I will feel outdated once I will be stuck with outdated programs, updates for which will only be available for 10.6.
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.
PowerPC Support in Mac OS X 10.6 After All?
- Thread starter MacRumors
- Start date
- Sort by reaction score
I will feel outdated once I will be stuck with outdated programs, updates for which will only be available for 10.6.
I put a nice Seagate w/32MB cache in mine, and it really flies.
The people who think they are slow don't have one. The ones who think PPC code that isn't running on x86 is hurting their OS stability aren't programmers.
As someone with around 50 Macs. None of which are Intel based, I voted to drop PPC.
No, it's not just to convince the wife that I need a new Mac Pro.
Right now Apple is supporting 4 different CPU architectures:
32-bit PowerPC (G4)
64-bit PowerPC (G5)
32-bit x86 (original Core processors - non-Core2)
64-bit x86 (Core2/Xeon)
Totally insane. They should never have released any Macs using 32-bit Intel.
Each type adds size to applications a UB program has to have executables for both PPC and x86 (I'm not sure if you need 32 and 64-bit versions, too).
I'm sure the money they're spending on bandwidth for the inflated sizes of all programs and the OS works out to more than I make in a year.
I'm probably overstating it all, but I'm sure it's easier to optimize a program for a single architecture than 4.
In the end, Leopard will still run on my 400 MHz G4 long after 10.6 is out. People seem to think being left out of the newest OS breaks their computers. Would I love to run Leopard on my G3 iBook? Sure. Do I expect Apple to support it? Not a chance.
Not.
Apple said they were maintaining Intel builds since 2001 or earlier. It is NOT hard to do once you are doing it. Linux & the BSDs maintain themselves across a DOZEN architectures with sometimes just a volunteer or two keeping an architecture alive.
That's what compilers are for. The amount of code that is processor specific is very limited, and it doesn't change too frequently.
PPC owners: You're not required to upgrade to every Mac OS.
Most of the rumors about 10.6 say there isn't going to be anything major (no Time Machine, no cover flow, etc.) so it's not like you're going to be screwed out of anything. IMO one of Windows' biggest problems is Microsoft wants to support everything ever written for a Windows/DOS platform. At some point, you gotta tell people running DOS 6.0 applications that their schyte is outdated.
We're running Jaguar at work on PPC Power Mac G5 machines. The OS X apps seem to work pretty well, but the Classic stuff is full of problems. Apple went pretty far keeping support for that old crap (I hate OS 9 with a passion) and I do believe finally ditched support in Leopard. Maybe 10.6 would be too soon, but I personally wouldn't see a problem as long as they kept offering Leopard support.
But if Apple keeps it, I don't wanna hear whining in 2011 about it. By then, get a new computer folks!
Most of the rumors about 10.6 say there isn't going to be anything major (no Time Machine, no cover flow, etc.) so it's not like you're going to be screwed out of anything. IMO one of Windows' biggest problems is Microsoft wants to support everything ever written for a Windows/DOS platform. At some point, you gotta tell people running DOS 6.0 applications that their schyte is outdated.
We're running Jaguar at work on PPC Power Mac G5 machines. The OS X apps seem to work pretty well, but the Classic stuff is full of problems. Apple went pretty far keeping support for that old crap (I hate OS 9 with a passion) and I do believe finally ditched support in Leopard. Maybe 10.6 would be too soon, but I personally wouldn't see a problem as long as they kept offering Leopard support.
But if Apple keeps it, I don't wanna hear whining in 2011 about it. By then, get a new computer folks!
PPC owners: You're not required to upgrade to every Mac OS.
Most of the rumors about 10.6 say there isn't going to be anything major (no Time Machine, no cover flow, etc.) so it's not like you're going to be screwed out of anything. IMO one of Windows' biggest problems is Microsoft wants to support everything ever written for a Windows/DOS platform. At some point, you gotta tell people running DOS 6.0 applications that their schyte is outdated.
We're running Jaguar at work on PPC Power Mac G5 machines. The OS X apps seem to work pretty well, but the Classic stuff is full of problems. Apple went pretty far keeping support for that old crap (I hate OS 9 with a passion) and I do believe finally ditched support in Leopard. Maybe 10.6 would be too soon, but I personally wouldn't see a problem as long as they kept offering Leopard support.
But if Apple keeps it, I don't wanna hear whining in 2011 about it. By then, get a new computer folks!
Michael, again, I will repeat like a parrot what I said before - shouldn't my 2006 Powermac G5 be new enough even for what you call a minor upgrade 10.6?
Just because it's a minor update doesn't mean that my powerful and capable G5 shouldn't have it.
iPhone is a ppc derivative
I think Apple will still support PPC for 10.6 and MAYBE 10.7 due to the fact they must still develop for PPC, considering that the iPhone is a PPC derivative (arm (Advanced RISC Machine)). Im not sure if the iPod touch and the iPod classic are PPC derivatives, if they are it would be even more logical to assume that Apple must still code Mac OS X for PPC and if they are might as well just make a full Mac OS X PPC distribution. On whether or not a PPC machine could handle 10.5 and 10.6, most definitely the MDD g4 towers would be a bit stretched without upgrades but the G5s even without any upgrades would do quite well and if you were to upgrade them then MOST definitely. If the rumor is true that Apple is focusing on optimization, performance and stability for 10.6, it would be logical to assume that if a mac can run 10.5 that, that same mac would be able to run 10.6 even better. I am reminded of 10.2 and 10.3, how 10.2 was terribly slow and unoptimized but then 10.3 came out which dramatically improved speed and performance. I no doubt that some PPC macs will have to be dropped for 10.6 as the optimizations wouldn't be enough to get 10.6 to run well on them. Even so there are still quite a few PPC macs that could run 10.6 . On a side note somewhat related to this discussion, I am throughly displeased with Apple's decision to cut CCE (Classic Compatibility Environment) on 10.6 for the PPC distribution (considering that not much if any emulation would be needed and not many more lines of code to get CCE to work with Mac OS X 10.5 . I can see why they dropped CCE on the Intel distribution as it would be quite a bit more work to get old Classic PPC and 68k applications to run on an Intel mac (requires allot more emulation and special workarounds, not to mention many more lines of code). In my opinion (which may be horribly wrong) Apple would have many more users willing to make a full switch to a new Intel mac if all there PPC applications (I would see dropping 68k classic applications as a reasonable compromise) regardless of OS X or Classic was fully supported. Obviously an all inclusive legacy support for Classic applications would be preferred. As it stands now I personally have had to move off all my Classic OS PPC applications to an iMac and all my 68k stuff to an lc575 to make the switch to 10.5 possible. I still as of yet to switch to 10.5 due to the fact that M-Audio has as of yet to release a fully supported non beta driver for my M-Audio Revolution 5.1 hardware sound card.
I think Apple will still support PPC for 10.6 and MAYBE 10.7 due to the fact they must still develop for PPC, considering that the iPhone is a PPC derivative (arm (Advanced RISC Machine)). Im not sure if the iPod touch and the iPod classic are PPC derivatives, if they are it would be even more logical to assume that Apple must still code Mac OS X for PPC and if they are might as well just make a full Mac OS X PPC distribution. On whether or not a PPC machine could handle 10.5 and 10.6, most definitely the MDD g4 towers would be a bit stretched without upgrades but the G5s even without any upgrades would do quite well and if you were to upgrade them then MOST definitely. If the rumor is true that Apple is focusing on optimization, performance and stability for 10.6, it would be logical to assume that if a mac can run 10.5 that, that same mac would be able to run 10.6 even better. I am reminded of 10.2 and 10.3, how 10.2 was terribly slow and unoptimized but then 10.3 came out which dramatically improved speed and performance. I no doubt that some PPC macs will have to be dropped for 10.6 as the optimizations wouldn't be enough to get 10.6 to run well on them. Even so there are still quite a few PPC macs that could run 10.6 . On a side note somewhat related to this discussion, I am throughly displeased with Apple's decision to cut CCE (Classic Compatibility Environment) on 10.6 for the PPC distribution (considering that not much if any emulation would be needed and not many more lines of code to get CCE to work with Mac OS X 10.5 . I can see why they dropped CCE on the Intel distribution as it would be quite a bit more work to get old Classic PPC and 68k applications to run on an Intel mac (requires allot more emulation and special workarounds, not to mention many more lines of code). In my opinion (which may be horribly wrong) Apple would have many more users willing to make a full switch to a new Intel mac if all there PPC applications (I would see dropping 68k classic applications as a reasonable compromise) regardless of OS X or Classic was fully supported. Obviously an all inclusive legacy support for Classic applications would be preferred. As it stands now I personally have had to move off all my Classic OS PPC applications to an iMac and all my 68k stuff to an lc575 to make the switch to 10.5 possible. I still as of yet to switch to 10.5 due to the fact that M-Audio has as of yet to release a fully supported non beta driver for my M-Audio Revolution 5.1 hardware sound card.
If there are different code bases for all four then I say reduce to 64 bit only on both PPC and x86.
If 64/32 are included as a group regardless, I say support PPC for 5 years after the last PPC shipped. Applecare is obligated for 3 years and any renewal of that makes 5 reasonable.
That does not necessarily say every feature need be supported, but one wonders why Apple does not for example release an OS 9.2 compatibility upgrade for awareness of things released since then. There has to be a few relatively easy things they could do.
It appears 10.4.11 is that bridge within the OSX world to prior versions of hardware, and 10.5/10.6 is a "new era" of sorts. I for one would guess 10.7 might be the Intel only OS, but it will also include several different Intel processor types support.
Apple can't seem to escape the complexity problem. Hence the iPhone. Very unified indeed.
Rocketman
I've been paying close attention to this and similar threads on the subject, and I'm greatly dismayed by those who believe that the presence of PPC-supporting code, in the source of MacOS X, might somehow effect in any way the experience of Intel-owning Macintosh community: whether it be "bloat", or "baggage", or stability problems.
It is clear that there is some disconnect in understanding afoot here. The platform-dependent parts of MacOS X are relegated to the lowest levels of abstraction, and they exist in the form of code that is guarded by C preprocessor macros, such as
#ifdef __APPLE_ALTIVEC__
#if TARGET_CPU_X86
#if TARGET_CPU_PPC
By the time MacOS X has been compiled and placed on a DVD to ship to you, all of the code guarded by TARGET_CPU_PPC has been compiled out of the way so that the code in your x86 segment of the fat binaries is uncluttered by it, unhampered by it, unimpeded by it, and uncompromised by it.
It does not weigh your system down. It does not make your system slower than it could be. All that it does is take up a tiny fraction of disk space for the PPC segment of each fat binary, and if it really concerns you then you could get off your ass and read the man page for LIPO(1) and get on with your lives.
Seriously, people. Are you going to start bitching about #if TARGET_CPU_ARM targets that exist on the OSX iPhone branch next?
There is no way in hell that Apple is going to squander good will by dropping PPC support for machines that, in my experience, run leopard just fine if they have enough RAM.
This code that we're speaking of began life on 68030, then added suppport for SPARC, PA-RISC, and x86 and had been ported to PPC in the NeXT labs long before their inverse-acquisition of Apple. The way that this code has been made portable is very mature and clean, and your experience does not suffer one iota from the presence of code-paths that never even make it into the binaries that you run.
Do the world a favor and get a grip & a clue. Peace out.
It is clear that there is some disconnect in understanding afoot here. The platform-dependent parts of MacOS X are relegated to the lowest levels of abstraction, and they exist in the form of code that is guarded by C preprocessor macros, such as
#ifdef __APPLE_ALTIVEC__
#if TARGET_CPU_X86
#if TARGET_CPU_PPC
By the time MacOS X has been compiled and placed on a DVD to ship to you, all of the code guarded by TARGET_CPU_PPC has been compiled out of the way so that the code in your x86 segment of the fat binaries is uncluttered by it, unhampered by it, unimpeded by it, and uncompromised by it.
It does not weigh your system down. It does not make your system slower than it could be. All that it does is take up a tiny fraction of disk space for the PPC segment of each fat binary, and if it really concerns you then you could get off your ass and read the man page for LIPO(1) and get on with your lives.
Seriously, people. Are you going to start bitching about #if TARGET_CPU_ARM targets that exist on the OSX iPhone branch next?
There is no way in hell that Apple is going to squander good will by dropping PPC support for machines that, in my experience, run leopard just fine if they have enough RAM.
This code that we're speaking of began life on 68030, then added suppport for SPARC, PA-RISC, and x86 and had been ported to PPC in the NeXT labs long before their inverse-acquisition of Apple. The way that this code has been made portable is very mature and clean, and your experience does not suffer one iota from the presence of code-paths that never even make it into the binaries that you run.
Do the world a favor and get a grip & a clue. Peace out.
Right now, the NetBSD motto is "Of course it runs NetBSD" with support for 58 platforms. It's not that insane to write code that can be compiled for multiple platforms.
just because it's powerful enough doesn't mean it can. There are already plenty of Intel only applications. Why? Because the code is optimized to run on intel and it takes a lot to support different architectures and that comes with the loss of optimization and disk space.
Not sure that would be the case, I think it's one codebase, three platforms (i386, PPC, and ARM6).
Most Intel-only apps are more to do with their uses, like Vmware Fusion, Parallels or CrossOver. NeXT had been supporting multiple architectures with little extra fuss 15 years ago. No reason it should be more difficult now.
Count me in for supporting machines released within the last five years with the latest version of Mac OS X. That might not be realistic if Apple decides to cut off the G4, as the iBook G4 was being sold up to the release of the MacBook in May 2006, but at the very least those who spent big bucks on the G5 towers shouldn't be left in the cold. As others have pointed out, it's about loyalty: if Apple makes it difficult for current owners to stay on the cutting edge for more than two or three years, they'll be more likely to look elsewhere.
In 2011, Apple should go completely 64-bit Intel; until then, support for 32-bit Intel and at least 64-bit PowerPC should remain.
In 2011, Apple should go completely 64-bit Intel; until then, support for 32-bit Intel and at least 64-bit PowerPC should remain.
I'll wade through all the posts in a minute, but I want to state one thing ASAP:
IMHO, Apple should not drop PPC support, for two reasons:
1) The latter G5's (Dual 2.7's, Quads...) are still very powerful machines, and would be crazy to not support them in a next OS. Maybe the OS requirements could be Dual (Core) G5 or higher, but Apple cannot drop the G5 this quickly.
What they can do, is limit the features. They have done that a bit in Leopard already (IIRC the video effects in iChat for example).
2) "Just in case"
I know Intel and x64 seems to be the future, and I'm sure it will be for some time.
But Steve was so clever to have an x86 version of ANY version of OS X in the closet. It sure could be a gr8 idea to keep PPC versions of > 10.5 developed...
What I mean is that development of Mac OS X could be kept alive on the PowerPC architecture, like the Power5, not just support for older types of the PPC.
It could also lead to versions of Mac OS X Server on the high-end PowerPC.
I think Steve once said: "It's good to have options".
So, if you CAN have those options, please use them.
Mac OS X Leopard is proving to be an outstanding Universal OS. Keep it up.
IMHO, Apple should not drop PPC support, for two reasons:
1) The latter G5's (Dual 2.7's, Quads...) are still very powerful machines, and would be crazy to not support them in a next OS. Maybe the OS requirements could be Dual (Core) G5 or higher, but Apple cannot drop the G5 this quickly.
What they can do, is limit the features. They have done that a bit in Leopard already (IIRC the video effects in iChat for example).
2) "Just in case"
I know Intel and x64 seems to be the future, and I'm sure it will be for some time.
But Steve was so clever to have an x86 version of ANY version of OS X in the closet. It sure could be a gr8 idea to keep PPC versions of > 10.5 developed...
What I mean is that development of Mac OS X could be kept alive on the PowerPC architecture, like the Power5, not just support for older types of the PPC.
It could also lead to versions of Mac OS X Server on the high-end PowerPC.
I think Steve once said: "It's good to have options".
So, if you CAN have those options, please use them.
Mac OS X Leopard is proving to be an outstanding Universal OS. Keep it up.
Universal Binary has nothing to do with optimizations. Its simply taking two binaries and putting them into one package. There are zero benefits to going to two DVDs. Zero. None. Not even imaginable ones.
Code does not care about the other architectures. Lets put it this way. If Apple was to build OS X for Intel. Then separately build it for PPC. It would be identical to them building it as a Universal Binary.
Even if it does take a lot I still think it's worth it. Not because of "loyal PPC" users who will try to stick with PPC only because it's PPC.. but because of people like me who got what they got. If I had 8-Core Mac Pro I would be delighted but I got what I got and I'm happy.
It's a powerful, capable and stable machine. I don't think it will be fair to drop support for PPC (G5) after only one OS upgrade. At least 10.7 or 10.8..
I've heard of only a few. A quick search found Soundbooth and the aforementioned Java SE 6. In time, it will come. John Nack pointed out (in October 2006) the same ideas as many of the posters here:
People do seem to have a short memory.
Development of 10.5 took a long, long time. They initially said it would be released at the end of 2006, then it was said to arrive in the spring of 2007, it was finally released in winter 2007. This was apparently because of iPhone development.
Now correct me if I'm wrong (I'm sure someone thinks they know better) but don't alternate versions of Mac OS X get developed concurrently i.e., one team does 10.4 --> 10.6 while another team work on 10.3 --> 10.5. If this is true then the team that are doing 10.6 would always have planned to deliver around about this sort of time frame, with 10.5's late arrival making the gap between about 12 months.
Oc course that could all be bollocks. PPC support will stay. And we'll get new features.
Development of 10.5 took a long, long time. They initially said it would be released at the end of 2006, then it was said to arrive in the spring of 2007, it was finally released in winter 2007. This was apparently because of iPhone development.
Now correct me if I'm wrong (I'm sure someone thinks they know better) but don't alternate versions of Mac OS X get developed concurrently i.e., one team does 10.4 --> 10.6 while another team work on 10.3 --> 10.5. If this is true then the team that are doing 10.6 would always have planned to deliver around about this sort of time frame, with 10.5's late arrival making the gap between about 12 months.
Oc course that could all be bollocks. PPC support will stay. And we'll get new features.
All Intel chips Apple has used support SSE3 (introduced with Prescott on the P), and SSE4.1 wasn't introduced until Penryn. SSE has nothing to do with 32-bit or 64-bit OSes, and doesn't exclude any Intel system either.
Hell, Altivec in the G4/G5 line still is comparable to SSE3 in functionality, and we have to wait until AVX/SSE5 until a lot of the nice fused ops show up on x86/x64.
Even though I have moved on and use Core 2 Duo machines, I haven't seen a really good reason why Apple would ditch PPC just yet. The hard work of the port is done and code is written in platform neutral code above the HAL. So the PPC-specific code that still needs maintaining is the HAL itself (which is a mix of headers and CPU drivers), and chipset drivers used in the G4/G5 systems that aren't used after the switch to Intel chipsets. These should have been in maintenance mode since the systems started being manufactured. This form of support is no different from supporting 2-4 year old Intel systems in the future.
Sort-of. Usually what you see is a core development team that works on the next big release, but they have time budgeted to fix bugs in the old code tree (or backport fixes from the new release). Sometimes you might see a second, smaller team that takes over a big release once it RTMs and handles the majority of the post-release fixes.
The reason why the iPhone development would delay Leopard would be if this core team got partially re-assigned to help the iPhone team hit their deadlines (it does happen).
There is no reason to get rid of PPC machines, they work just fine on Leopard and Tiger. However, I don't think that means Apple should continue putting any effort into PPC when working on a NEW OS. Now I understand that maintaining basic compatibility with PPC doesn't affect the stability or "bloat" of Intel machines, and much of their current code is platform-agnostic. However, if Apple is able to drop PPC support when they are working on 10.6 for performance and stability, wouldn't they be able to optimize different parts of the operating system and their software for x86-64? And if they are spending time on maintaining PPC compatibility and the testing that is involved, that means there is less time for everything else.