time/money involved by Apple, developers, and most importantly, consumers. I have spent several hundred dollars to make a full transition to OSX. The bulk of this cost came in the form of upgrades to my apps to make them native (some of which I had not planned on upgrading yet).
I believe that the move to The New Chip? will be more of an evolutionary change. It is in Apple's best interest to make this change as unobtrusive to the customer as possible (ie no need to pay for a suite of upgrades).
Regardless of who makes this chip for Apple, it will something that we don't have today. Maybe a cousin of the PPC (Power5 rumors) or a cross-breed of AMD's risc core and the PPC.
Apple is aiming it's guns at the HIGH-end market. Don't let the iApps fool you, SJ wants the crown back and he is going to storm the gates of the film/TV/effects industry to get it. They will need raw-power for this.
The transformation of Apple began when NeXt took them over. It has been most visible w/OSX, but, IMO, the real changes have only begun.
Here's the BIG question you'd have to ask yourselves.
Why would Apple transition to an x86 processor from Intel when Intel themselves are trying to transition to the EPIC (Explicit Parallel Instruction Computing) based processors offered in the Itanium line?
If Apple had stated they were going to go with an Intel based processor that would be one thing. But to say they were going with x86 developed over 20 years ago...PLEASE I don't think Apple is anywhere NEAR that STUPID!
Don't put too much stock (hehe ) in what Wall Street analysts say... These idiots urged Moto to dump SPS, while SPS holds more patents than the rest of Moto combined. Remove SPS, and Moto doesn't have much reason to exist. That's like donating a kidney and mentioning to the doc on the way into the operating room: "Why don't you cut out my heart, too!"</strong><hr></blockquote>
Wall street occasionally has some things worth while to say; however, most of the time, it's a chorus of "aren't we sophisticated and beautiful" ... as they parade their latest catch phrases (which convniently ignore how badly last week's *wisdom* worked out) like badges of membership.
You need to keep in mind that it isn't the PowerPc that has problems scaling, it's Motorola that seems to be having problems scaling 1 specific implementation - the G4. That could be a design issue with the G4, or it could be from Moto bleeding staff like a severed artery.
IBM seems to be having no problems scaling their versions. Take a look at the POWER4.
As far as ignoring what Wallstreet analysts have to say, that's easy. It was Steve Jobs that said,"that first the company had to finish the transition to the OS X operating system, expected around the end of this year.
"Then we'll have options, and we like to have options," he said. " "
Whether he meant it or it was an unintentionall misleading slip, Apple' s only options now are Motorola and IBM. This implies additional options not including Motorola or IBM, whether true or not.
I say unintentional misleading slip, because Jobs' reference to options may very well have been meant to include Motorola or IBM, also.
I wouldn't read too much into the quote, but in the immortal words of Arty Johnson it is "veerrrrryyyyy interesting".
For those not old enough to know who Arty Johnson is I apologize, lookup "Laugh In" on Google for sound bytes.
Also, what does Steve's quote"that first the company had to finish the transition to the OS X operating system, expected around the end of this year. " mean??????
What would require the "transition to the OS X operation system", before having options?????
... for one reason and one reason only: registers. You can't emulate a processor with many registers (PPC) with one with few registers (x86). If you tried, you'd get PPC601 66MHz performance (tops) on a P4 3GHz. It just won't work if you want to have backwards compatability, and it would be suicide not to have it; and that doesn't even get into the nightmare of emulating Altivec on SSEII.</strong><hr></blockquote>
runtime translation isn't necessary. translate ppc -> x86 and cache the resulting native (x86) binary, and it will run plently fast.
i do hope motorola realizes that it isn't that hard for apple to jump ship and go x86. maybe they'll get off their asses and make a g4 with a ddr fsb. i won't be holding my breath, though. embedded systems is what they care about, and where most ppcs are sold.
personally, i'm sticking with alpha (which is dieing) for now. my plan was to get a mac in 6 months or so, to run linux and mac os x (under mol). that will change if (big if) they switch to x86. i'm not going from a beautfil arch like alpha, to the ugly mutant x86.
<strong>He has also heard that Apple plans to release boxes that Dual Boot.
But a PowerPC made by AMD would work well.
</strong><hr></blockquote>
Dual Boot = Definate NO! Too confusing and ugly for a streamlined OS.
PPC by AMD = interesting. I remember when AMD bought out a company by the name of NexGen which started the whole upward swing (it was a RISC processor that emulated CISC instructions, and quite fast too) toward the Athlon. Could be a possibility but I have a gut feeling that since IBM already makes G3s for Apple, that they will be the ones with the new chip for the new PowerMac. And to that end, I am guessing it will be a POWER 5 variant.
Comments
bohdi. i don't understand. do you mean boot into osx or boot into windows?
inteeeerrrressssttttting....
There is also a /usr/standalone/i386 folder. This is where the files necessary for an el-torito bootable CD would be kept. The only thing missing from this folder is the boot.flp file which the bios uses to emulate a 2.88 MB floppy. I know Darwin runs on Intel, I have it installed on an old PII, and that this folder is needed for it, but why would Apple leave this on a CD meant for a PowerPC machine?
This made me wonder whether some of the files on the CD were built fat by accident. I decided to try to build my own x86 install CD. I knew it was a long shot, but what the hell.
I had access to a new iBook, so I built and installed the mkisofs package needed to build a Darwin install CD. I then copied to the iBook the 10.1.3 CD and the Darwin x86 CD. I then mixed and matched the contents of the two to make a new CD structure that included the Intel kernel and low level binaries in the /bin and /sbin folder and the /System/Installation and /Applications etc. from the 10.1.3 CD. I then made the custom CD ISO and burned it. It booted fine on my Athlon, loaded the kernel, and then began to try to run the rc.cdrom install script from 10.1.3. I got excited and thought I was on to something when it then began to give me error messages that the wrong CPU type was specified in a bunch of Framework files. It was not able to load the CDIS splash screen in the/System/Installation folder or the Installer.app in the Applications folder. While I was defeated in my own attempts, I found it interesting how easy it would be for Apple to build an x86 build of OS X. Now all I need is someone who knows how to decompile the necessary file on the 10.1.3 CD and build them for x86?
[ 07-18-2002: Message edited by: FotNS ]</p>
<strong>My friend 'in the know' if you could say that has been whispering this to me for about two weeks now. He has heard that Apple has a partnership with AMD. </strong><hr></blockquote>It's hard to imagine converting to a new chip, but maybe AMD will buy out Motorola's semiconductors and pick up the PowerPC.
<strong>My friend 'in the know' if you could say that has been whispering this to me for about two weeks now. He has heard that Apple has a partnership with AMD. He has also heard that Apple plans to release boxes that Dual Boot.
The second one I am leery of but he swears on the first, and this is the same guy who told me about the solid state scroll on thhe new iPods about two weeks ago. I don't think Apple is looking to go x86, they would piss off too many developers. But a PowerPC made by AMD would work well.</strong><hr></blockquote>
d00d! If you could get your buddy to elaborate on this....it would be very cool!
Could this partnership with AMD have to do with a mobo chipset, and Apple will continue to use Moto as their CPU supplier? Or do you think this AMD-Apple partnership is about CPUs?
And, aren't all current Macs "Dual Boot", in that one can boot into OS X or OS 9? Or do you mean dual boot, as in, these new "Macs" will be able to boot either OS X, or Windows?
Or could this Dual Boot mean that one boots into OS X on PPC for older apps, but boots into OS X on x86 for newer apps?
Just curious.
What troubles me about all this is the following:
1. Mac developers, including Apple, have invested a great amount of resources in Altivec. From OS X to Photoshop to screen savers, a multitude of Mac code has been written with Altivec in mind. Now that the G4 is finally in both pro AND consumer Macs, a developer can write altivec code with the knowledge that ANY Mac user will benefit. Futhermore, altivec optimizations are now an incentive for all mac users to choose a particular piece of software over another. It would be a small tragedy if right when Altivec was finally gaining widespread adoption and support, Apple abandoned it.
2. The transition to OS X continues, and by the looks of things, it will be years before the Mac user base has fully migrated to OS X, with OS 9 native apps being relics from the past. This transition is costly to both Apple and developers, AND to Mac users. Mac users who switch to OS X are also compelled to replace all of their software with OS X native versions, if not immediately, then after a few months of using the classic environment. For Apple to undertake another major transition within the next few years would be dangerous. Not only would Apple risk alienating developers (some who have invested lots of time and money in Altivec optimizations, imagine all the Mac programmers who've taken the time to learn how to code for altivec), but more importantly, Apple could alienate the entire Mac user base. Asking them to buy all new software once is bad enough, but doing it again....I don't see many people sticking with the platform. The migration to OS X is confusing enough for most people, try throwing in a concomitant migration to a different CPU architecture and you suddenly have a bunch of confused Mac users who will see Windows as being more simple and cheaper.
3. Motorola DOES have some sort of "G5" ready, and IBM has the Power 5 which will supposedly be good for workstations. With options like these, why would Apple choose x86? Furthermore, x86 is supposedly near the end of it's life. I don't understand the technical reasons, but it's something I've read a few places. Not in the next few years, but certainly within the next decade. IF Apple were to choose a new architecture, wouldn't it be better to choose something that is new, with a long life ahead of it, rather than something in it's twilight years, with the end in sight, like x86?
Eh, just some ideas.
The IBM Power 4/5 derivative sounds good... But the AMD fabbed / Apple designed G5 sounds more like the choice of a company (... a man...) that wants complete control over their 'widget'...
Couple this with the Apple / nVidia secret collaberation rumours, mixed in with GeForce5 / nForce GPUs & chipsets...
I looks to me that Apple is gonna take the horse by the reins, and steer it directly towards the high-end entertainment 3d market...
Now if the wankers would just hurry up and purchase Maya (and it's engineers/programmers...) from Alias|wavefront...!
;^p
Oh yeah,
Maya v4.5 announced for Mac OS X first! :cool:
<strong>what about AMD's new 64-bit chip? it seems like a good chip from a reliable supplier (unlike moto)</strong><hr></blockquote>
I'd hardly call the Itanium a 'good' chip. It's been plagued by so many set-backs and mhz stalls it would be the perfect successor to the PPC.
<strong>There are some things in the OS X 10.1.3 CD that I found interesting. First there is a /private/drivers/i386 folder. Furthermore, in this folder is the standard System.config file which is the same as the one included with Darwin x86 1.4.1, but there is also a .config file for a network install option that is not in the Darwin distribution.
There is also a /usr/standalone/i386 folder. This is where the files necessary for an el-torito bootable CD would be kept. The only thing missing from this folder is the boot.flp file which the bios uses to emulate a 2.88 MB floppy. I know Darwin runs on Intel, I have it installed on an old PII, and that this folder is needed for it, but why would Apple leave this on a CD meant for a PowerPC machine?
This made me wonder whether some of the files on the CD were built fat by accident. I decided to try to build my own x86 install CD. I knew it was a long shot, but what the hell.
I had access to a new iBook, so I built and installed the mkisofs package needed to build a Darwin install CD. I then copied to the iBook the 10.1.3 CD and the Darwin x86 CD. I then mixed and matched the contents of the two to make a new CD structure that included the Intel kernel and low level binaries in the /bin and /sbin folder and the /System/Installation and /Applications etc. from the 10.1.3 CD. I then made the custom CD ISO and burned it. It booted fine on my Athlon, loaded the kernel, and then began to try to run the rc.cdrom install script from 10.1.3. I got excited and thought I was on to something when it then began to give me error messages that the wrong CPU type was specified in a bunch of Framework files. It was not able to load the CDIS splash screen in the/System/Installation folder or the Installer.app in the Applications folder. While I was defeated in my own attempts, I found it interesting how easy it would be for Apple to build an x86 build of OS X. Now all I need is someone who knows how to decompile the necessary file on the 10.1.3 CD and build them for x86?
[ 07-18-2002: Message edited by: FotNS ]</strong><hr></blockquote>
Well, too bad it didn't work, but your attempt was pretty kewl.
No way any x86 platform is an option. Apple has been playing up the disadvantages of x86 for years, it would be major feast of crow to go x86 now. But, most other high performance cpus would be real possibilities. From IBM, we have Power line, from Intel and AMD we have Itanium (1 and 2) and Hammer (that is their 64bit chip right?). Other chips are options as well (MIPS, Sparc...), but as they are all from shrinking companies or product lines, they probably aren't good options as they would require Apple to the major promoter of the platform, even more than when they went PPC (remember at the time that Intel paid something like 90% of ad costs for companies that played the Intel jingle in their commercials. Apple tried for a similar deal with IBM and Moto, but was brushed off) With cpus from IBM, Intel or AMD, anyone else using the cpu in their machines, by default help promote Apple's choice of cpu.
Anyway, that was a tangent from the post I originally replied to, but didn't want to post another jsut to be on topic for the thread.
<strong>all I need is someone who knows how to decompile the necessary file on the 10.1.3 CD and build them for x86?
</strong><hr></blockquote>
Impossible. Still, almost every component of Mac OS X has a mention of i386, including AppleScript. This means that Apple is keeping x86-compatibility up-to-date. Of course, some required components are missing, such as frameworks with their libraries.
If ever Apple switches to x86, there will be no software/hardware PPC emulators. Software ones would be inefficient and utterly useless, hardware ones would be too expensive (consider a whole x86 CPU). If so, Apple probably will suggest 2 breeds of Mac OS X: one for PPC and the other for Intel/AMD. And yes, every app will have to be at least recompiled. And yes, every CPU-optimized app (such as Photoshop's AltiVec support) will have to be rewritten. Just remember that Photoshop, for example, already exists for x86 platform and is already optimized for it. Just mix Aqua for Mac OS version with mathematics' guts for Intel version and get your uber-new Photoshop for Mac OS X86. It's pretty simple in theory. If all system calls are ported properly, most apps will simply have to be recompiled.
One side-effect of this switch may be a huge load of other Windows-specific software made available for Mac OS X86.
Another side-effect is a terrible head-ache for developers supporting twice as many products and for you who, to know which CPU you are using, will 'ask your system administrator'
<strong>
I'd hardly call the Itanium a 'good' chip. It's been plagued by so many set-backs and mhz stalls it would be the perfect successor to the PPC.</strong><hr></blockquote>
AMD's 64-bit chip is the "Hammer" (the x86-64 architecture). Intel's is the Itanium.
I believe that the move to The New Chip? will be more of an evolutionary change. It is in Apple's best interest to make this change as unobtrusive to the customer as possible (ie no need to pay for a suite of upgrades).
Regardless of who makes this chip for Apple, it will something that we don't have today. Maybe a cousin of the PPC (Power5 rumors) or a cross-breed of AMD's risc core and the PPC.
Apple is aiming it's guns at the HIGH-end market. Don't let the iApps fool you, SJ wants the crown back and he is going to storm the gates of the film/TV/effects industry to get it. They will need raw-power for this.
The transformation of Apple began when NeXt took them over. It has been most visible w/OSX, but, IMO, the real changes have only begun.
Why would Apple transition to an x86 processor from Intel when Intel themselves are trying to transition to the EPIC (Explicit Parallel Instruction Computing) based processors offered in the Itanium line?
If Apple had stated they were going to go with an Intel based processor that would be one thing. But to say they were going with x86 developed over 20 years ago...PLEASE I don't think Apple is anywhere NEAR that STUPID!
That's my 2 cents!
[ 07-18-2002: Message edited by: Bombthroat ]</p>
<strong>Not going to happen
Don't put too much stock (hehe
Wall street occasionally has some things worth while to say; however, most of the time, it's a chorus of "aren't we sophisticated and beautiful" ... as they parade their latest catch phrases (which convniently ignore how badly last week's *wisdom* worked out) like badges of membership.
Ignore them - mostly; bet against them - mostly.
You are now a Jedi.
You need to keep in mind that it isn't the PowerPc that has problems scaling, it's Motorola that seems to be having problems scaling 1 specific implementation - the G4. That could be a design issue with the G4, or it could be from Moto bleeding staff like a severed artery.
IBM seems to be having no problems scaling their versions. Take a look at the POWER4.
"Then we'll have options, and we like to have options," he said. " "
Whether he meant it or it was an unintentionall misleading slip, Apple' s only options now are Motorola and IBM. This implies additional options not including Motorola or IBM, whether true or not.
I say unintentional misleading slip, because Jobs' reference to options may very well have been meant to include Motorola or IBM, also.
I wouldn't read too much into the quote, but in the immortal words of Arty Johnson it is "veerrrrryyyyy interesting".
For those not old enough to know who Arty Johnson is I apologize, lookup "Laugh In" on Google for sound bytes.
Also, what does Steve's quote"that first the company had to finish the transition to the OS X operating system, expected around the end of this year. " mean??????
What would require the "transition to the OS X operation system", before having options?????
[ 07-18-2002: Message edited by: rickag ]</p>
<strong>Not going to happen
... for one reason and one reason only: registers. You can't emulate a processor with many registers (PPC) with one with few registers (x86). If you tried, you'd get PPC601 66MHz performance (tops) on a P4 3GHz. It just won't work if you want to have backwards compatability, and it would be suicide not to have it; and that doesn't even get into the nightmare of emulating Altivec on SSEII.</strong><hr></blockquote>
runtime translation isn't necessary. translate ppc -> x86 and cache the resulting native (x86) binary, and it will run plently fast.
i do hope motorola realizes that it isn't that hard for apple to jump ship and go x86. maybe they'll get off their asses and make a g4 with a ddr fsb. i won't be holding my breath, though. embedded systems is what they care about, and where most ppcs are sold.
personally, i'm sticking with alpha (which is dieing) for now. my plan was to get a mac in 6 months or so, to run linux and mac os x (under mol). that will change if (big if) they switch to x86. i'm not going from a beautfil arch like alpha, to the ugly mutant x86.
<strong>He has also heard that Apple plans to release boxes that Dual Boot.
But a PowerPC made by AMD would work well.
</strong><hr></blockquote>
Dual Boot = Definate NO! Too confusing and ugly for a streamlined OS.
PPC by AMD = interesting. I remember when AMD bought out a company by the name of NexGen which started the whole upward swing (it was a RISC processor that emulated CISC instructions, and quite fast too) toward the Athlon. Could be a possibility but I have a gut feeling that since IBM already makes G3s for Apple, that they will be the ones with the new chip for the new PowerMac. And to that end, I am guessing it will be a POWER 5 variant.
<strong>The transformation of Apple began when NeXt took them over.</strong><hr></blockquote>
When Apple brought in SJ, NeXT wasn't even a company!
<strong>What would require the "transition to the OS X operation system", before having options?????</strong><hr></blockquote>
OS9. Ever seen the OS9 kernel running on a platform other than PPC? The OSX kernel does.